PDA

View Full Version : A rant towards a rescue. Needed. NOW. ... From here to where?



Moonbase
2009-08-17, 03:58
In all earnestness, I start seeing the "SqueezeBox" product range go down the wrong path. Having been a product manager myself for a long time, I hear the alarm bells shrill for quite some time now.

As a developer, being enthusiastic for some time, I'm constantly losing interest in cooperation, because the development paths seem unclear and not well-defined, old bugs keep resurfacing, odd decisions are made (and taken back).

The product itself (Duet and the Controller, mainly) keeps getting more unusable with every "update" -- one starts getting anxious ("What will break this time?") instead of happily anticipating things getting better and more usable -- and being able to USE it for what it's meant: An enyojable way of listening to and showing off your music.

Successful products are "emotional" products, remember? So why isn't the emotion a good one, but frustration mostly?

Slimserver and the Squeezebox products have long kept you feeling that you are at the top of current technology, running with the fast crowd, simply owning and using the best you can possibly get. And be proud of it. Albeit having a few minor problems, you simply KNEW these would be gone soon -- and usually you'd get something better beyond all expectations.

Here's how the current development looks to an "outsider" (i.e., someone who knows about product management, development, usability, marketing, but also insists in keeping the standpoint of a "user"/"buyer") today:


There is no clear path where to go. What are the goals, where is the target market?
We keep adding "glitz" and pomp. Because we don't know how to focus on the basic problems and hope the user won't mind lack of basic functionality because there's so much "nice" stuff.
Bugs long gone and repaired keep surfacing again. And again. Ideas comes to mind, like "Do they ever USE their stuff themselves?" ... "Why the heck do we have to fight the character set wars again and again since 30 years? Don't they ever learn that there ARE characters beyond 0-9 and A-Z?"
It's all too technical. I want MUSIC. And the BEST and most USABLE way to handle it. Today, it's all about bug fixing, Perl programming, product name changes, database issues no one understands (and which they take back after investing so much efforts into a solution that's worse than before -- and while they take back the changes to "what we got before", they break even what we HAD before!).
Even (external) developers get frustrated -- after investing so many countless hours into repairing things that get broken again next day. (Yes, the code IS great; yes, they have thought about almost everything already; but they keep overlooking SIMPLE things over and over again -- probably because no one has shown them that some things ARE important to USERS. Even with Open Source, goals need to be defined, and milestones met. And everyone in the team needs to know "where we all want to go".)
Look at the "new" controller firmware. It LOOKS worse. It's unusable beyond expectation. It needs more keystrokes for simple functions. It doesn't behave logical or even consistent. BUT it has overlaid screens that nobody needs, it overlaps badly rendered cover images with status lines, it uses LARGE fonts so that you have to wait for some information to scroll until -- in the middle of reading -- the screensaver pops in and you have to press yet another key ... Aaaargh! It has a zillion "effects" that DISTRACT (instead of FOCUSING) the user's attention, while lacking some basic functionality. Oh well, and it has "apps"! Does anyone need "apps" while the basics don't function yet? ... I doubt it.
It's great to let the "tech kiddies" show off. And they can actually be a very valuable resource -- if LEAD correctly. PLEASE, Logitech, hire someone who knows about USABILITY. And probably how to manage a product line.
It still IS (could be?) such a great product line. Don't throw it away like this.


This reads like a rant. It is. Because I still love the product and the ideas behind it. (Emotion, remember?) I hope that the right people can understand my current frustration. And ACT.

I really wouldn't like to see others "winning" in spite of having higher prices and worse technology. But knowing what the customer wants.

Keep the promise! Even if it's only in our heads.

Thank you for reading.

maggior
2009-08-17, 08:13
The product itself (Duet and the Controller, mainly) keeps getting more unusable with every "update" -- one starts getting anxious ("What will break this time?") instead of happily anticipating things getting better and more usable -- and being able to USE it for what it's meant: An enyojable way of listening to and showing off your music.


This is why I've decided to get very picky about the build that I'll run on my server. Yes, it's fun to play, but I also like to sit an enjoy my music. Unless there is a particular bugfix or feature that I NEED in a release, I sit back and wait for the dust to settle. Even then, after I install it, if it doesn't work out, I immediately go back. I just don't have time any more to dig in and try to resolve issues that weren't threre previously. If I were single or didn't have any young kids, I might feel differently.

I suspect others (inlcuding yourself) are getting to the point.



Successful products are "emotional" products, remember? So why isn't the emotion a good one, but frustration mostly?


I agree that this could start to be a problem. If the product starts getting a rep as being frustrating to use and being unstable, the beginning of the end could be neigh. There are so many media streamers popping up out there that do music and video, people will start going elsewhere.



Look at the "new" controller firmware. It LOOKS worse. It's unusable beyond expectation. It needs more keystrokes for simple functions. It doesn't behave logical or even consistent. BUT it has overlaid screens that nobody needs, it overlaps badly rendered cover images with status lines, it uses LARGE fonts so that you have to wait for some information to scroll until -- in the middle of reading -- the screensaver pops in and you have

After reading this (and some other posts) I cringe. The controller FW is one area that has definitely required stabilization over time, and it has. I suspect I'll be staying on 7.3.3 for a long time! At least I have that option. I wonder why there has been such an overhaul on the UI on the controller? I wasn't aware that it had so many problems?


The flip side of this is that it is *good* that they release not-fully-baked versions of software to solicit input from the user community. As long as they continue to take the input and incorporate it (which I think they tend to do). These releases are always clearly labeled as BETA, so you always have fair warning.

I'd like to think (and perhaps I'm being too much of an optimist) that what is going on is that they are beefing up the feature list and making some major enhancements that will shake things up a bit, making things unstable for a few releases. They will continue to work on things to bring it back to stability. They've done this before - look at the new streaming and synchronization feature. They rewrote the *basic underpinnings* of how the products works. Frankly, that's a move that takes balls and shows great courage. And they worked to get it working correctly. Now we have a great new feature and things are good again.

My hope is that the SW dev team continues to work in this fashion. And I can sit on the sidelines (by choice) and wait for the dust to settle each time it gets shaken up.

Muele
2009-08-17, 08:50
[QUOTE=maggior;449159]
After reading this (and some other posts) I cringe. The controller FW is one area that has definitely required stabilization over time, and it has. I suspect I'll be staying on 7.3.3 for a long time! At least I have that option. I wonder why there has been such an overhaul on the UI on the controller? QUOTE]

I much agree with both of you.
I just wanted to comment on the UI overhaul. If the upcoming SB-Touch-device shares the software with the controller, I suspect they are optimizing the UI for the wider screen of that device. Also for that device the ability to read the screen from a distance is valuable. Hence the bigger font. (All that said, I haven't seen the new UI as I'm hessitant to spend x ammount of time getting a beta version to work. (small children in the house and all that))

Sorry for going slightly off topic.

tamanaco
2009-08-17, 09:33
I share Moonbase's feelings also... specially this point... "There is no clear path where to go. What are the goals, where is the target market?". The move to SQLite (possibly to accommodate the new product line) and then back to MySQL completely left the existing MySQL users, like myself, wondering what to do. Do I stick with MySQL? Is SQLite the future... and if so why? Because of the new product line?... because NASes need a smaller OS footprint?... because it has better performance?... Should I plan to move to SQLite or not? Nothing has been clearly articulated. The "existing" user base is confused... I know betas are betas, but there is no clear vision (strategy) to which clear goals (tactics) are attached. What I fear is that the plugin developers will get frustrated with the platform changes. They have either stopped development (enhancements) of their current plugins or have not introduced new ones in quite some time. Palm OS is a great example of developers moving on to greener pastures when Palm failed to communicate a clear vision for them. Look at the total number of Palm applications and then find how many of those have been enhanced or developed recently.

JJZolx
2009-08-17, 10:37
It looks like recently there was another personnel shakeup at the Logitech SMBU, and I'm afraid it isn't budget related this time. No doubt someone overseeing the unit is feeling many of the same issues.

https://bugs.slimdevices.com/show_bug.cgi?id=3383

The retreat from pushing the SQLite update out in 7.4 is a _good_ sign, IMO, and points to someone taking a serious look at real world time tables and development resources. But I also agree that much of what's being added these days is half-finished junk used only to put bullet points on retail packaging. And that may ultimately be the downfall of the product.

rtitmuss
2009-08-17, 10:37
We will be moving to SQLite in a future release of SqueezeCenter, probably 8.0. This change is being made to improve performance, especially on NASes and similar devices. It was decided that the SQLite changes will not be complete for the 7.4 release, which is why we reverted back to MySQL at the moment.

Richard

maggior
2009-08-17, 11:06
It looks like recently there was another personnel shakeup at the Logitech SMBU, and I'm afraid it isn't budget related this time. No doubt someone overseeing the unit is feeling many of the same issues.

https://bugs.slimdevices.com/show_bug.cgi?id=3383


I saw this in another thread. This is interesting. I wonder if this is in part in reaction to Sean leaving the company.


But I also agree that much of what's being added these days is half-finished junk used only to put bullet points on retail packaging. And that may ultimately be the downfall of the product.

Could you elaborate? I don't mean that in a confrontational sense - I'm genuinely curious.

tamanaco
2009-08-17, 11:39
We will be moving to SQLite in a future release of SqueezeCenter, probably 8.0. This change is being made to improve performance, especially on NASes and similar devices. It was decided that the SQLite changes will not be complete for the 7.4 release, which is why we reverted back to MySQL at the moment.

Richard

This sounds reasonable... but my fear is that Logitech will move the development resources from the current 7.4 MySQL platform to the future SQLite "before" the 7.4 MySQL is fully baked. Plugin developers "might" not want to work on a platform that will be sunset in the near future. I think a clearly articulated framework to move the plugins from this platform to the next to lessen the blows would help. I might be way out on left field here... but if the plugins that are developed for the officially released 7.4 were to work without much effort on the SQLite platform... it'd be big step forward. The recent move to SQLite broke too many plugins.

Matt Wise
2009-08-17, 11:58
In all earnestness, I start seeing the "SqueezeBox" product range go down the wrong path. Having been a product manager myself for a long time, I hear the alarm bells shrill for quite some time now.

As a developer, being enthusiastic for some time, I'm constantly losing interest in cooperation, because the development paths seem unclear and not well-defined, old bugs keep resurfacing, odd decisions are made (and taken back).


I understand your dilemma... let me assure you that we're working very hard to continue to work with the development community and that we actually want to increase the amount of community contribution significantly to future software releases.



The product itself (Duet and the Controller, mainly) keeps getting more unusable with every "update" -- one starts getting anxious ("What will break this time?") instead of happily anticipating things getting better and more usable -- and being able to USE it for what it's meant: An enyojable way of listening to and showing off your music.


We're definitely aware that software updates have broken things for some users... however we work very closely with our Support staff and are fairly confident that the most recently released software (7.3.3) is the most stable we've had since 7.0. I understand that each persons experience is different, but perhaps you might open another thread (or contact support) to deal with your specific issues?



Successful products are "emotional" products, remember? So why isn't the emotion a good one, but frustration mostly?

Slimserver and the Squeezebox products have long kept you feeling that you are at the top of current technology, running with the fast crowd, simply owning and using the best you can possibly get. And be proud of it. Albeit having a few minor problems, you simply KNEW these would be gone soon -- and usually you'd get something better beyond all expectations.

Here's how the current development looks to an "outsider" (i.e., someone who knows about product management, development, usability, marketing, but also insists in keeping the standpoint of a "user"/"buyer") today:

There is no clear path where to go. What are the goals, where is the target market?

This has been communicated before I believe... but there's a very clear emphasis on making our products easier to use for the average-joe. To your point above, the easier the product is to use, the better the emotional response will be.



We keep adding "glitz" and pomp. Because we don't know how to focus on the basic problems and hope the user won't mind lack of basic functionality because there's so much "nice" stuff.


I'd argue that since Duet was released (and subsequently, Boom) we've made almost no "glitzy" upgrades. We've been focused almost exclusively on improving the quality of our software from bottom to top. Only with the most recent 7.4 beta firmwares for Squeezebox Controller have we begun work on improving the UI "glitziness".



Bugs long gone and repaired keep surfacing again. And again. Ideas comes to mind, like "Do they ever USE their stuff themselves?" ... "Why the heck do we have to fight the character set wars again and again since 30 years? Don't they ever learn that there ARE characters beyond 0-9 and A-Z?"


Are you referring to recent 7.4 betas, or our released software?



It's all too technical. I want MUSIC. And the BEST and most USABLE way to handle it. Today, it's all about bug fixing, Perl programming, product name changes, database issues no one understands (and which they take back after investing so much efforts into a solution that's worse than before -- and while they take back the changes to "what we got before", they break even what we HAD before!).


The recent switch in our 7.4 *BETA* Squeezebox Server builds from SQLite back to MySQL was exactly for that reason ... that we're trying to make sure when the 7.4 software is shipped that it is as stable as possible. We remain committed to migrating Squeezebox Server to SQLite (and there is still a branch of code based on that migration), but we want to make sure that when that change ships, its even more reliable than our existing MySQL system. To that end, we've pushed the SQLite switch off to the 8.0 software release. This was explained in the beta forums when the switch occurred.



Even (external) developers get frustrated -- after investing so many countless hours into repairing things that get broken again next day. (Yes, the code IS great; yes, they have thought about almost everything already; but they keep overlooking SIMPLE things over and over again -- probably because no one has shown them that some things ARE important to USERS. Even with Open Source, goals need to be defined, and milestones met. And everyone in the team needs to know "where we all want to go".)


Sorry, can you elaborate with detail? I'm not sure what you're referring to...



Look at the "new" controller firmware. It LOOKS worse. It's unusable beyond expectation. It needs more keystrokes for simple functions. It doesn't behave logical or even consistent. BUT it has overlaid screens that nobody needs, it overlaps badly rendered cover images with status lines, it uses LARGE fonts so that you have to wait for some information to scroll until -- in the middle of reading -- the screensaver pops in and you have to press yet another key ... Aaaargh! It has a zillion "effects" that DISTRACT (instead of FOCUSING) the user's attention, while lacking some basic functionality. Oh well, and it has "apps"! Does anyone need "apps" while the basics don't function yet? ... I doubt it.

The newest controller firmware for the 7.4 *BETA* software is, to put it simply, "not done." Its an extremely early development build that we've pushed out to the developer community with the intention of getting as many eyes on it as possible. Looks are subjective -- you may dislike the new UI changes, just as I might like them quite a bit, so lets leave looks out of this for now.

Its absolutely a useable firmware, if your a developer. The 7.4 software is considered "unstable" and is only meant to be run by users who are willing to deal with the issues in it, as they come up. There is a reason why we do not push the 7.4 software into the "TESTING" repositories for the Linux builds.

Although you may only see the new UI changes to the 7.4 Controller firmware, it has been hugely re-written. Major upgrades to the way networking, audio playback (headphone jack), and players are handled. Also sweeping re-work of the UI to reduce the number of clicks required to get to various features, and upgrades to features that were just not quite right before.

Lastly, there are huge changes in how the controller handles switching between SqueezeNetwork and Squeezebox Server that although are not yet done, have required significant engineering time to implement. This work has been mostly behind-the-scenes because it simply has not been ready for the community to use until very recently.



It's great to let the "tech kiddies" show off. And they can actually be a very valuable resource -- if LEAD correctly. PLEASE, Logitech, hire someone who knows about USABILITY. And probably how to manage a product line.
It still IS (could be?) such a great product line. Don't throw it away like this.
[/list]

This reads like a rant. It is. Because I still love the product and the ideas behind it. (Emotion, remember?) I hope that the right people can understand my current frustration. And ACT.

I really wouldn't like to see others "winning" in spite of having higher prices and worse technology. But knowing what the customer wants.

Keep the promise! Even if it's only in our heads.

Thank you for reading.

Matt Wise
2009-08-17, 11:59
This sounds reasonable... but my fear is that Logitech will move the development resources from the current 7.4 MySQL platform to the future SQLite "before" the 7.4 MySQL is fully baked. Plugin developers "might" not want to work on a platform that will be sunset in the near future. I think a clearly articulated framework to move the plugins from this platform to the next to lessen the blows would help. I might be way out on left field here... but if the plugins that are developed for the officially released 7.4 were to work without much effort on the SQLite platform... it'd be big step forward. The recent move to SQLite broke too many plugins.

I believe that SQLite development has been put-on-hold until after 7.4 ships.

regalma1
2009-08-17, 12:19
I've been using the SB for a few years. I love it when it works. But I would never recommend it to anyone, at least not without a long list of caveats. These last few years have been full of frustrations and opportunities to keep my vocabulary of expletives well tuned.

That being said, I do have to admit it is getting more stable. I have been able to do upgrades without bringing my system crashing down on me. That is progress. And secondly I have tried to replace it a few times over the years with something that is suppose to be more stable. But I find myself keep coming back due to the UI, which I guess I have just gotten comfortable with. A for the flexibility that I haven't found elsewhere.

The Sonos is an obvious replacement, but I just haven't been willing to put out the $1K. Besides that little screen on their controller leaves me cold. I am using a WiFi equipped 7" netbook to run SC and act as a portable remote control, and am even thinking of going larger (to save my aging eyes). I like being able to see lots of albums or artists listed in front of me. It often inspires to play something I hadn't thought of playing.

Now if you could just come up with a way to stop my netbook losing its connection with my NAS and forcing me to rescan my library. That may be the thing that eventually drives me to Sonos.

Goodsounds
2009-08-17, 12:30
One company employee says that the focus is to be more friendly to average joes. I agree with this, it's the market potential the company needs to serve better. It's also the biggest growth opportunity.

Another says the changes are to make the software perform better on NASes and other smaller formats.

Most average joes I know (and I'm one too) don't have or use NASes. When they do, it's for storage, not for running apps. Has someone really convinced themselves that this deserves a development focus?

These comments contradict one another - which is it? Or maybe the smaller format focus is for unreleased self-contained products and so won't be commented on at this time?

In any event, I think we all appreciate the participation from company personnel.

mherger
2009-08-17, 12:41
> Most average joes I know (and I'm one too) don't have or use NASes.
> When they do, it's for storage, not for running apps. Has someone really
> convinced themselves that this deserves a development focus?

I'm adding a third employee's voice: we've had too many complaints about
SC being a memory hog on all platforms (including Windows), and more than
enough about issues with MySQL and AV scanners and firewalls. Most of our
users will profit from the change. Some more, some less. And a few not at
all.

> These comments contradict one another - which is it?

They don't. If users run into trouble with AV scanners and firewalls
blocking access to the DB, then this is a usability issue.

Michael

maggior
2009-08-17, 13:00
I've been using the SB for a few years. I love it when it works. But I would never recommend it to anyone, at least not without a long list of caveats.

I was actually starting to feel this way. I tried to talk my brother into getting an SB3, but he insisted on the Duet. Of couse I started back-pedalling a little bit with a list of caveats and such. I just knew of too many things that can go awry with the Duet. I even told him - worst case scenario - you return it for a refund.

As it turned out, he got it up and running with no issues. I was very (pleasantly) surprised. It's been running great for him and he and his wife are loving it. Of course he's had a question or 2, but no big deal.

As far as PCs go - he's your average Joe, except that he knows he can fall back on me when things really go south. He did it all on his own, which I thought was a good testament to usability improvements made over that last year for the Duet.

bluegaspode
2009-08-17, 13:04
Hi there,

I'm relatively new to SqueezeCenter and just want put in my two cents:
- I'm an owner of a squeezebox-duet and don't regret my purchase. It's exactly what I expected - and having a 7.3.3-firmware from beginning I didn't feel any glitches. All the basic function were really easy to set up - and it just worked. Thats how it is supposed to be.
I recently stumbled about the headphones not working as I expected but still are expectand that this will be solved soon. In the meantime there is a workaround in this forums - what do I want more ?

- I once got in touch with support, because the Duet didn't want to wake-on-lan my NAS. This was handled very fast by 1st tier and 2nd tier support.

- Two days ago I made the 'mistake' to install FW7.4. This totally broke my controller in the first place - in the meantime with a lot of cursing and I sleepless night I got back to 7.3.3. After some more reading in the beta forums, it turned out to be the language settings, so I tried it again.
And what do I have to say ? I like it ! I can see progress and the direction where my duet will be going - not only is the UI nicer to look at, also switching servers has gotten easier - when I want to browse my music I'm asked which server/library I like - thats a good place (instead of it beeing hidden in some extended menu).

UI is always difficult to explain / get right - but in my opinion its now more stylish - also such subtle things like sounds played when clicking/scrolling have been changed to be much more ear-friendly.
Yes the fonts are to big (as german words are quite a bit longer than the english equivalents) - but thats not a thing I'm troubled about. As they release betas at least there is a chance to complain (and that should be done in beta-forums), before they can't chance anything in the final version.

- I read that one thing to change in the future is server resources consumption. I think this is a good strategic decision. I'm one of those guys running SqueezeCenter on a NAS. That's because I want a low-power, mostly noiseless server. If you see that alternate systems (streamium for instance and most other) all go by UPNP (which are supported by mosts lowtech NAS's out of the box) I think that beeing able to be run smoothly on any NAS is MUST-HAVE for SqueezeCenter. Its still difficult enough to be run on a NAS but if that wasn't possible at all, I wouldn't go with squeezecenter.

- I'm really like the open source character of this product. I'm not programming myself - but I'm able to read this forum/wiki see that others are developing, try their stuff out and I'm able to help myself. Yes that's some time to spent and you need to be a bit more advanced ... but hey - at what other product you get this ?

So my conclusions:
- easy to use for beginners on official firmware
- much more options for advanced users with plugin architecture
- lots to play (and curse) with the cutting edge latest 'beta' firmware.

I hope that my still good mood as a relatively new user stays this way.

pfarrell
2009-08-17, 13:23
bluegaspode wrote:
> I hope that my still good mood as a relatively new user stays this way.

I'm an old user, bought my first SD when the non-G SB1 was new. I've
been in a good mood over the products ever since, and I've bought a
bunch of them.

I don't grok the bad karma in this thread and others. The server
software is open source. That is great. One of the strongest drivers in
my choice of SlimDevices way back when. Sure, Logitech pays a number of
engineers to develope the SqueezeCenter the way they want, but there is
nothing to prevent others from joining and making it work their own way
as well.

I'm on the developers' mailing list, altho I haven't turned a hand in
any real code in years, but its clear that the move to SQlite from MySql
is not canceled, just pushed back to get more work to make it a better
transition that is ready for the mass market users. The use of a
database package allows a lot of nice things that you can't do with raw
tags. But the resource requirements of a complete MySql install exceed a
lot of the low end machines that users want to use. SQlite will be a
better fit.

It would be silly to make the change from MySql to SQlite one way, I
can't imagine anyone doing that. Rather I expect that the official
supported version will use SQlite, and for folks who are running MySql
anyway, there will be some configuration parameter to point to the database.

The reality is that SqueezeCenter is a complex server application. They
all take some care in setup. For most folks, the SC install scripts work
out of the box. Its never going to be a toaster.



--
Pat Farrell
http://www.pfarrell.com/

andynormancx
2009-08-17, 15:35
Most average joes I know (and I'm one too) don't have or use NASes. When they do, it's for storage, not for running apps. Has someone really convinced themselves that this deserves a development focus?

If our guesses about what the SD/USB support on the up coming Touch means then they have very good reason to have been working on making SC run better on lowly hardware. _If_ SD/USB support on the Touch really does mean it can run SC then they'll be a lot more "average joe" users running SC on NAS level hardware.

shake-the-disease
2009-08-17, 15:59
I love it when it works. But I would never recommend it to anyone, at least not without a long list of caveats.

<snip>

I am using a WiFi equipped 7" netbook to run SC and act as a portable remote control, and am even thinking of going larger (to save my aging eyes).

<snip>

Now if you could just come up with a way to stop my netbook losing its connection with my NAS and forcing me to rescan my library. That may be the thing that eventually drives me to Sonos.

The problems you are having look to be nothing to do with SC or your Squeezebox(es). Running SC on a laptop wirelessly connected to a NAS which holds your music files is about the worst possible setup for stability I can think of.

SC, for stable running needs to run on either
1) a dedicated server, with music stored locally, wired to the router
2) a powerful NAS, with music stored locally, wired to the router

I've got such a setup and it's wife-proof (another way of saying rock solid and easy to use)

peterw
2009-08-17, 21:45
SC, for stable running needs to run on either
1) a dedicated server, with music stored locally, wired to the router
2) a powerful NAS, with music stored locally, wired to the router


I'm so tired of this "wired to the router" myth. My Squeezebox server uses a wireless bridge, and all but one of my players connect to the home network via wifi. Only the kitchen player has had trouble, and that's because the microwave would sometimes knock it, *not* the server, off the network. No, I don't play flac or wav. Yes, wired connections are always more reliable, but they are NOT needed. Shoot, I can stream DVD-quality *video* across my G network. 320k MP3, even with multiple players synced, is nothing.

My 2 cents on this thread:
7.3 has been very good to me. 7.4, not so much, but then I usually avoid running the betas anywhere other than my development system until public release seems imminent. I, too, wish there was more guidance about where things are headed. I think communicating plans and documenting APIs has always been an area that could stand improvement, but look at the competition -- closed development, and zero support for 3rd-party enhancements. I'm confident we're still pretty lucky to have the Slim Devices folks running this enterprise.

shake-the-disease
2009-08-17, 22:13
I'm so tired of this "wired to the router" myth.

How about SC running on a laptop wirelessly connected to the music store myth? LOL. OK, I can accept for some people who know what they're doing, a wirelessly connected server can work. For most though, it's going to be trouble. I think wirelessly connecting SC to a music storage is trouble, and doing that from a laptop not dedicated to SC duties is going to be flat out flakey.

cliveb
2009-08-18, 00:42
I'm sure the powers that be must have good reasons for switching from MySQL to SQLite, but a few years ago (back around 2005/6) SlimServer went the other way: switched from SQLite to MySQL. I reckon there must have been very good reasons to make that switch at the time.

What were those reasons, and why are they no longer pertinent?

peter
2009-08-18, 05:33
rtitmuss wrote:
> We will be moving to SQLite in a future release of SqueezeCenter,
> probably 8.0. This change is being made to improve performance,
> especially on NASes and similar devices. It was decided that the SQLite
> changes will not be complete for the 7.4 release, which is why we
> reverted back to MySQL at the moment.
>

Weren't older versions of SlimServer using SQLite?
Doesn't this mean we'll be reverting back to SQLite?

What were the reasons to move from SQLite to MySQL the first time?

Regards,
Peter

snarlydwarf
2009-08-18, 07:19
I'm sure the powers that be must have good reasons for switching from MySQL to SQLite, but a few years ago (back around 2005/6) SlimServer went the other way: switched from SQLite to MySQL. I reckon there must have been very good reasons to make that switch at the time.

What were those reasons, and why are they no longer pertinent?

At the time, I believe it was that performance of SQLite was just not all that hot.

And, yes, there was a lot of "omg this broke everything" and Erland had a mess in his plugins to deal with things the database layer didn't provide an abstract method for him, etc etc.

So, yes, this is a bit deja vu'ish in reverse...

But, then, supposedly SQLite has made great improvements.. it's certainly at least competitive for me (not seeing -huge- improvements, but it's a bit faster than MySQL, other people may have different results).

Matt Wise
2009-08-18, 07:29
The 'why' for most of our decisions can be found in the Developer and Beta forums. I suggest that if anyone is curious to why we've made certain architectural decisions, you check there first.

Dogberry2
2009-08-18, 08:37
I didn't understand the rant, until I read downthread and found out it's about a beta release. I still don't understand the point of "ranting" about a beta release, since one would normally expect problems (else why become a beta tester?), but the Beta forum does seem like a more appropriate place to talk about beta software.

kolding
2009-08-18, 09:02
One company employee says that the focus is to be more friendly to average joes. I agree with this, it's the market potential the company needs to serve better. It's also the biggest growth opportunity.

Another says the changes are to make the software perform better on NASes and other smaller formats.

Most average joes I know (and I'm one too) don't have or use NASes. When they do, it's for storage, not for running apps. Has someone really convinced themselves that this deserves a development focus?


Honestly, you ought to go over and cruise the Sonos forums sometimes. One of the things they beat into the ground constantly is that the Sonos can work with just a NAS, doesn't need a server (like a NAS doesn't have a processor). So, apparently hitting that point where you can have a NAS with no supported computer is useful.



These comments contradict one another - which is it? Or maybe the smaller format focus is for unreleased self-contained products and so won't be commented on at this time?

I fail to see how they contradict each other. SQLLite integrates into SqueezeCenter, which gives you an all-in-one package. No external MySQL to have to worry about. That simplifies it for Joe Sixpack. At the same time, it also reduces the memory footprint, and hopefully the CPU needs, which makes in more amenable to a low-powered machine, such as a NAS. Seems like a win all-around to me.

Eric

erland
2009-08-18, 10:09
7.3 has been very good to me. 7.4, not so much, but then I usually avoid running the betas anywhere other than my development system until public release seems imminent. I, too, wish there was more guidance about where things are headed. I think communicating plans and documenting APIs has always been an area that could stand improvement, but look at the competition -- closed development, and zero support for 3rd-party enhancements. I'm confident we're still pretty lucky to have the Slim Devices folks running this enterprise.

For me the lack of communication and lack of documentation is the biggest problem as a third party developer. However, as you say, if you compare with other companies performing closed source development, Logitech (formerly Slim Devices) is far far ahead of the competition IMHO. The idea of helping Microsoft or Apple with development on my spare time has never crossed my mind.

Changing API's isn't a major problem for developers that are active in the community, the problem is for all previously written plugins that stop working because they aren't maintained anymore and the original developer isn't around anymore.

I'm still worried regarding the lack of communication, there were a lot more communication with third party developers a few years back compared to today. I haven't lost the hope yet, but I fear that the development community might disappear unless something changes. Back in 2006 you pretty much knew someone would answer if you posted in the Developers section of the forum, now I'm often pretty sure no one is going to answer if I post something else than simple questions there.

However, I also understand the reasons. It takes time and effort to communicate with the third party developers and if you need to prioritize between fixing a bug to reach a release date or help a third party developer the prioritization is pretty easy. Most employees tends to prioritize the person paying your salary instead of someone that wants you to do something for free.

I think the problem with the lack of communication is that most third party developers is involved because it's fun and they feel they are part of something. If there isn't any communication, it doesn't feel like you are part of something and the result is that it isn't as fun as it used to be.


And, yes, there was a lot of "omg this broke everything" and Erland had a mess in his plugins to deal with things the database layer didn't provide an abstract method for him, etc etc.

Well, you can't blame Logitech for this.

They provide the database abstraction layer because it contained all functionality they needed for the server implementation. You can't really expect them to make an abstraction layer to cover every possible usage.

It also contains all functionality that I would need in my plugins if I had implemented them the correct way. However, I selected the quick and dirty path to get something out that worked and solved the problem, instead of making the proper solution to invent and implement my own query language for smart playlists. The reason I selected the quick path was that I do this on my spare time and it would take too much time to do the proper solution.

Generally speaking, changes in the perl code has always caused a lot more problems in my plugins than changes in the database engine. This was the situation during the last SQLite->MySQL switch back in 2006 and the current MySQL->SQLite->MySQL switch.



But, then, supposedly SQLite has made great improvements.. it's certainly at least competitive for me (not seeing -huge- improvements, but it's a bit faster than MySQL, other people may have different results).

As a database engine it's a lot better than it was back in 2006 (when I think the MySQL switch was). I remember it had problem running more complex SQL statements back in 2006 but now the same statements works perfectly in the current SQLite version. The only thing I had to change to make the SQL statements in my plugins to work in SQLite in 7.4 was some MySQL specific functions and some joined UPDATE statements.

snarlydwarf
2009-08-18, 11:45
Well, you can't blame Logitech for this.

They provide the database abstraction layer because it contained all functionality they needed for the server implementation. You can't really expect them to make an abstraction layer to cover every possible usage.

Oh, no I don't blame them at all.. your plugins are almost by nature going to push the edge of any sort of schema/backend. If only SQL were a real standard, it would be trivial to use MySQL/SQLite/Postgres/Oracle/DB2/whatever. "Yeah, just use this DB connector and you're done!"

But, alas, it's not.



It also contains all functionality that I would need in my plugins if I had implemented them the correct way. However, I selected the quick and dirty path to get something out that worked and solved the problem, instead of making the proper solution to invent and implement my own query language for smart playlists. The reason I selected the quick path was that I do this on my spare time and it would take too much time to do the proper solution.

Didn't mean it as a criticism of your appreciated work, just the "yeah db backends change, and just like the bug 6.5 change, lots of things broke" dejavu'ishness of it all.

I survived back then (and was happy to switch to MySQL), and even if MySQL support is dropped, I'd still survive (but would have to fiddle with some PHP code I have to list albums.. oh, well).



As a database engine it's a lot better than it was back in 2006 (when I think the MySQL switch was). I remember it had problem running more complex SQL statements back in 2006 but now the same statements works perfectly in the current SQLite version. The only thing I had to change to make the SQL statements in my plugins to work in SQLite in 7.4 was some MySQL specific functions and some joined UPDATE statements.

And the joys of different ways to do 'RANDOM' order by clauses.

SQLite performance on low end hardware is apparently a TON faster than MySQL on the same hardware. On more expensive hardware, I think the difference is less, but if it stops the neverending "why do I get 'cant write to *.MYI' errors" from dumb Antivirus software, that's a huge bonus as well.

The backend db for anything (whether it is SC/SBServer/SS/whatever) is almost always a religious issue.

pfarrell
2009-08-18, 12:01
snarlydwarf wrote:
> The backend db for anything (whether it is SC/SBServer/SS/whatever) is
> almost always a religious issue.

For sure. And the vendors of the DBMS love to sucker you into using
their special extra cool feature.

Writing code to transparently swap between two packages is not all that
hard, but once you start into the religion of "independent" backend
packages, it becomes actually hard.


--
Pat Farrell
http://www.pfarrell.com/

mherger
2009-08-18, 14:16
> However, I also understand the reasons. It takes time and effort to
> communicate with the third party developers and if you need to
> prioritize between fixing a bug to reach a release date or help a third
> party developer the prioritization is pretty easy.

Excellent point. One good approach to improve your chance to get an answer
is to keep postings short and clear. Also in some of the forums it's
almost impossible to keep up with the discussions. Good ideas go lost in
the noise of useless postings. Sometimes it might be worth to restart a
new thread if in posting #938 of a thread there really is some important,
new information. I keep to skip threads based on their subjects after I've
found the first couple of postings unimportant or not interesting for me.

Michael

erland
2009-08-18, 22:28
One good approach to improve your chance to get an answer is to keep postings short and clear

I know, I suspect long posts sometimes tends to be ignored because no one has the time to read through all the text. It feels wrong to post many small threads instead of describing the whole problem in one thread, but I suppose it might be worth a try.


Also in some of the forums it's almost impossible to keep up with the discussions. Good ideas go lost in the noise of useless postings.

Strangely enough, it often feels like the section that I personally get least responses in is the "Developers" section which definitely doesn't contain a lot of useless postings IMHO. Of course, I really don't expect someone from Logitech answer a thread in the "General" section of the forum but I do expect them to answer stuff posted in the "Developers" section.

Sometimes it really better to post a stupid answer like "We currently have no one that has the time to answer this, please get back in a few weeks" instead of just leaving the thread unanswered. This way the poster at least know that there is a reason why no one answers his/her questions instead of getting the feeling that no one is even reading his/hers thread.

The Ubuntu community forum have handled this pretty good by making it easier to find unanswered threads and assigning some forum members with the responsibility to make sure everyone posting a new thread gets some kind of answer or indication where to search further for the information. (No I'm not a volunteer)

It also has to be said that IMHO Logitech employees are often very good at giving feedback on things in the beta related forum sections.

One thing that I've sometimes missed is a way to communicate with Logitech regarding non development stuff. For development stuff it's pretty clear that the "Developers" section should be used but for non developer stuff I currenlty have to use the "General" section where the posting easily could be lost among all the other thread. Some samples of this are questions/comments regarding the Logitech web site or the forum software/structure.


I keep to skip threads based on their subjects after I've
found the first couple of postings unimportant or not interesting for me.

You are not alone, I do the same kind of skipping.

Mnyb
2009-08-18, 22:55
my 2c/

If I'm going to volunteer some time to beta testing, the beta has to be somewhat functional as I only have one server/image no special test system to play around with.
This has been the norm for the last 2 years, the dev's have been slowly introducing new things bugs get found and sorted, nice work flow an everyman can participate as Guinea-pig/crash test dummy without to much hassle .

Migration and preservation of settings/database MUST work.
It also must work when downgrading.
Why ? If it's getting to much you must have a clean retreat route to run and hide a while when the beta is to unstable ;-)
For now I upgraded myself to a tight spot where I'm sure either a up or down grade would blow my settings and require a clean install, the icing is the controller upgrade witch forces me to do something about it.

No more Déjà vu bugs, it's sometimes looks like a zombie movie here, old bugs are resurrected from the dead quite often.
Volunteer's tire quickly if their bug is found and fixed and then reappears 6 months later.
This leads to the conclusion that there is very little rudimentary testing at all before releasing a software ? How is it possible to miss that password protection is not working or that the power off function is gone, or that dhcp is malfunctioning or that dns resolving is broken or non A-Z characters is causing trouble again, easy nobody bothered to try ;)

My absolute favorite in this genre ( why test it should work.. ) was the release of the duet:
Password protected squeezecenter did not work.
Hidden SSID did not work.
Setting up on a network with no DHCP did not work.
No easy way to configure fixed ip's (turning on said dhcp made this impossible ), when actually fixed ip's is the only stable way to run your Squeezeboxes ?
No more can't find.. or blue lights when both server ip and dns is hardcoded into the device eh.
No roaming ? basically large parts of normal network functionality was missing in a release ,
this was all clear to me after using the product for 30 minutes o pro would have seen this in 5 ? I'm unsure of how it is now.

Point, if an average joe with no programing skills can poke holes in software in 5 minutes, it can not have been tested properly.

/2c

Anyway keep up the good work, but a more structured less chaotic way of working would benefit us all i think.
I will still be testing betas for the fun of it.

And now i think the 7.3.4 is very good so they eventually got there almost a year after the release of the duet, so for fuzz free squeezeserver I would use 7.3.4 . Or if possible one the pre sqlite 7.4 beta's which where better, but they to would probably force you to upgrade the controller. I think that the development peaked in the pre sqlite 7.4, that was golden.

egd
2009-08-18, 23:20
Having read the comments made in this thread I'm inclined to agree with the spirit of the intial post. I'll repeat what I posted in the beta forum a while back:
Looking at the 8.0/1 roadmap I don't see the software making any significant functional strides. It seems to me that there are two things happening:
- catering for implementation on embedded systems
- tinkering at the edges of existing functionality to address bugs/ tweak features etc.

All told, when the above is achieved, we're going to have more or less exactly what we have today in functional terms, except plugin developers whose plugins interact with the db are being forced to move support to a different db layer (that is if they want their plugins to continue to operate).

The technological aspect of the Squeezebox offering is no longer a differentiator. The biggest opportunity that's being squandered here is that of content, allowing users to interact with their music collections in ways that are meaningful to them. Someone else is going to do it and that'll herald sunset for the SB line.

No doubt many at Logitech don't want to hear the above, but I genuinely think that is where things are headed. The hardware offerings are continually being expanded but the front end remains pretty much what it is today. Whilst it's good, I don't see it evolving. Folks on the Logitech development team would do well to take a look at products like Qsonix to get some ideas of what shape meaningful interaction with one's music might take.

If Logitech is patently disinterested in taking SC in that sort of direction then at least publish a comprehensive roadmap and finalise the database structure so that adding 3rd party functionality doesn't become a lifetime maintenance project.

radish
2009-08-19, 10:50
Point, if an average joe with no programing skills can poke holes in software in 5 minutes, it can not have been tested properly.

Or it was tested properly, the bugs were found, but someone said "release it anyway". I'm not saying that happened in this case, but it happens all the time with software releases (and is often justifiable to be honest). There are always bugs, it comes down to what you consider a show stopper.

maggior
2009-08-19, 11:35
Or it was tested properly, the bugs were found, but someone said "release it anyway". I'm not saying that happened in this case, but it happens all the time with software releases (and is often justifiable to be honest). There are always bugs, it comes down to what you consider a show stopper.

I've seen that happen too. And yes, All Software Has Bugs! Usually though, there are release notes that will list any known issues that are currently being worked on. That's the way we used to handle it where I used to work.

If Logitech is releasing product with full knowledge that significant issues exist, I think it would be a smart move on their part to do something similar by issuing release notes.

TiredLegs
2009-08-20, 12:57
As a former product manager for consumer electronics devices myself, I generally agree with Moonbeam's assessment, and I commend him for having the gumption to post it, knowing he'd catch some flak.

In my view, a key problem with open source software is that the techies develop it for their own desires, not those of Joe and Jane Sixpack who don't give a rats a** about SQLite vs. MySQL. The masses just want something that works well (even if it has a few bugs), without having to tinker under the hood. Think iPod or TiVo. When both of those products were first introduced, they already had very well thought out (if imperfect) user experiences which kept the technical details hidden from view. You didn't have to be a tech geek to appreciate them, but tech geeks could appreciate them anyway, despite their frustrations at not having full access to the inner workings. I hate to say this, and I'm sure I'll catch flak for it, but Logitech might be better off taking development entirely in house. I especially hate to say it, because the open source platform is one of the things that attracted me to buy Squeezeboxes in the first place.

But Joe and Jane Sixpack don't frequent this forum, and that leads to skewed input into the open source development process unless your target market is open source developers. Sure, I can set up a Squeezebox system that's so easy to use that my girlfriend can handle it and even like it, but no way in heck could she set it up herself. Does anybody here believe that open source development of Squeezecenter will ever lead to a product that is easy enough to set up that his or her mother could do it?

Matt Wise
2009-08-20, 13:07
As a former product manager for consumer electronics devices myself, I generally agree with Moonbeam's assessment, and I commend him for having the gumption to post it, knowing he'd catch some flak.

In my view, a key problem with open source software is that the techies develop it for their own desires, not those of Joe and Jane Sixpack who don't give a rats a** about SQLite vs. MySQL. The masses just want something that works well (even if it has a few bugs), without having to tinker under the hood. Think iPod or TiVo. When both of those products were first introduced, they already had very well thought out (if imperfect) user experiences which kept the technical details hidden from view. You didn't have to be a tech geek to appreciate them, but tech geeks could appreciate them anyway, despite their frustrations at not having full access to the inner workings. I hate to say this, and I'm sure I'll catch flak for it, but Logitech might be better off taking development entirely in house. I especially hate to say it, because the open source platform is one of the things that attracted me to buy Squeezeboxes in the first place.

But Joe and Jane Sixpack don't frequent this forum, and that leads to skewed input into the open source development process unless your target market is open source developers. Sure, I can set up a Squeezebox system that's so easy to use that my girlfriend can handle it and even like it, but no way in heck could she set it up herself. Does anybody here believe that open source development of Squeezecenter will ever lead to a product that is easy enough to set up that his or her mother could do it?

Excuse me ... but how do we not provide a good experience already for the average user? The currently released production-level software 7.3.3 is our most stable software release of the entire 7.x.x releases. We have not encouraged average-users to try out 7.4 (which is in a huge state of flux).

Furthermore, we do not ask that any of our average users know the difference between MySQL or SQLite -- in fact, we do our best to hide that fact completely. Only developers, and highly-advanced beta users are affected by our betas at this time.

We are very actively working towards making our Squeezeplay experience (Squeezebox Controller) as absolutely dead-simple as possible (given its inherent complexities), and improving the setup experience of our Desktop Software as well as our Squeezenetwork service. That is why there have been so many changes to 7.4 that have "broken" things -- we're making huge under-the-hood changes that allow us to provide a better ultimate experience.

TiredLegs
2009-08-20, 14:15
Excuse me ... but how do we not provide a good experience already for the average user?
Matt,

The point you're missing from my post is that your average existing user isn't the average potential buyer (unless the only buyers you want are your existing users). It has nothing to do with beta users vs. regular users, and nothing to do with the particulars of 7.4 vs. 7.3.3.

I'll go back to a question Moonbase asked in his original post of this thread: "where is the target market?" If Logitech has answered that question internally, then is open source development the best path to a product that nails that target? (If Logitech hasn't truly answered that question, then the software will wander at the whim of the developers.)

Pale Blue Ego
2009-08-20, 15:33
Excuse me ... but how do we not provide a good experience already for the average user?

The release of the Duet comes to mind. My original SliMP3 gets daily use while the Duet resides in my sock drawer.

Goodsounds
2009-08-20, 15:43
Tired Legs - Good comments, I agree. Same to Moonbase (relative to his general comments).

Similar comments describing a seemingly disinterested attitude concerning the Joe Sixpack market segment (the largest portion of the potential market, by the way) have been made before. And, I recall, were also responded to by saying that they are "making the installation process easier", which is NOT the point.

It may be that the company doesn't share the view or doesn't want to change its direction.

Matt Wise
2009-08-20, 15:52
Tired Legs - Good comments, I agree. Same to Moonbase (relative to his general comments).

Similar comments describing a seemingly disinterested attitude concerning the Joe Sixpack market segment (the largest portion of the potential market, by the way) have been made before. And, I recall, were also responded to by saying that they are "making the installation process easier", which is NOT the point.

It may be that the company doesn't share the view or doesn't want to change its direction.

Sorry, perhaps I'm misunderstanding. What is it that you currently think Logitech's long term goals with Squeezebox products are -- and how do those differ from the direction you think we should be going?

edit: (keep in mind, I'm asking so that I can understand your POV a bit better ... not because Logitech is unsure of its own direction with the product.)

erland
2009-08-20, 16:32
In my view, a key problem with open source software is that the techies develop it for their own desires, not those of Joe and Jane Sixpack who don't give a rats a** about SQLite vs. MySQL. The masses just want something that works well (even if it has a few bugs), without having to tinker under the hood. Think iPod or TiVo. When both of those products were first introduced, they already had very well thought out (if imperfect) user experiences which kept the technical details hidden from view. You didn't have to be a tech geek to appreciate them, but tech geeks could appreciate them anyway, despite their frustrations at not having full access to the inner workings. I hate to say this, and I'm sure I'll catch flak for it, but Logitech might be better off taking development entirely in house. I especially hate to say it, because the open source platform is one of the things that attracted me to buy Squeezeboxes in the first place.

If we would talk about some open source project driven by the community I would partly agree with your comments but here we talk about an open source project driven by a company. In an open source project where the community makes the decisions a lot of your comments is valid, but this definitely isn't the case here.

You have to understand that open source doesn't mean that the community makes the decisions or even that the decisions are available to the public.

Logitech takes 100% of all decisions regarding the development of Squeezebox Server themselves and even though I'm not involved in them I'm pretty sure most decisions taken are based on business advantages/disadvantages. The code of Squeezebox Server is mostly written by Logitech employees and it's written the way it is because some manager within Logitech has told the employee they want a specific feature or solution.

Sure, Logitech sometimes listens to developers AND users which aren't employees, but there is a huge difference between listening in on other developers/users and letting the other developers/users take the decisions.

So please, don't blame the open source model for the decisions made.

The only difference as far as I can see if this was a closed source project would be that:
- Logitech would have a hard time to get feedback on code and decisions from outside developers.
- Logitech would have a hard time to get free help from third party developers.
- Logitech would have a harder time getting people to help them testing the product without economical compensation.

The open source model makes it possible for Logitech to get help from people outside the company.



But Joe and Jane Sixpack don't frequent this forum, and that leads to skewed input into the open source development process unless your target market is open source developers. Sure, I can set up a Squeezebox system that's so easy to use that my girlfriend can handle it and even like it, but no way in heck could she set it up herself.

I completely agree.

If Logitech wants to target Joe and Jane Sixpack, they definitely need to look outside this forum for feedback. I'm sure they already have these kind of channels, we just don't know because we only sees what's posted in this forum. So please, don't make the mistake to think that this forum is the only input Logitech gets.



Does anybody here believe that open source development of Squeezecenter will ever lead to a product that is easy enough to set up that his or her mother could do it?

Once again, don't blame the open source model for this.

Logitech can still keep the open source development model, they'll just have to make sure this forum isn't their only input when making the decisions.

If we would have talked about an open source project where the decisions were taken by the community and where the decisions was taken based on other things than what's good for the business, your comments are completely valid. However, this isn't really the case here.

Don't get me wrong, I still party agree with you on this point. As long as Logitech only makes decisions which the users on this forum is happy with, it will be a hard time to make the Squeezebox products user friendly. However, the open source model isn't the cause to this, the situation would be exactly the same if this was a closed source project and Logitech only used the community for beta testing and feedback. The reasons a lot of options exists in Squeezebox Server IS because beta testers has requested them, not because this is a open source project.

IMHO, to solve this Logitech needs to start taking decisions which the community isn't 100% happy with, the reason is that the current community member might not be the typical future user. There will be complains when this happens, but as long as the decisions are explained I'm sure most people on these forums will agree after a while.



I'll go back to a question Moonbase asked in his original post of this thread: "where is the target market?" If Logitech has answered that question internally, then is open source development the best path to a product that nails that target?

Yes, I believe it is



If Logitech hasn't truly answered that question, then the software will wander at the whim of the developers.

Once again, I completely agree.
If you haven't set the goal, it's hard for everyone to make the right decisions.

Of course, just because Logitech hasn't officially made the goal and target user public, it doesn't have to mean they don't have it internally.

However, a problem with only having the goal and future target user specified internally is that it's harder for beta testers and third party developers to help. If there isn't a public available goal and target user, beta testers and third party developers tends to give feedback in a direction that's good for themselves. I'm pretty sure this is a big reason why the current software looks like it does, there really hasn't been any clearly defined goal and target user available to the beta testers and third party developers.

Of course, sometimes there are reasons why you like to keep goals and target user internally:
- You might want to hide what future products you are working on
- You might not want to scare everyone away since they might not match the future target user category
- You might want to get feedback based on what a beta tester thinks is good for himself instead of requesting that he/she tries to be someone else

erland
2009-08-20, 16:50
Tired Legs - Good comments, I agree. Same to Moonbase (relative to his general comments).

Similar comments describing a seemingly disinterested attitude concerning the Joe Sixpack market segment (the largest portion of the potential market, by the way) have been made before. And, I recall, were also responded to by saying that they are "making the installation process easier", which is NOT the point.

It may be that the company doesn't share the view or doesn't want to change its direction.

Maybe Joe Sixpack isn't the target user for the current release ?

I've never seen Logitech stating that they are targeting Joe Sixpack and remember just because there are a lot of them it doesn't mean that they are easy to reach. Joe Sixpack might not be the easiest way to increase the economical result for Logitech.

To reach Joe Sixpack, a lot of other things might be needed than just making the Squeezebox products easier to use.

IMHO, there are two ways Joe Sixpack would be able to use the Squeezebox products:
- If he/she only used streaming services and it was very easy to connect it to the network, it probably needs to be bundled with a pre-configured router to make this possible.
- If Logitech solved the whole ripping part to make it easy to make that CD available to the Squeezebox system

You have to remember that Joe Sixpack is used to turning on the radio or to buy a CD and insert it into a CD player, where the CD player or radio just needs to be plugged into the wall to work. As long as it's harder than this to use the Squeezebox products, I'm pretty sure Joe Sixpack will prefer his radio or CD player. We are not talking about someone that knows what a MP3 file is and we are not talking about someone that used to work with a computer or wireless networks.

I think this type of user can be very hard to reach so Logitech probably have to do this in steps. To me, the first logical step seems to make it possible for users that at least are familiar with MP3 files to use the products and to make it easier for users that only use online streaming services to setup the device in the network.

radish
2009-08-20, 17:43
Joe Sixpack listens to his ipod on a JBL speaker dock and considers it high end. He's not going to pay more than $100 for anything which streams, and it better do video at that price.

Thinking that the only way to make a profit is to target your product to the lowest common denominator is a mistake, often a fatal one. Sure volume is higher, but so are support costs and margins are tiny. I have no more insight into the inner workings of Logitech than anyone else, but my impression is that they see the SB line as a somewhat premium, audio focused line for people who care a bit more than the average Joe about music. There are plenty of manufacturers who don't cater to Joe who do pretty well - I mean just look at Sonos. They might argue they're easier to setup (I have no evidence either way, just going on marketing spiel) but at the price they charge Joe Sixpack isn't going to give it a second look.

dsdreamer
2009-08-20, 20:29
There is plenty of room in the market between $100 Airport Express clients used as Airtunes endpoints, and a Sonus or even Meridian/Soloos system if it comes to high end.

The positioning of Logitech / Slim products is fine by me in terms of price point.

What is less clear to me is whether this line of products can acquire the ease of use and robustness needed to make this, a fun product for the average public school teacher to own. It's fine to aim at professionals, but it would be great if the majority of enthusiastic owners were not necessarily engineering professionals.

It's interesting how the Open model of software development gives users the feeling of being somehow part owners of the company, even expecting they can influence strategy and company culture with nothing more than a well-crafted forum posting! There's plenty of things I wish were different about the way these guys develop and release software, starting with making them all take the Hippocratic oath to "first, do no harm." In other words, thou shalt not regress the codebase. But then I have to remember it's not my company, I will not be held accountable for what is released. It's not my problem. Now back to the music...

iPhone
2009-08-20, 20:48
With SC as large as it is and with all it can do and users asking for more, at this point it is almost impossible to catch everything that could/might/are affected by a Code change. Can you image Slim Devices trying to do what they do without the Forum and the Beta user community? With all the features, devices available, plugins, etc, it would be next to impossible for even a group of people to test every single feature, in every available configuration, between Production releases much less before the next Nightly release. The Beta Community submits bugs and catches what it can to help with this process.

Why do you think Roku and Sonos don't release updates very often? Part of it is that its not open source, but part of it is if its not a major major bug, they're not fixing it until 'they' get around to it.

Lastly, it is open source, so if one doesn't like something, change it. Its never to late to learn a new programming language.

Goodsounds
2009-08-20, 21:34
With SC as large as it is and with all it can do and users asking for more, at this point it is almost impossible to catch everything that could/might/are affected by a Code change.

I'm glad the software people at Boeing and Airbus don't have this attitude! ;-)

C'mon, iphone, it's a music player, not something relatively complicated (in the grand scheme of things). Most hardware products have substantial software content and/or companion software (even semiconductors do). The bigger or more complicated the hardware, the bigger and more complicated the software. The vast, vast majority of companies handle it internally.

What is it about these products that lead you to think that this approach is not an outlier? Fact is, it's rare. I think it contributes to some of the negatives others have discussed so eloquently.

Mnyb
2009-08-20, 21:42
With SC as large as it is and with all it can do and users asking for more, at this point it is almost impossible to catch everything that could/might/are affected by a Code change. Can you image Slim Devices trying to do what they do without the Forum and the Beta user community? With all the features, devices available, plugins, etc, it would be next to impossible for even a group of people to test every single feature, in every available configuration, between Production releases much less before the next Nightly release. The Beta Community submits bugs and catches what it can to help with this process.

Why do you think Roku and Sonos don't release updates very often? Part of it is that its not open source, but part of it is if its not a major major bug, they're not fixing it until 'they' get around to it.

Lastly, it is open source, so if one doesn't like something, change it. Its never to late to learn a new programming language.

There are degrees to things, logi could do some rudimentary testing with a clean product without the most complicated plugins. So a basic test with installing the sw on officially supported platforms setup a bunch of players scan a test library, play some music etc (using consumer grade routers on DCHP ;) with pw protection on in SC ) . Just some normal sanity checks input vs output compared to a specification.
A beta is supposed to be a *functional* software with bugs known or unknown, betas are perfect for finding the thousands of small issues that would require months of in-house testing to find.

erland
2009-08-20, 22:38
With SC as large as it is and with all it can do and users asking for more, at this point it is almost impossible to catch everything that could/might/are affected by a Code change. Can you image Slim Devices trying to do what they do without the Forum and the Beta user community? With all the features, devices available, plugins, etc, it would be next to impossible for even a group of people to test every single feature, in every available configuration, between Production releases much less before the next Nightly release. The Beta Community submits bugs and catches what it can to help with this process.

Finding every bug can't be done in an economical way, the quality is related to the cost of testing it.

For each product you will have to decide which quality you can afford to reach. As an example if there is a major bug in some software in an aircraft your company is going to be heavily affected after the aircraft crash. It's not going to make the people that was inside the aircraft and their relatives happy if you fix the bug a week later. With a music player a bug will cause a bit of frustration but often people are happy when it works a week later.

I think there are a number of reasons why we have the current situation, like:
================================================== ==========
1.
There isn't a specification that describes how it is supposed to work, much of the functionality seems to be in the head of different people. This is especially true if we talk about how it should behave with different combinations of settings and tagging schemes.

A result of this is that you don't remember to test everything because you don't remember the functionality exists. It's also hard to prioritize correctly during your testing work.

They started to do this once but I don't think it was finished:
http://wiki.slimdevices.com/index.php/Slim_Server_Specification_(6.5.1)
http://wiki.slimdevices.com/index.php/SlimServer7Spec

I'm not sure if the above specifications are used anymore, someone from Logitech would have to answer if there is any point for us to help correcting them.


2.
There aren't any official test cases available to the beta testers, so it's very hard for the beta testers and volunteers to do any structured testing without inventing the wheel themselves. I hope at least the Logitech test team has some kind of structured test cases, if they don't there should be a huge improvement potential.


3.
Some parts of the code isn't automated tested, some because it's hard to do automated testing on it and some because no one has bothered yet. Automated test is really the best way to do testing on all functionality.


4.
With the current amount of resources assigned to testing within the Logitech team it's hard or even impossible to get the quality some people expect. I still think there is a improvement potential but IMHO it's related to the first two points.

So what can we do ?
====================
- Complain in the forum and tell the Logitech people how to do the work like I just did ? This really doesn't help much because I'm pretty sure they already know what the problem is. I'm guessing that's the reason it hasn't been done already is because they've had to prioritize other things.

- Letting Logitech know if you think the quality is to low so they know if they have to improve or not. However, remember that it's always up to Logitech to decide if it's worth the cost or not. Sometimes it's easier to raise the income by adding more functionality than raising the quality of existing functionality.

- Ask Logitech to stop the development for a year and only focus on testing every functionality that already exists ? With the result that the competition catches up and leaves Logitech behind...

- Use the beta version and register bugs at http://bugs.slimdevices.com when we find something that doesn't work.

- Buy more Squeezebox devices so Logitech can afford to add more people to their test team.

- Help updating the above mentioned specifications on the wiki (if Logitech uses them).

mherger
2009-08-21, 00:12
> Joe Sixpack listens to his ipod on a JBL speaker dock and considers it
> high end. He's not going to pay more than $100 for anything which
> streams, and it better do video at that price.

If Joe Sixpack has an iPod with tons of music, then he's not the same Joe
somebody else described (the guy who can't rip a CD, pirate some MP3
files, buy music online).

We call him the "music lover". See
http://www.logitechsqueezebox.com/how-it-works/overview.html for more
details (hey, web sites sometimes offer some the information you claim to
lack ;-)).

--

Michael

mherger
2009-08-21, 00:32
Great posting, Erland.

> There isn't a specification that describes how it is supposed to work,
> much of the functionality seems to be in the head of different people.

Logitech SMBU isn't SlimDevices any more. At the time there were Sean,
Dean and a few enthusiasts. They did the hardware, the software, the
marketing, sales... Today we're part of a large company. There's worldwide
marketing, sales, hardware engineering, software etc. all in different
departements, all with many people working there. This doesn't simplify or
speed up communication, that's for sure.

> There aren't any official test cases available to the beta testers, so
> it's very hard for the beta testers and volunteers to do any structured
> testing without inventing the wheel themselves.

That's something we definitely want to improve.

> Some parts of the code isn't automated tested, some because it's hard
> to do automated testing on it and some because no one has bothered yet.
> Automated test is really the best way to do testing on all
> functionality.

Another point QA's working on. Eg. the file scanning needs automated
testing to recognize regression in eg. character set issues early. We're
trying to build the world's largest collection of broken music files :-).

> So what can we do ?
> ====================
> - Complain in the forum and tell the Logitech people how to do the work
> like I just did ? This really doesn't help much because I'm pretty sure
> they already know what the problem is.

Thank you, thank you, thank you.

> I'm guessing that's the reason it
> hasn't been done already is because they've had to prioritize other
> things.

Correct. Product development often is a trade-off between cost and
quality. Somebody mentioned development of an airplane control system.
That's not the mass market product you want to be selling for a few
hundred bucks. You've got more time and resources for quality. But if you
want to sell a 200USD device you can't afford hundreds of developers, QA
guys etc. We're working hard to make our products the best product you can
have for the money. And that's no easy task.

> - Letting Logitech know if you think the quality is to low so they know
> if they have to improve or not. However, remember that it's always up to
> Logitech to decide if it's worth the cost or not. Sometimes it's easier
> to raise the income by adding more functionality than raising the
> quality of existing functionality.

Somebody explain the 80/20 rule...

> - Ask Logitech to stop the development for a year and only focus on
> testing every functionality that already exists ? With the result that
> the competition catches up and leaves Logitech behind...

You can ask us to do this. But then new product sales are better money
than improved quality in already sold devices... it's a sad fact we're
fighting with marketing every day.

> - Buy more Squeezebox devices so Logitech can afford to add more people
> to their test team.

That's definitely the way to go :-).

--

Michael

amcluesent
2009-08-21, 01:31
>But then new product sales are better money than improved quality in already sold devices<

People here always use Apple as a great example of producing Joe Sixpack friendly systems.

Maybe you should follow Apple and monetise upgrades by charging $10 for SqueezeCenter 8? Start the Plugin Marketplace for selling plugins at $4.99 , make SqeezeNetwork a subscription service.

toby10
2009-08-21, 03:09
JOE SIXPACK:

Read some of the Amazon customer reviews of SD/Logitech players, some are quite funny. A few paraphrased examples:
- "...it doesn't work with WiFi like my computer does..."
- "..why would I pay $299 for a Boom when I can buy a used laptop and add speakers..."
- "..to get radio & Music Services you must expose your connection to the internet, too much a security risk..."
- "...no AM/FM? I listen to NPR so this is a deal killer...."

These common Joe's are obviously not the target market of SB players. They are quite content with their iPods, just as my father has no interest in learning how to load a CD into his 16 speaker car stereo system. It is just not the SB market.

As for simplifying the use of SC, just look at the Extension Downloader. SD and third party Plugin customization (add, remove, install, update) a wide selection of choices & options, all automated. To me this clearly illustrates a commitment by SD to simplify and better mainstream it's product line, without dumbing down the hardware.

Room for improvement? Always! But from a non-programmer and less than tech savvy consumer I sure like the direction SD is headed. :)

radish
2009-08-21, 06:07
If Joe Sixpack has an iPod with tons of music, then he's not the same Joe
somebody else described (the guy who can't rip a CD, pirate some MP3
files, buy music online).

I think that's a key insight. Joe Sixpack is actually lots of different people, and you know what they say about pleasing all of the people.



We call him the "music lover". See
http://www.logitechsqueezebox.com/how-it-works/overview.html for more
details (hey, web sites sometimes offer some the information you claim to
lack ;-)).

Well given that I hadn't read that page I guess I was pretty close :)

dsdreamer
2009-08-21, 07:39
Excellent aspiration. Very hard to do.

I noticed thet Sonos has a "Design Director for User Experience" in the recent demo video they posted: http://www.sonos.com/demo/demo.aspx

Does Logitech SMBU have an equivalent role?

mherger
2009-08-21, 07:55
> I noticed thet Sonos has a "Design Director for User Experience" in the
> recent demo video they posted: http://www.sonos.com/demo/demo.aspx
>
> Does Logitech SMBU have an equivalent role?

Hehe... titles, names...

Juliet:
"What's in a name? That which we call a rose
By any other name would smell as sweet."

So... our man's called "Senior User Interface Designer"

--

Michael

dsdreamer
2009-08-21, 13:51
Hehe... titles, names...

Juliet:
"What's in a name? That which we call a rose
By any other name would smell as sweet."

So... our man's called "Senior User Interface Designer"

--

Michael

Well not much is in a name, but the overall user experience goes well beyond the UI. And being a director of that overall user experience would allow someone to call the shots across a cross-functional team and to be the voice of the customer as regards things like delivered quality, user support experience, out-of-box experience, etc, etc, which is not quite what a senior UI designer would normally get to do. Someone thinking holistically about the complete delivered user experience including, but not limited to UI, in other words.

TiredLegs
2009-08-22, 07:22
Well not much is in a name, but the overall user experience goes well beyond the UI. And being a director of that overall user experience would allow someone to call the shots across a cross-functional team and to be the voice of the customer as regards things like delivered quality, user support experience, out-of-box experience, etc, etc, which is not quite what a senior UI designer would normally get to do. Someone thinking holistically about the complete delivered user experience including, but not limited to UI, in other words.
This is a very well stated observation by dsdreamer. At the company where I used to work, the CEO referred to this as a product's "gestalt", which effectively means that the whole of a product is greater than the sum of its parts.

ChrisOwens
2009-08-24, 13:38
All the points that everyone has brought up, they *are* real problems for us and we (mostly) know it. At least a few times a week, for instance, we make decisions including the consideration that we want the community to continue helping us and being involved in testing new software/products and in contributing to the product (with patches or plugins.)

I think we're trying to find the 'right' solution between Joe Sixpack ease-of-use and hackability. Erland's certainly right that complaining helps, and we can take an action-item to communicate back, too. :)

It seems sometimes like there's a clear trade-off between closed-source, reliable, limited, friendly products and those that are open-source, unreliable, hacker-friendly, but user-hostile. However, I think there are good paths down the middle as well. Apple does have success with the first course, but it's far from crystal clear to me that their success is DUE to those four things alone. We have to develop our own synthesis (or gestalt!) for why our products are better, and I think help from the community and an open source server can continue to be part of that.

Software quality is always a challenge, and we have an extremely small team to try to make it work. As you've no doubt realized, community-filed bugs are an incredibly important part of it. I was able to hire another QA engineer recently despite the economy, and he's freed up some additional time for me. We have a plan for test automation and sharing test plans with the community so that they can help contribute (something Dean and I have wanted to do for years), and I'll post separately on those as they get closer.

I feel like I have to point out there have been some big improvements since the acquisition, too. The in-house development team is many times as large as it was when I joined Slim! It was Dan Sully and Dean part time and Andy part time (when he wasn't doing SN ops)... and there was a period when it was only Dean part time and Andy part time and the community! I sure don't miss the days when I'd find a bug and file it in Bugzilla, but then didn't have anyone to assign it to. And to have been able to hire so many engineers who already knew the product and had an interest in improving it is unheard of in my career!

As Michael said, we're not Slim Devices any more, but there is still a core spirit left from those of us that joined before the acquisition, and from new team members that know that Logitech bought us because we had something special, and whatever we do next, we don't want to lose the special! There have been some personnel changes, but there's still a lot of the original flavor remaining, in my opinion, even if some of us have to spend more time working on detailed budgets and attending meetings in Fremont than we did in the Slim days.

Ben Sandee
2009-08-24, 14:27
All well and good Chris, but short of turning complete operation of
your company over to the forums-most-vocal, it will be insufficient.
Sorry.

Ben

ajayrav
2009-08-24, 14:29
I originally bought my Squeezebox duet so I could play it through my Musical Fidelity X-DAC-v3 quad stack (power supply, DAC, tube buffer, headphone amp). The MF DAC has auto-input sensing. However, since the duets outputs are never completely zero, the DAC always senses an input even when the duet is turned off and the appropriate option in the player set-up to disble outputs when off is turned on. A bug 1397 was filed and I waited for a fix. Target milestones kept getting rescheduled and finally the bug number changed to Bug 10077. To this date it is still unassigned.
I'm REALLY disappointed.

Here are some excerpts from the bug filing

Richardson James 2008-11-19 11:31:25 PST
BUG 571, Comment #75

Customer is requesting that the S/PDIF ports be turn electrically off when the
player is in an 'off' or standby state.Comment 1 Richardson James 2008-11-19 11:43:02 PST

Correction: Bug 1397, Comment #75Comment 2 Sean Adams 2008-11-19 14:05:32 PST
This bug concerns a further enhancement of the "disable outputs when off"
capability. Some receivers still interpret our S/PDIF output as being active
when we try to disable them. These receivers seem to be detecting the input
electrically, as opposed to using an s/pdif receiver chip to test for a valid
signal.

What is desired is that when disabling s/pdif outputs, that we not merely
ceaqse transmitting a valid carrier, but that any electrical modulation of the
outputs be stopped. The coax output would go to 0 volts DC, and for the LED
would turn off.

Note to Slim QA: the Digliyzer can NOT be used to test this. It doesn't know
the difference between "no carrier" and "no voltage".

Comment 3 C--- Ba--- 2008-11-21 08:51:38 PST
In addition to the need for 0V on the output to support some DACs, there is
another useful side-effect of doing this. The presence or absence of a voltage
on the SPDIF output could be used (via a suitable external device) to trigger
power to ancilliary devices such as amplifiers. Felix's PowerSwitchII plugin
can be used for this purpose on the Squeezebox Classic and Transporter, but not
on the Squeezebox Receiver (since it has no headphone output).

Comment 4 Blackketter Dean 2008-11-23 12:18:05 PST
Consider for 7.3.1. Sean?

Comment 5 A--- R--- 2008-11-29 10:46:34 PST
Still not able to automatically switch my Musical Fidelity X-DAC v3 by turning
off receiver inputs. I guess the DAC uses current sensing rather than spdif
receiver chip to check for signal. I'll have to wait for 7.3.1 :-(

Comment 6 Richardson James 2008-12-19 08:03:02 PST
Changing target to next release

Comment 7 S--- McV---- 2009-04-14 19:21:33 PDT
As with comment #5, I have a Mobile Fidelity X-DAC V3, which I have hooked up
to the digital output of an SB3. (running 7.3.2) I have the player's settings
configured to "disable outputs when off." When connecting the player to the
DAC with the SB3's power off the DAC still detects a live input. This occurs
with both the coax and optical outputs of the player.

As a sanity check, I also connected the DAC to the digital output of my DVD
player. As expected, with both digital and coax inputs, the DAC indicates that
the there is no live input when the DVD player is powered down. (confirming
that the DAC is able to accurately identify the presence of a live input)

peterw
2009-08-24, 16:43
I hope you don't mind, I ran a couple sections of your post through Babelfish...



It seems sometimes like there's a clear trade-off between closed-source, reliable, limited, friendly products and those that are open-source, unreliable, hacker-friendly, but user-hostile.

"SqueezeCenter is junk. Every day I rue our decision not to use Windows Mobile for the Controller. Unless you are a masochist, you would be wise to buy gear from Sonos. Or Apple. Anyone else, really."


Christopher Owens
QA Manager

"Trust me, nobody knows better than I just how awful our products are."

Seriously, what are you talking about? Have you been comparing iPhone to OpenMoko's FreeRunner or something? :-)

ChrisOwens
2009-08-24, 17:05
Sorry if I wasn't clear. I am under no illusions that our products are as high-quality as Apple's.

I was trying to say that I don't think in order to have a high-quality easy-to-use product that it's necessary that we stop being open source or hacker-friendly.

maggior
2009-08-24, 18:05
I was trying to say that I don't think in order to have a high-quality easy-to-use product that it's necessary that we stop being open source or hacker-friendly.
I agree!

Apple has it's issues - it's certainly not like their products are problem free. When my wife got her 2G 8GB nano, it had the lovely bug that the on-the-go playlist didn't work. Yes, they fixed the problem eventually, but it was certainly very annoying since that was a feature we used and relied on.

I certainly perfer the situation with the Squeezebox where things may not be as stable, but features are updated and added regularly, some based on user feedback and input. I don't hold my breath waiting for Apple to support FLAC on the iPod. If some new wizbang lossless CODEC takes over the audio universe, I can count on Logitech to support it :-).

I see where Chris was coming from in his post and I appreciate his candor. I hope you guys find that balance.

radish
2009-08-24, 19:24
Sorry if I wasn't clear. I am under no illusions that our products are as high-quality as Apple's.


Compare like with like. A Squeezebox _is_ higher quality than an Airport Express, however you measure it. When Logitech/SD start making laptops, then we can compare those :)

erland
2009-08-24, 22:35
It seems sometimes like there's a clear trade-off between closed-source, reliable, limited, friendly products and those that are open-source, unreliable, hacker-friendly, but user-hostile.

It's all about who decides what to do and how to do it.

In many open source projects, the decisions are made by one or several developers which wants the software to be usable in all possible scenarios. It's more important to make it work for everyone than to make it work really good for a smaller part of the crowd. Features that are cool or interesting to implement often get higher priority than features that improve the usability. Advanced users get exited when they realize all possibilities and the developers get excited when they realize someone likes it and implement more advanced stuff. Look at my plugins for a great example on this, very flexible but many users don't even bother to try because they are too complex to configure.

However, IMHO SqueezeCenter isn't the typical open source project. There is a huge difference when a company makes all decisions and the decisions are based on what reaches the company strategy and gives economical results in the end.

In my opinion, Logitech has to improve on deciding on a strategy and making decisions based on that strategy. This has nothing to do with if the software is released as open source or closed source.

However, as long as the strategy only is available internally, there is a huge risk the community will do its best to try to force you another way because they don't know why you take certain decisions. I'm not saying that making the strategy public would make the community follow it without complaints, but it would probably explain some things and make it easier to understand some decisions. Of course, sometimes it really feels like there isn't a strategy available even internally, but that's probably just because you implement it in smaller steps so we don't see the forest for all the trees. Yes, I also do understand that it might be a bad idea to make the strategy public since it will also make it available for all competitors.



However, I think there are good paths down the middle as well. Apple does have success with the first course, but it's far from crystal clear to me that their success is DUE to those four things alone. We have to develop our own synthesis (or gestalt!) for why our products are better, and I think help from the community and an open source server can continue to be part of that.

I agree.

I'm pretty sure Apple's success is because they work a lot with usability. Limiting features is part of this but usability involves a lot more things than just limiting the number of available features. Usability isn't easy and you won't success unless you have someone that works with it. As mentioned earlier in another context in this thread, user interface design is one part of the formula but there is a lot of other stuff involved too. Apple has a clear strategy that's visible in all their products, it starts with physical design but you will also see it when using the products. Unfortunately, I'm not sure I can say the same for the Squeezebox products yet but I hope we will get there.

I think anyone blaming the community or the open source philosophy for your problem is on the wrong road even though it probably sometimes feels like an easy way out. I'm pretty sure you would have exactly the same problems plus a number of new ones if you decided to go closed source and ignore the community.

A problem with the community which you probably already know is that most persons in the forum might be the target user of the old products but they might not be the typical user you are targeting today and in the future, especially if you are moving in the Joe Sixpack direction. Some examples of this are some poll threads that showed that Linux was the most used operating system for SqueezeCenter and that Google Chrome Beta is used by a large percentage. I'm pretty sure Windows and IE/Firefox is what your typical user use.

However, I think you have to learn to not listen to the community in all situations. Of course, this isn't easy because to make the community stay they also need to feel that you listen to them.

pfarrell
2009-08-24, 22:47
erland wrote:
> However, IMHO SqueezeCenter isn't the typical open source project.
> There is a huge difference when a company makes all decisions and the
> decisions are based on what reaches the company strategy and gives
> economical results in the end.

I'm not so sure. FireFox is open source, but they get huge amounts of
funding from Google. NetBeans and Java are open source, but got huge
amounts of funding/engineering time from Sun.

In both cases, I don't think decisions are as community driven as you
are claiming.

Even IETF and IEEE standards are defined by folks who vote at the
meetings, and those same folks have their time and travel paid for by
some companies. Sure, there are a few pure amateurs, but most of the
folks are representing a view of their supporters and sponsors.

> I'm pretty sure Apple's success is because they work a lot with
> usability.

I may be quoting you out of context here, but Apple is the best example
I know of a completely closed source company that extracts premium prices.

Sure, they have great design and great usability. But they are not open
in any sense of the word.

I'm not sure that there will be any light shed on this thread if it
continues, as it seems to be mostly smoke. Starting with the subject
line of the thread. It assumes that a rant is justified and that a
rescue is needed and the need for this rescue is seen by all.

SlimServer/SqueezeCenter/SqueezeBoxServer or whatever its called is by
no means perfect. But I am not seeing how this effort here is helping it
move anywhere.


Pat

p.s. I absolutely hate the renaming of the server. Hated it the first
time, do now, and probably always will. Branding, bletch.


--
Pat Farrell
http://www.pfarrell.com/

TiredLegs
2009-08-25, 04:37
I think you have to learn to not listen to the community in all situations. Of course, this isn't easy because to make the community stay they also need to feel that you listen to them.
This can be one of the most difficult things for a company to do: to prioritize the anticipated needs of a conceptual target market over the existing desires of real users. Yet for a company to grow beyond the early adopter type of users, it's often crucial to get to Joe Sixpack, or perhaps in this case, Joe Latte.



In my opinion, Logitech has to improve on deciding on a strategy and making decisions based on that strategy.
At small companies, like pre-Logitech Slim Devices, that strategy usually comes from the top, e.g. Sean's evident leadership and personal care for the success of the products. In most large companies, that charge is the role of product management, which can be especially challenging if it's a new category of product that nobody's really sure what will make it succeed. At Apple, Jobs probably sets the strategy himself, which is more like the way small companies operate.

Mnyb
2009-08-25, 06:20
Are not things better than they *apear* to be:

*They are developing at least 2 new products, that we know of trough leaks.
*They are doing several big changes in the UI which is obvious when you are trying the newest controller beta.
*They are developing a completely new UI for the squeezenetwork.
*They are basically redesigning the product line.

So when when we are seeing that the software roadmap in the wiki never gets updated, and it is very dated, and a general lack of info on whats coming in the future. Then at the same time the beta sw quality is at new all time low, this could mean that everything is going south ?
But i don't think so.

There is just an information vacuum of some kind, I think I lack some vital information to interpret this.

This will be much clearer when we see the new products I hope ?

erland
2009-08-25, 09:34
erland wrote:
> However, IMHO SqueezeCenter isn't the typical open source project.
> There is a huge difference when a company makes all decisions and the
> decisions are based on what reaches the company strategy and gives
> economical results in the end.

I'm not so sure. FireFox is open source, but they get huge amounts of
funding from Google. NetBeans and Java are open source, but got huge
amounts of funding/engineering time from Sun.

In both cases, I don't think decisions are as community driven as you
are claiming.

I consider FireFox, NetBeans and Java to be in the same category as SqueezeCenter, in all these products I suspect decisions aren't made by the community. That's also a reason why they've all in a way succeeded to be something more than software for geeks.

I guess SqueezeCenter might not be as unusual as I made it look like. I was just trying to say that there are different kinds of open source software, some succeeds and some doesn't. In the ones that succeed there is often some kind of economical interest pushing the development in a strategic direction.





> I'm pretty sure Apple's success is because they work a lot with
> usability.

I may be quoting you out of context here, but Apple is the best example
I know of a completely closed source company that extracts premium prices.

Sure, they have great design and great usability. But they are not open
in any sense of the word.

I'm sorry if I was unclear. I wasn't trying to say that Apple is open, I was trying to say that I think the Apple success is primarily based on other things than it's closed development philosophy.



I'm not sure that there will be any light shed on this thread if it
continues, as it seems to be mostly smoke. Starting with the subject
line of the thread. It assumes that a rant is justified and that a
rescue is needed and the need for this rescue is seen by all.

I suppose you are correct.



p.s. I absolutely hate the renaming of the server. Hated it the first
time, do now, and probably always will. Branding, bletch.

I'm just tired of changing the names everywhere in the code, wiki and in other places.

I kind of like the last one, I'm just hoping it won't change again next year.

Goodsounds
2009-08-25, 11:21
It's all about getting and spreading passion. Whether it's people like Chris or the thoughtful postings from people like Erland and others. Comments are offered because people care.

Chris, you'll know you're in the middle of the spectrum when you're getting the same number of complaints from each side.

Some old-timers seem to suffer from the residual effect of drinking too much Kool Aid. The software is open source because of a dollars and cents decision made by the founders (in their own self interest), not because of a philosophical point of view. Logitech needs to play the ball where it lies, it's nothing more than that. It's not a democracy, it's not like other open source projects that are unrelated to a commercial product. So-called "standards" might be a better example, where typically the big boys get involved to railroad the outcome in a manner consist with the direction they want to take with their own products.

Chris, we appreciate and support all you guys and (most of us) appreciate that you do listen.

Letten
2009-08-25, 14:21
My first squeezebox was the Duet and it was a lot of trouble the first year or so. I really hated it, if you only knew how much time I have spent trying to sort it out (I really should have sent it back).

But I hoped the problems would be solved eventually and they were (7.3.3). But I must admit that I have been very cautious about recommending it to friends because of my initial experience.

Today I also own two Booms and a Classic and they work fine. I still experience a few glitches (synchronisation) but I'm sure they will be solved as well.

I'm very happy about the recent focus on Q&A instead of new functionality. But it's a two edged sword, new music services keep emerging and squeezeboxServer needs to keep up (e.g. spotify).

The decision to move to SQLite is also fine by me as it allows me move SC to a cheap NAS.

Sike
2009-08-26, 01:03
I have 8 Classics, 3 Booms, 2 SB2s, 1 SB-Reciever and 2 SB-Controllers running on my network. I still have a Slimp3 tucket away, but only use it to impress the ladies :)

When I bought my Duet, I never attached the Reciever to the network. I never tried. The Reciever attached now is at a friends house, running on my network. He bought it a couple of months ago and it works fine. It does have the odd problem with wireless, it seems like the antennas are not that good.

I have never been impresses with the SB-Controller. It seems to be Slow and flimsy. iPeng has been the reak killer-app for me. It runs perfectly even over a slow connection. Anybody who asks me if the should buy the Duet, I tell them to buy a classic and a iPod Touch with iPeng, it works a treat!

I have been running Squeezeserver for 6 (or even maybe 7..) years now. I have been fed up at some times due to it crashing or not being able to display foreign characters. Nowerdays I just update it when I need to, or a new feature comes out. I have a large collection of older versions I could roll back to, but this has not been necacary for a long time.

Maybe my listening habbits are different than others. I use the standard install with musicIP (<-- also a killer app) and Trackstat, that's it.

I had the occasional crash with the last version, but have learnt that trashing the DB and rebuilding it solves most problems.

DrewSB
2009-09-16, 05:15
I am relatively new to Squeezeboxes after getting frustrated trying to get my Tivo to serve as a reliable music player. I now have a Squeezebox 2 and a Boom, and have been very happy with both.

There are a few things about them that I think should be resolved for a mass market product:

1. The need for a seperate server, or at least a seperate propriatary server. The box should include a USB port or internal storage and be able to use it. It should also be able to use external UPNP sources. Perhaps indexing them in an internal database in the Squeezebox, ala embedded shivaplug. The ideal solution would be an embedded Squeezecenter, I really like the customization that is possible with this software.

2. The seperation between SqueezeCenter and SqueezeNetwork. As a new user, I was surprised that there was such a wall between these sources - to the extent that the firmware will up/down grade between uses. I would have expected that internet content would be served by Squeezecenter if available, otherwise Squeezenetwork. Local music would be served by Squeezecenter if available. I would be able to turn on the Squeezebox, listen to Rhapsody, online radio, etc, and only need to turn on Squeezecenter in order to make the local media menu items appear. I am running SC on a NAS so I leave it on all the time, but for the typical user running it on thier PC, it forces them to decide which server to connect to before they can enjoy thier music.

peterw
2009-09-16, 09:39
There are a few things about them that I think should be resolved for a mass market product: ...

Except for UPNP client support, you've just described Squeezebox Touch.

Take away the USB port & embedded SqueezeCenter, and you've got Squeezebox Radio -- shifts between SC and SN as needed, and everything that doesn't require SC works just fine so long as Radio has a wifi or ethernet connection and can talk to SN. I haven't tried it away from wifi yet, but I'm sure the plan is that you could, as with Boom, play music from the analog line in if there's no network. You won't even need AC power all the time if you buy the battery.

bolfings
2009-09-16, 15:30
Interesting thread - The rationale behind alot of the posts is understandable, but the questions are far simpler (to me).

First, I love my squeezeboxes - they have been a wonderful part of my life for many years now. But as a "Joe Sixpack", I want to know how my Boom, SB3, and classic are going to work with the greater integration of entertainment sources. I bought a Mac mini to run my SC, because it was inexpensive, small footprint, and offered integration capabilities with my DVD collection and internet entertainment sources. It works well and easily, and thats all I want. After using a windows PC, and then a Linux appliance, both of which had unique and overly challenging issues, I am very happy with my choice of a mini with an external drive.

But unless Logitech wants to make a portable player to take the place of my wifes and kids iPods, or a cell phone that integrates with my music library, I am getting more and more frustrated with the arrangement. iTunes has a stranglehold grip on my friends and families music enjoyment, and it looks like thats only going to grow. I am trying to maintain separate libraries for my flac files, and the families MP3s. I'm almost ready to give up on flac and put everything into MP3 just for ease. iTunes rips and tags, organizes and finds, and makes the process pretty painless, and they all understand it - many parts I don't. SC does a great job of organizing and serving, and I love the sound quality of flac, but I still have to find separate apps to rip, tag, and manage my files. If I find a tag problem when I am using a SB, there is no easy way to ID the file and fix it, so I generally live with it.

I understand how alot of people enjoy the openness, and the ability to customize and modify the system, but when the basic system gets too complicated for the most basic user to very simply enjoy it, while the technically proficient agonize over issues with specialized customizations that appeal to a limited few, I'm just afraid that its lost its appropriate focus on the ability of us "Joe Sixpacks" to solidly recommend it to our friends.

Maybe there needs to be a 2 track SC development - an open source release that can be available to those who choose to pursue it, modify it, customize it, and deal with the problems that come with it - and a more closed source option that runs on Windows and OSX, with few or no options but with additional functionality, like integrated ripping and tagging, and iTunes integration that is simple and painless enough that families can use it easly. Sorry to harp on it, but everyone in my family, and most of my friends use iPods or iPhones, and I'm the only one using a squeezebox. I think its well over 20 to 1.

Anyway - thats my 2c -

radish
2009-09-16, 15:45
Maybe there needs to be a 2 track SC development - an open source release that can be available to those who choose to pursue it, modify it, customize it, and deal with the problems that come with it - and a more closed source option that runs on Windows and OSX, with few or no options but with additional functionality, like integrated ripping and tagging, and iTunes integration that is simple and painless enough that families can use it easly.
Why does the license (i.e. open source or not) affect what functionality it can have? Why can't the open source version have all the bells & whistles?

Philip Meyer
2009-09-16, 16:56
It seems to me that the developers haven't been communicating as much recently. I think there's been a lot of discussions between beta testers about new features or changes to existing features, with very little feedback from logitech developers.

In the last few months it feels like the developers have had less authority to influence decisions on useability and functions. It seems to me that there have perhaps been some strange decisions that not all developers have agreed with, but the powers that be have layed down the rules; the developers are implementing those decisions.

I can understand why - they have a day job to do, the priority being to hit milestone dates so new products can be launched. The bigger the product, user base, etc, the more reactive and less proactive development becomes.

However, I feel that it has been harder for effective use of beta testers. I'm finding that it is less useful to raise issues in the forums, but to go straight to bug/enhancement reports, as eventually someone with decision-making authority may catch up with them, whereas they will never go back to catch up with old forum posts.

There do appear to be conflicting issues. There's been a purge of some functionality to save memory footprint and/or perhaps to make the product more simplistic. On the other hand, there have been more functions added that have fringe benefits, such as work to produce different styles of clock screensavers, non-music related add ons (Facebook app, Flickr).

Some new functionality that should be quite simple seems overly complicated. I still find Extension Downloader awkward for example. i.e. needing to manually configure third-party repositories to find extra plugins, and then the plugin screen becomes rather crowded with tables.

It seems like some long-running bugs/enhancements that frequently get discussed have been pushed out to the future again, and time is being spent on other functions to give marketting something to use.

I think a lot of plugins are going to need updating for 7.4/8.0, and there are many opportunities for new plugins for the new products, but with big changes happening frequently to the 7.4 code and less chance of developer support at the moment, there's not much incentive to try to develop/support third-party plugins.

DrewSB
2009-09-17, 04:53
Except for UPNP client support, you've just described Squeezebox Touch.


I hadn't looked at the new products until you mentioned it. It looks like a very nice device.

MrSinatra
2009-09-18, 07:23
Logitech employees,

my comments come from someone who is glad you acquired slim devices. while i am sad to see sean and dean go, and see no gain by their leaving, i think in the long run logitech will create a group of products that will get cheaper and better due to logitechs mass market ability and focus.

having said that, i do think you need to reconsider the current paradigm.

i should preface myself by saying i'm NOT a developer, but i do represent the 'average joe' b/c while i have learned a lot since i bought a SB2 5 years ago or so, i share my music with many people and so i see first hand how they react and its much like how i reacted (which i remember well anyway).

i support the idea logitech has to be targeted to be useful FIRST to the average joe, and second to the advanced user.

i support the move to SQLite and to being less resource intensive overall.

i support the idea to have SC AVOID issues and common problems it will come across in the marketplace. ie. it should be the GOAL of SC to adapt and change where it makes sense to, rather than proselytize and martyr itself on philosophical points. in other words, "go with the flow" more or less, recognize de facto standards and marketplace conditions.

now, what should logitech do to accomplish these goals?

i have two suggestions, and i think they would help logitech tremendously:

1. setup SC so that it "reports back" to you all at the mothership everything about a specifc users given installation. ie. what options are being used and how are they set, (were default values changed? to what? etc...) what plugins, what formats and how many files, what version of SC, log files and errors and dropped streams, basically ANYTHING and EVERYTHING you can think of. (maybe even a "last.fm" or winamp style "orgler" for song listings to be part of an online community presence, but thats a tangental and separate idea)

you should not only use this info to aid development, but you should present it in aggregate to all of us, (which wouldn't violate any given user's privacy) maybe even in a page in SC itself! it would go a long way in showing "what the majority do" and help 3rd party development.

i would suggest it be "opt out" but i am sure the majority here would say it should be opt in. however, it would be good to have the opt out folks at least reported by SC as "opt outers" so we know how many of that type there are, altho there should be a way to totally disable it for those installs that AREN'T internet capable.

2. this next idea is even bigger, and is the paradigm shift i alluded to.

i think logitech needs to sell a SC piece of HARDWARE. it makes no sense to me that so much development costs go into SC only to give it away for free, on a variety of platforms and permutations that only make supporting the SC clients MORE difficult than it needs to be.

what exactly am i suggesting? a small silent energy efficient box, running linux, with SC on it. all that would still be open source, but it could be streamlined, maybe even rewritten in something OTHER than Perl.

one version could do SC only, while another could do SC and a client (like SBR or SB2) combined; the obvious benefit being no router, internet or even computer is necessary to use that one. both could use a SSD and could boot nearly instantly. it could be user upgradeable, like for ram or wireless card or processor or SSD. the linux and SC could be in firmware, or if installed to the SSD, the FW could install fresh linux/SC to a new SSD via web install from a logitech repo.

it would be reachable via web interface (or SBC) and connect to your router via dhcp (default) or static ip. it would have usb ports and other card ports for playing music attached to it which is automatically scanned. since it would operate on top of a full linux OS, it could be configured to map network drives to get music from elsewhere on your network. the linux OS itself could have a webui app for basic config options (like mapping drives) for non-techy types, maybe via SC itself.

the SC device would be constantly connected to logitech, (unless users opt out), and logitech could update it, troubleshoot it, monitor it, all remotely. it could be controlled or connected to SN, and a community social networking type of site.

what i am talking about is a turnkey solution. u could add a CD ripper to it, but that should be optional. it could do UPNP and DLNA and serve out not just music, but video to any (logitch) media adapters. maybe even an ipod hookup, (and i hate ipods, but u get my drift, any such device)

once SC8.0 is out logitech should concentrate on this device, and should devote the majority of resources to it, rather than to SC development for various operating systems. in other words, "new" SC releases appear on it first, (including betas) in its new language and streamlined development environment, and later ported over to other OS's in "major release" versions only. surely this would aid in development of SC, since it would still be open but in a more "definite" environment in a less clunky language.

the benefits of all this are that people who don't like or even use computers could use this device. even people who do like computers could use this without needing a computer turned on. it would be a LOT easier for logitech to support and it would create another revenue stream.

if you can sell a transporter, for $1000 or whatever, which is insane and ridiculous, then you could sell this, probably for less, and reach a lot more people.

goodbye problems with peoples computers and networks, you at logitech could "get in" the device and figure it out remotely, if there were even any issues, which btw there wouldn't be on the "combo server/client" unit.

(you could make it work with RF remotes too, like the harmony and/or add RF to the SBC to make networking totally unnecessary)

one more thing, it would be PORTABLE. thats no small thing. i could see it in cars, used at friends, etc...

i look at my SB2 and wonder why i need to have a computer turned on to use it with local music. it simply no longer makes sense to do that.

so this was a long post, but i really think these two ideas would help logitech out tremendously. what say you logitech?

MrSinatra
2009-09-18, 07:30
d'oh... one other thing i forgot to mention...

with such a device, SN could always make your local music available to you wherever you get access to SN. so if you're away from home, just go to SN to get to your tunes. additionally, the device itself could be reached via the proposed social networking site i mentioned, or DDNS or something like that. (get yor music on your cell phone pluged into your car stereo, etc) bitrate lmiting could be enabled dynamically.

make your music available to multiple households or friends clients, etc...

erland
2009-09-18, 09:48
I still find Extension Downloader awkward for example. i.e. needing to manually configure third-party repositories to find extra plugins, and then the plugin screen becomes rather crowded with tables.

Hasn't this been solved in 7.4 ?
If you check the checkbox to indicate that you allow third party plugins you should get a single list with all third party plugins that isn't officially recommended by Logitech.

I think the only repositories that aren't included is the "beta testing" repostiories which some developers have.

Philip Meyer
2009-09-18, 10:35
>Hasn't this been solved in 7.4 ?
No.

>If you check the checkbox to indicate that you allow third party
>plugins you should get a single list with all third party plugins that
>isn't officially recommended by Logitech.
>
That only gives you some - the ones that have been suggested by the community as safe third-party plugins.

Most poeple have to add several repositories - one for each third-party developer, and each one is listed as a separate table.

I just tried clicking on Settings > Plugins. After at least 30 seconds waiting for the page to display, I counted 24 different tables - one for Active, Inactive, Recommended 3rd party plugins, other 3rd party plugins, Alien BBC plugin, Erland's plugins, my plugins, etc...


>I think the only repositories that aren't included is the "beta testing" repostiories which some developers have.
There's no such thing (except I think maybe Erland has his own repository and a test repository for WIP stuff).

Third-party developers have their own repository for their own plugins - they could be for 7.2, 7.3, 7.4, future, etc. i.e. any version of any of their plugins.

ChrisOwens
2009-09-18, 10:36
Did you ever notice that Logitech employees "clam up" when a thread starts getting close to actual product plans? :) Although, it's not looking quite like you envision, Mr.Sinatra.

We really, really, can't talk about it, but we are definitely keeping these issues in mind as we move forward. See, now you made me talk like a marketing guy. Dammit.

MrSinatra
2009-09-18, 10:47
Chris,

thats great. but how about suggestion number 1? basically the idea SC should have an equal to "Windows Customer Experience Improvement Program" where the app collects usage data and sends it to logitech, etc?

ChrisOwens
2009-09-18, 10:49
I'll get someone more involved with product definition and strategy to have a look at the thread for that one. :)

erland
2009-09-18, 10:59
>Hasn't this been solved in 7.4 ?
No.

>If you check the checkbox to indicate that you allow third party
>plugins you should get a single list with all third party plugins that
>isn't officially recommended by Logitech.
>
That only gives you some - the ones that have been suggested by the community as safe third-party plugins.

Most poeple have to add several repositories - one for each third-party developer, and each one is listed as a separate table.

I just tried clicking on Settings > Plugins. After at least 30 seconds waiting for the page to display, I counted 24 different tables - one for Active, Inactive, Recommended 3rd party plugins, other 3rd party plugins, Alien BBC plugin, Erland's plugins, my plugins, etc...

If you have checked the "Show all 3rd party plugins" checkbox in the Plugins tab you shouldn't need to specify any of the repositories listed on the wiki. This checkbox will include these repositories automatically and put them in a "Other 3rd party plugins" list.

However, I'm guessing that in your setup you probably had these repositories already registered which probably explains why you see multiple lists. Try removing them and checking the "Show all 3rd party plugins" checkbox and you should get a nice surprise.

If you don't check the "Show all 3rd party plugins" checkbox, there will only be a single list with plugins recommended by community and reviewed and approved by Logitech.

Isn't this how it works in your setup ?

iPhone
2009-09-18, 11:51
Nice post with some interesting ideas. I was right there with you until......


i look at my SB2 and wonder why i need to have a computer turned on to use it with local music. it simply no longer makes sense to do that.

so this was a long post, but i really think these two ideas would help logitech out tremendously. what say you logitech?

The Radio and Touch are "Smart" devices (the Controller with minimum resources). All Squeezeboxes prior are "Dumb" devices and that is why they need a Server whether it is SC locally or phone home to SN. For what is to wonder why about the SB2? It is impossible for an SB2 to function on its own to play local music because its a "dumb" device that currently needs either a Server or a "Smart" device running TinySC to be able to play local files. And even if something was created on SN to play local files, I would think a local server would have to be on so its back to SC and "Dumb" devices.

I see the point, but to me, we are just now getting there with this new generation of Smart devices.

MrSinatra
2009-09-18, 12:04
Nice post with some interesting ideas. I was right there with you until......

<my quote>

The Radio and Touch are "Smart" devices (the Controller with minimum resources). All Squeezeboxes prior are "Dumb" devices and that is why they need a Server whether it is SC locally or phone home to SN. For what is to wonder why about the SB2? It is impossible for an SB2 to function on its own to play local music because its a "dumb" device that currently needs either a Server or a "Smart" device running TinySC to be able to play local files. And even if something was created on SN to play local files, I would think a local server would have to be on so its back to SC and "Dumb" devices.

I see the point, but to me, we are just now getting there with this new generation of Smart devices.

i think either i wrote that poorly, or you misunderstood, probably both. ;)

i wasn't saying the SB2 i have, as in the one i literally look at and use, doesn't need SC or SN. sure, of course it does; and it mades sense 5 years ago to host the server on a users computer, yada yada yada...

what i meant was that today i look at my SB2, and i believe that what made that a sensible product/paradigm years ago no longer exists today.

in other words, a SB2 today should work with my "proposed device" INSTEAD OF my computer.

you could hook up a USB drive or esata or whatever to the small linux SC box, and power your SB2 or whatever with just the proposed SC device. thats what i meant when i said "i wonder why i need a computer on" meaning my own computer.

times have changed, the paradigm should change too. its financially feasible now.

DrNic
2009-09-18, 13:17
I'm not sure I'm in on the "Mr Sinatra ethos". Though it would seem that Logi are already thinking of removing the unknown variable to make their user experience a predictable one.
I have always lagged behind on the server front. Why? - because until I HAVE to change (because a new device now requires it) the current version does what I wanted it to do now as well as I wanted it to do then - play my music.
I have a low energy fanless server running linux already - why do I want to(or eventually have to) buy another device to sit on my network to do the same job. My server does more than the logi device will be able to do (video, ftp, web, email, automation).
I don't consider my home network environment to be anything special, and I have managed to install updates to the server (when needed) without ever having issues. Does that make me especially lucky?
I worry we are just heading for a hooked in system that demands you use the Logi server box. The original appeal was that I could use what I liked, and not be dictated to about the serving hardware.
I realise this POV has many holes for others to pick at.

Nic

Philip Meyer
2009-09-18, 13:52
Well, I didn't really agree with much of it, I'm afraid.

My PC is left on most of the time, or has wake-on-LAN, so it makes sense to use it to host music, rather than need to transfer music to another piece of kit (NAS, USB Disk, etc). It's nice that Players can connect direct to internet to stream radio, I expect customers to identify with that. But I expect that most customers who have music ripped to local disk have it connected to a PC. eg. hoards of people who use iTunes to sync up to iPods.

I would expect that having software that reports back to Logitech with usage stats would scare a lot of people. I always disable things like that in WMP, iTunes, etc. I'd imagine at least half of the community base would not opt in (I think it's a legal requirement to opt in to things like that in the UK, rather than give the option to disable). So, any stats would be skewed by the type of people that opt in/opt out.

Philip Meyer
2009-09-18, 14:07
>If you have checked the "Show all 3rd party plugins" checkbox in the
>Plugins tab you shouldn't need to specify any of the repositories listed
>on the wiki. This checkbox will include these repositories automatically
>and put them in a "Other 3rd party plugins" list.
>
I have the "Show all 3rd party plugins" checkbox ticked. No idea which ones that gives me, what table title they are listed under, etc, as there's not been much announcement or documentation explaining it.

I assumed that setting only controlled the recommended 3rd party plugins, and any others would need to come from manually entered repositories. I add 3rd party repositories to the list when I see plugin announcements, so I can see what plugins are available. I don't really see why they should come up as separate tables.

In Recommended 3rd Party plugins, I have 2 plugins listed (assume there are more, but I have them installed, so they appear within "Active").
In "Other 3rd party plugins", I have 6 listed.

Adding all the other repository plugins together gives another 40 that are available for installation. If I wanted to find one in particular, I'd have to know who the author was/what repo in order to know what table to look in to find it.

erland
2009-09-18, 18:04
I have the "Show all 3rd party plugins" checkbox ticked. No idea which ones that gives me, what table title they are listed under, etc, as there's not been much announcement or documentation explaining it.

I assumed that setting only controlled the recommended 3rd party plugins, and any others would need to come from manually entered repositories. I add 3rd party repositories to the list when I see plugin announcements, so I can see what plugins are available. I don't really see why they should come up as separate tables.

Recommended plugin list is there by default.
Other third party plugin list are only there if you have the "Show all 3rd party plugins" check box ticked.

You can remove all manually added repositories which is available on the wiki page that lists repositories. The checkbox described above will add all plugins in these repositories automatically and they will appear in the "Other third party plugins" list.

If you have manually registered repository urls, these will override the "Show all 3rd party plugins" checkbox and you will get multiple lists.



In Recommended 3rd Party plugins, I have 2 plugins listed (assume there are more, but I have them installed, so they appear within "Active").
In "Other 3rd party plugins", I have 6 listed.

Adding all the other repository plugins together gives another 40 that are available for installation. If I wanted to find one in particular, I'd have to know who the author was/what repo in order to know what table to look in to find it.

If you remove all your manually registered repository url, you will get a lot larger "Other 3rd party plugins" list. All plugins now listed in separate lists will then be moved into "Other 3rd party plugins".

The ONLY repositories you should need to register manually are those not listed on the wiki page. In 7.3 this wasn't the case, then you had to register all repositories to get not recommended plugins, but the new checkbox in 7.4 takes care of this.

MeSue
2009-09-18, 18:41
If you have checked the "Show all 3rd party plugins" checkbox in the Plugins tab you shouldn't need to specify any of the repositories listed on the wiki. This checkbox will include these repositories automatically and put them in a "Other 3rd party plugins" list.

However, I'm guessing that in your setup you probably had these repositories already registered which probably explains why you see multiple lists. Try removing them and checking the "Show all 3rd party plugins" checkbox and you should get a nice surprise.

If you don't check the "Show all 3rd party plugins" checkbox, there will only be a single list with plugins recommended by community and reviewed and approved by Logitech.



Geez, I didn't know that! Thanks!

Philip Meyer
2009-09-19, 00:04
I didn't know that it worked like that. Some better help information should be made to the Plugins page. Currently it only says:

"You may add additional third-party extension repositories by entering the URL for the repository below. A list of repository URLs can be found here."
where here is "http://wiki.slimdevices.com/index.php/SqueezeCenter_Repositories".

i.e. it is suggesting that you have to add those manually.

I tried it out - deleted a few repositories I had manually entered, and more appeared under Other 3rd Party Plugins. Great. So I deleted some more repositories that I didn't need any more, and now I don't have an "Other 3rd Party Plugins" table at all :-(

Also, does it include your test repository: http://erlandplugins.googlecode.com/svn/repository/trunk/testing.xml

erland
2009-09-19, 00:13
I tried it out - deleted a few repositories I had manually entered, and more appeared under Other 3rd Party Plugins. Great. So I deleted some more repositories that I didn't need any more, and now I don't have an "Other 3rd Party Plugins" table at all :-(

Sounds like a bug



Also, does it include your test repository: http://erlandplugins.googlecode.com/svn/repository/trunk/testing.xml

No, you will have to add that manually.

Philip Meyer
2009-09-19, 00:33
>> I tried it out - deleted a few repositories I had manually entered, and
>> more appeared under Other 3rd Party Plugins. Great. So I deleted some
>> more repositories that I didn't need any more, and now I don't have an
>> "Other 3rd Party Plugins" table at all :-(
>>
>Sounds like a bug
>
I removed a couple more, applied, navigated away and went back to Plugin page and the table then showed up again.

>> Also, does it include your test repository:
>> http://erlandplugins.googlecode.com/svn/repository/trunk/testing.xml
>>
>No, you will have to add that manually.
>
I thought that would be the case, and probably better to appear as its own table anyway.

Thanks,
Phil

toby10
2009-09-19, 04:38
I *love* the Extension Downloader (ED) feature!! Particularly since being a Radio & Touch beta tester where updating SC is a much more frequent necessity and updating my chosen plugins manually and (almost) daily would have been a hair pulling experience (and I have little to spare).

But, I do see two areas where the ED could be improved.

1. Maybe add a "New Plugins" section where SC shows newly added Plugins to ED so I don't have to re-read the entire Plugin offerings?

2. Maybe make (or optionally allow) for ED to exclude Plugins that are player specific and do not apply to my player(s) that I use? Ex: If I only own a SB Radio, why confuse me by showing Plugins that are only related to Boom, Duet, TP?

Dunno, just thinking out loud and wondering if so many options (Plugins) may be a bit overwhelming to a first time user, particularly if they do not apply to my sole SB player.

erland
2009-09-19, 06:59
But, I do see two areas where the ED could be improved.

Could you please make sure to post these ideas in the beta section or register them as enhancement requests, they are too good to get lost in this thread in the General section.

toby10
2009-09-20, 05:57
Oki doki erland, will do.
And what of my thinning hair??? :)

MrSinatra
2009-09-22, 01:37
I'm not sure I'm in on the "Mr Sinatra ethos". Though it would seem that Logi are already thinking of removing the unknown variable to make their user experience a predictable one.

no need to be in on it, whatever logitch does will be beyond our control, or anyone else here.

but clearly, there is a need for such a product, and i'm not surprised they have something in mind. the important point to remember from your POV is it wouldn't keep you from still doing things your way.


I have always lagged behind on the server front. Why? - because until I HAVE to change (because a new device now requires it) the current version does what I wanted it to do now as well as I wanted it to do then - play my music.

then nothing i suggest should worry you.


I have a low energy fanless server running linux already - why do I want to(or eventually have to) buy another device to sit on my network to do the same job. My server does more than the logi device will be able to do (video, ftp, web, email, automation).

who suggested you would have to get it? not me.

but who is to say the logitech box couldn't do everything your does?


I don't consider my home network environment to be anything special, and I have managed to install updates to the server (when needed) without ever having issues. Does that make me especially lucky?

yes.


I worry we are just heading for a hooked in system that demands you use the Logi server box. The original appeal was that I could use what I liked, and not be dictated to about the serving hardware.
I realise this POV has many holes for others to pick at.

Nic

that isn't what i suggested, but i wouldn't mind it if it meant a better user experience.

MrSinatra
2009-09-22, 02:14
Well, I didn't really agree with much of it, I'm afraid.

nothing to fear, but fear itself.


My PC is left on most of the time, or has wake-on-LAN, so it makes sense to use it to host music, rather than need to transfer music to another piece of kit (NAS, USB Disk, etc).

so don't buy it, problem solved.


It's nice that Players can connect direct to internet to stream radio, I expect customers to identify with that. But I expect that most customers who have music ripped to local disk have it connected to a PC. eg. hoards of people who use iTunes to sync up to iPods.

i expect that as well, since there wouldn't be any other way to listen to it on the SB. but wouldn't it be nice for logitech to have a device which makes itunes unnecessary? that could handle their ipod? even play off it?

how about a system that makes a [traditional] computer or the internet unnecessary?

how about a system thats reasonably portable? car worthy?


I would expect that having software that reports back to Logitech with usage stats would scare a lot of people.

doubtful. most users don't care, and don't even know, and why should they?


I always disable things like that in WMP, iTunes, etc.

and i proposed you should have that ability, we totally agree on that.


I'd imagine at least half of the community base would not opt in (I think it's a legal requirement to opt in to things like that in the UK, rather than give the option to disable).

ugh, typical gov't heavyhandedness. so proposed SC devices sold in the UK could be configured like that if needed. the software however is free as i am so often reminded and voluntary to DL, so i doubt that law would apply to that.

regardless of laws in any given province, i think most people would not care if it was "opt out" b/c most people wouldn't know, and why should they? what could/would logitech do thats so dasteredly? if by "community base" you mean the forum, maybe so, but so what?

i think anyone who wants to help SC would. i think anyone who uses SN would. i think anyone on Last.fm would. i see no reason (in most cases) for anyone to object, outside of paranoia.


So, any stats would be skewed by the type of people that opt in/opt out.

so what? that doesn't stop microsoft from doing it, and they spend literally hundreds of millions on this kind of data mining and R&D.

some stats would be better than no stats. my suggestion that active installs that opt out at last have that one aspect reported to logitech would then give you an idea about the ratio, and then those without net connectivity can basically be forgotten or possibly guessed at to some degree by considering the amount of downloads, etc...

i think you guys are missing a big part of my suggestion by focusing on these things anyway...

it would be a huge boon to logitech and users alike if logitech could not only get lots of automated feedback, but be able to ACTIVELY go into your device and troubleshoot your issues, should you encounter any. in many ways, its not much different than a cable company managing your cable box.

i'm all for it being optional, and frankly the opt-in/opt-out debate is a quibble and not the point. the point is SC could be better, faster, and easier to own.

Ozz
2009-10-06, 18:22
Being everything to eveyone is not possible! Trying to do so will be the end of the products. trying to be a solid reliable product is possible IF there is a defined goal. I for one like the fact that SB is open source. I also understand that at times this will lead to issues that make some customers unhappy. A price we pay for the ability to have a voice through these forums that leads to new ways to manage and listen to our music. I may be Joe six-pak but I love my Transporter and I have little to complain about (outside of the lack of 12vdc trigger... sorry had to have my dig) This thing rocks!!! I listen to it every day and my friends are amazed by it. Is it perfect? NO! is it a work in progress? God I hope it always is beause it is impossible to be perfect for more than a second in time. We as a community will always ask for more. I hope that this product continues to mature and evolve as we find more and better ways to take advantage of it.

maggior
2009-10-07, 06:51
I may be Joe six-pak but I love my Transporter and I have little to complain about (outside of the lack of 12vdc trigger... sorry had to have my dig) This thing rocks!!!

As the owner of a Transporter, you are overqualified to be "joe-sixpack" by definition :-).

I agree with what you say - trying to be everything to everybody is a slippery slope that you are destined to slide down.

Yes, they've been adding features and functionality, but notice that they haven't even hinted about venturing into video. If they did that, then they would really be trying to be everything to everybody. That I think would be the beginning of the end.

This brings to mind another perl of wisdom - "it can always be worse".

Nonreality
2009-10-07, 17:47
You know I really like Squeezecenter 7.4 or Squeezebox server what ever they are calling it now. But if they are really making it more user friendly for the casual users I would think that a manual showing all the settings and how to make use of them would be a must. Maybe I've been missing that but it always seems that you have to search and search to get info on settings and that should be a much simpler process.

MrSinatra
2009-10-07, 18:49
very true. i'd like to see logitech write a series of wiki articles that say given this data set, given these settings, here is the expected result. the lack of such info causes a lot of problems.