PDA

View Full Version : Is Slimserver sustainable as Open Source?



amcluesent
2007-02-19, 12:55
Not wanting to start a flame war here, but seriously is the current open source approach to developing Slimserver viable?

I appreciate that the community provides plug-ins etc. but so does the community for PhotoShop, it just needs a documented API.

If 'mainstream' consumers are dropping a few $100 on the Squeezebox and Transporter, the slimserver with the bugs of 6.5.1 and issues with Vista seems to devalue the experience. (As a recent buyer of a squeezbox, I decided to go with 6.3.1)

Is it time for Logitech to fork the build, so that the community can take the GPL version for U*nix and there is a closed source Win32 binary for the man-in-the-street?

mherger
2007-02-19, 13:09
> Is it time for Logitech to fork the build, so that the community can
> take the GPL version for U*nix and there is a closed source Win32
> binary for the man-in-the-street?

What advantage would you hope to see by closing the source?

--

Michael

-----------------------------------------------------------------
http://www.herger.net/SlimCD - your SlimServer on a CD
http://www.herger.net/slim - AlbumReview, Biography, MusicInfoSCR

Mark Miksis
2007-02-19, 13:27
I appreciate that the community provides plug-ins etc. but so does the community for PhotoShop, it just needs a documented API.

Non-Slim employees contribute to the main codebase as well as external plugins.


If 'mainstream' consumers are dropping a few $100 on the Squeezebox and Transporter, the slimserver with the bugs of 6.5.1 and issues with Vista seems to devalue the experience. (As a recent buyer of a squeezbox, I decided to go with 6.3.1)

Is it time for Logitech to fork the build, so that the community can take the GPL version for U*nix and there is a closed source Win32 binary for the man-in-the-street?

Why would slimserver be *more* stable with fewer people looking at it and working on it?

pfarrell
2007-02-19, 13:45
amcluesent wrote:
> Not wanting to start a flame war here, but seriously is the current open
> source approach to developing Slimserver viable?

Do you have any idea what open source software means?
Do you know that at least 50% of the servers on the Internet, worldwide,
run Apache, which is open source?
Do you know what LAMP means and why most of the new websites are using
it? Hint, the letters stand for Linux, Apache, MySql and Perl/Php.

Do you have a real question? or are you just throwing out gasoline
hoping for a flame war?

--
Pat
http://www.pfarrell.com/music/slimserver/slimsoftware.html

David Alexander
2007-02-19, 13:57
On Mon, 19 Feb 2007 21:09:50 +0100, "Michael Herger" <slim (AT) herger (DOT) net>
wrote:

>> Is it time for Logitech to fork the build, so that the community can
>> take the GPL version for U*nix and there is a closed source Win32
>> binary for the man-in-the-street?
>
>What advantage would you hope to see by closing the source?

I think what he was getting at was that the Slim experience is as much
influenced by the software as it is the hardware, and by not
controlling the software Logitech may be risking developing a negative
product image when the community-developed-software is perceived to be
as buggy and difficult as some of the recent releases have been
perceived as being.

This didn't seem to hurt Slim Devices so much, but then they were more
of a grass roots type of company with many loyal early adopters. Two
factors that could change that are, 1) the hardware is moving further
to the mainstream market where people are less likely to be willing to
tinker to get things working, and 2) the software becomes more complex
with each release and therefore the possibility of perceived quality
control issues will be greater in the future than it has in the past.


Note, I'm not making any judgements here, just fleshing out the
argument.

mherger
2007-02-19, 14:04
> by not
> controlling the software Logitech may be risking developing a negative
> product image

Development is pretty well controlled by Logitech. There aren't many
outside developers with write access to the trunk.

--

Michael

-----------------------------------------------------------------
http://www.herger.net/SlimCD - your SlimServer on a CD
http://www.herger.net/slim - AlbumReview, Biography, MusicInfoSCR

kdf
2007-02-19, 14:39
Quoting David Alexander <listmail (AT) cox (DOT) net>:

> as buggy and difficult as some of the recent releases have been
> perceived as being.

as always, by a disproportionally vocal minority, pushing hard on
advanced features, rather than the "best practises for slimserver"
methods while claiming all they want is to listen to music. In my
time being involved with slimserver, I've not seen a single release
that has been greeted with "this is great" for longer than a half a
dozen posts before the complaints begin. Now, as 6.5.1 is around,
6.3.1 is the safehouse, yet when that was new the complaints were all
the same. Some will still claim 5.4.1 is the only stable version
around. Great that everyone has the freedom to choose.

This is no different for closed software. WinXP when new, horrible
and buggy while Win2k was nearing maturity and working fairly well.
I think the real difference is the expectation. Closed source, you
know there is nothing you can do, so complaining really doesn't do
anything. Open source, things to happen. Some will always believe
that complaining loudly is the most effective.

Long term viability: if Slim Devices or Logitech ever suddenly vanish
from the planet, SlimServer is already public. Anyone, anywhere can
add new features until the last piece of hardware goes dark.

-kdf

Kevin O. Lepard
2007-02-19, 14:44
>Long term viability: if Slim Devices or Logitech ever suddenly vanish
>from the planet, SlimServer is already public. Anyone, anywhere can
>add new features until the last piece of hardware goes dark.

Amen. That's why I picked Slimserver over closed solutions.

Kevin
--
Kevin O. Lepard

Happiness is being 100% Microsoft free.

325xi
2007-02-19, 15:04
Open source doesn't have to be of low quality! My experience with ever-changing Eclipse proves that to me very well. It's all a matter of good organization, good project management (read - testing control), and final quality control (testing, again).

JJZolx
2007-02-19, 16:24
Closed or open source, doesn't matter a wit to me. If making it closed source would somehow hasten SlimServer's development into something more stable and usable, I'd be all for it. SlimServer still feels pretty primitive to me and its progress into something more advanced is progressing at a snails pace. I'm not exatly sure what to attribute that to. One, there are very few developers on the project, despite the open source approach. Two, they seem to spend a great deal of their time fixing bugs that keep reappearing. Also, the cross-platform compatibility issues also appear to be eating them up lately.

pfarrell
2007-02-19, 16:32
JJZolx wrote:
> One, there are very few developers on the project, despite
> the open source approach.


Open or closed has nothing to do with speed of development. And adding
more developers usually slows development efforts. Microsoft claimed to
have 5,000 developers on Vista, it came out late, buggy and missing key
features. Read Fred Brook's Mythical Man Month for some insight.

> Two, they seem to spend a great deal of
> their time fixing bugs that keep reappearing. Also, the cross-platform
> compatibility issues also appear to be eating them up lately.

Cross platform compatibility is very, very hard to do, and gets harder
exponentially as features are added. Just testing for compliance is
vastly more expensive that you'd think at first guess. Even if you
restrict yourself radically, say to Windows 2000 or greater, there are a
lot of versions to test, a lot of patch levels, etc.

I think that many of those who whine the most don't understand the basic
problem. Its hard. And it is the essence of the problem, not an
accident. Development is not slow because it is written in Perl, or is
open source, or any of usual suspects. It is hard. It is a complex
program that does a lot of stuff.

Now, I would not want a closed SlimServer, because if a bug bothers me
enough, I can fix it myself. But YMMV, IMHO, etc.

--
Pat
http://www.pfarrell.com/music/slimserver/slimsoftware.html

JJZolx
2007-02-19, 16:50
Open or closed has nothing to do with speed of development. And adding
more developers usually slows development efforts. Microsoft claimed to
have 5,000 developers on Vista, it came out late, buggy and missing key
features. Read Fred Brook's Mythical Man Month for some insight.

I have. SlimServer ain't Vista or OS/360. It's a project with two or three full time developers and a handful of part-timers. Adding a couple of full time in-house software engineers to that total would do wonders.

There are obviously things going on right now at the former Slim Devices to which we aren't privvy. Hell, maybe they've already made the decision to abandon SlimServer as it now stands and are feverishly working on a proprietary Windows-only package being outsourced to a coding sweatshop in Russia or India.

pfarrell
2007-02-19, 18:09
JJZolx wrote:
> There are obviously things going on right now at the former Slim
> Devices to which we aren't privvy. Hell, maybe they've already made
> the decision to abandon SlimServer as it now stands and are feverishly
> working on a proprietary Windows-only package being outsourced to a
> coding sweatshop in Russia or India.

Where in the world do you get this? What is "obvious"?
And why would you even speculate that they'd move to Windows-only?

As KDF said, I really don't care what Logitech does, the existing code
works fine for me and I've got the source.

Looks like flame bait to me.


--
Pat
http://www.pfarrell.com/music/slimserver/slimsoftware.html

David Alexander
2007-02-19, 19:31
On Mon, 19 Feb 2007 15:24:26 -0800, JJZolx
<JJZolx.2m9zfn1171927501 (AT) no-mx (DOT) forums.slimdevices.com> wrote:

>
>Closed or open source, doesn't matter a wit to me. If making it closed
>source would somehow hasten SlimServer's development into something
>more stable and usable, I'd be all for it. SlimServer still feels
>pretty primitive to me and its progress into something more advanced is
>progressing at a snails pace.

Personally, I don't care too much about the open/closed source issue
at this time either. However, I always wonder why some people regard
Slimserver as such a primitive tool. I thought the original
SliMP3/Slimserver combination was already nearly perfect back with
version 1x or 2x or whatever. To me it was all about serving up music
to the hardware and providing a very usable remote control driven
interface in the living room. The web interface is/was just an
afterthought and not something I really used too much since it
basically means I have to go into the den to use it. Providing an
interface that can efficiently navigate through thousands of songs and
albums on a small two-line text display is not easy. Slim did it. So
did the Audiotron.

There seem to be two camps of people... 1) those like me who see
Slimserver as just one tool of several in the digital music toolbox,
and 2) those who think Slimserver should be some kind of do-all iTunes
killer. Personally, I'm fine with using other available tools to
rip/manage/tag/organize my library and just letting Slimserver be the
transport mechanism with a good usable interface at the device.

Makes me wonder...if Slimserver had never developed web interfaces at
all would people view the situation differently? That might actually
have made it easier to move into the mainstream.

totoro
2007-02-19, 20:10
\
Open or closed has nothing to do with speed of development. And adding
more developers usually slows development efforts.

On point one: well, sort of. I guess the issue here might not be open-source, so much as the leaders of the project being able to force/entice/whatever other contributors to do things they don't feel like doing. This is a classic problem in all-volunteer open source projects. Of course, you can have a way around this by simply having more employees of SD working on the code, so that the unglamorous/unpleasant/boring/miserable stuff is more likely to get done.

On point two, again, sort of, really depends. Sure it's the case that in a project that's going along with a dozen programmers, suddenly upping the number to a hundred probably won't help much. But if you have a company sponsored open source project with a couple of employees and some volunteers, adding new employees to the mix could help in numerous ways. One is addressed above. Regardless of what the mythical man month claims, you're throwing the baby out with the bathwater if you try to claim that it never or almost never helps to add developers.

I've certainly been in situations in my career when getting a couple bodies onto my projects when I was completely swamped was crucial, and the only thing that enabled the projects to succeed. I'd guess that many other people have, as well.

pfarrell
2007-02-19, 20:25
totoro wrote:
> On point one: well, sort of. I guess the issue here might not be
> open-source, so much as the leaders of the project being able to
> force/entice/whatever other contributors to do things they don't feel
> like doing. This is a classic problem in all-volunteer open source
> projects. Of course, you can have a way around this by simply having
> more employees of SD working on the code, so that the
> unglamorous/unpleasant/boring/miserable stuff is more likely to get
> done.

Yes, the only way I've ever seen folks get motivated to doing the "
unglamorous/unpleasant/boring/miserable stuff" is to pay them. And
sometimes that isn't enough. Software developers are rarely motivated by
money. They all want to do the cool stuff.

> One is addressed above. Regardless of what the mythical
> man month claims, you're throwing the baby out with the bathwater if
> you try to claim that it never or almost never helps to add
> developers.

I didn't claim that.
With tiny teams, sometimes adding developers helps.
But it is never without cost, and adding people only works if the tasks
are suitable for parallel development, and if the people are good.

But the Mythical Man Month speaks the truth. Adding people to a late
project makes it farther behind.

> I've certainly been in situations in my career when getting a couple
> bodies onto my projects when I was completely swamped was crucial, and
> the only thing that enabled the projects to succeed.

I've never seen a case where adding bodies helped. Adding talented and
motivated folks, sure. There is a difference.

But this is not comp.software.engineering so I'm not sure that this is a
useful thread for me to comment on.

As the wise man has said in many other threads: Patches welcome.

--
Pat
http://www.pfarrell.com/music/slimserver/slimsoftware.html

totoro
2007-02-19, 20:51
Pat Farrell wrote:

Yes, the only way I've ever seen folks get motivated to doing the "unglamorous/unpleasant/boring/miserable stuff" is to pay them. And sometimes that isn't enough. Software developers are rarely motivated by money. They all want to do the cool stuff.


I think you need a carrot and stick: forcing the employee to do the occasional necessary grunt-work, while giving rewards for it, and managing things so that nobody has to do _all_ grunt-work.

As to the money, I kind of like both, and I'd bet there are a lot of people with the same preference. My only point here is that sometimes I see people conflating "open-source" with "all or primarily volunteer". This doesn't have to be the case (as IBM, amongst others, has shown).


I didn't claim that. With tiny teams, sometimes adding developers helps. But it is never without cost, and adding people only works if the tasks are suitable for parallel development, and if the people are good.


Agreed. But it's not _that_ hard to find good people, and since slimserver's development is already heavily parallelized, I would think that it could absorb more people into it from SD, esp the roles that open source volunteers don't typically enjoy.


But the Mythical Man Month speaks the truth. Adding people to a late project makes it farther behind.


Agreed, mostly (caveat below). But I'm not sure I see how that is relevant. Nobody here is talking about adding 12 more developers to make the next arbitrary date for a release, as far as I can see.


I've never seen a case where adding bodies helped. Adding talented and motivated folks, sure. There is a difference.


Yep. When I spoke of adding bodies, I meant good ones: I was the primary developer on these projects, so it would have hurt me rather than helped me if they were stooges. Why would _anyone_ want to add untalented/unmotivated people to a project?

erland
2007-02-19, 23:37
Not wanting to start a flame war here, but seriously is the current open source approach to developing Slimserver viable?

Is it time for Logitech to fork the build, so that the community can take the GPL version for U*nix and there is a closed source Win32 binary for the man-in-the-street?This has nothing to do with open source. Logitech can already today choose to fork the build and make a Win32 binary, but I can't see any reason why the Win32 binary can't be open sourced.

One reason to choose the closed source path is if you want to make sure no competitor can use your work. SlimServer is already adapted for specific use with the SqueezeBox/Transporter and the SlimProto protocol used. This makes the source code almost unusable for any competitor which wants to use SlimServer with their own hardware.

Another reason to choose the closed source path is if you want to earn money on the software, SlimDevices has so far been focused on earning the money on the hardware. When moving to the massmarket it might be a good idea to focus a bit more on the software, but as long as SqueezeBox/Transporter is delivered with SlimServer and there is no alternative software I can't see any benefit of going the closed source path. Any new SlimServer replacement which costs money must also be a lot better than the current SlimServer since users always can choose to run the open sourced SlimServer version instead. So you would get a huge development cost to develop something new and it would take some time before you can start earn money on it since a similar open sourced version exist. In my personal opinion it is a lot better to focus these efforts on the hardware or the current open sourced version.

SlimServer must be good enough so it isn't a reason for people to choose other hardware, but it doesn't have to be the best software out there. There are two major complaints to SlimServer today, some people have problem to get it to work and other people complains about the usability of the web interface. These issues needs to be solved by Logitech, but as I see it both these issues will be a lot easier to solve in the current architecture compared to writing a Win32 replacement of SlimServer.

Aguida
2007-02-20, 03:01
I think this is really a good discussion and I hope somewhere Logitech and Slimdevices executive are reading it. On the other hand it is a pity if it gets side tracked by religious discussions about open vs. closed source software. It really doesn't matter whether the source is available to everybody and whether all the developers are Logitech employees or sitting somewhere in India (though I like open source :-) ).

What it really matters is the product development process used. Most people buys the hardware because of its superior sound quality (there are many more much cheaper player out there) and they need some software on the server side to make it work. They expect something with a defined set of features and very good reliability. It doesn't help anything top have exotic features if the main ones are buggy. It doesn't help to get it running on weird OS's if the main ones covering 80% or more of the buyers does not work properly.

I would like to see the community to take a step back and get things to work correctly for the majority. To do this two major steps are missing at the moment (I think):

1. Proper functional specification management.
The actual approach is pushing lots of new features into the product via bug reports. By now no one knows anymore how things are supposed to be working. Things are reported as bugs and often fixed but no one is able to say: here is how it is supposed to work. If it doesn't it's a bug. If we don't like how it is designed than we can vote for changes to the specifications. As an example, look at the whole browsing/tags/searching mess. As a new user I was completely lost. I tried the forum, tried the support and only after filing 2 bugs (which turned out to be lack of documentation and few bugs) I have figured out how to not get my Various Artists to appear under the single artists even though I had selected the "Group albums together" option. In the process process argued that this was "definetely a bug", "by design" and "an enhancement". Looking back in the histiry files I can see that changes have been made which contradict each other. This is clearly lack of specifications management.

2. TEST, AUTOMATED TEST, and TEST MANAGEMENT. Yes if there is no automated tests in place to make sure that changes in the code does not screw up somethign which used to work, then - as the product grwos - is only going to get worst. This is even more important if the product really becomes open-source and amny more people are allowed to work on the source code (or for that matter if it is outsourced somewhere). Everytime a bug is fixed a new test script should be created to make sure that the bug does not cripple in again.

Yes, this is going to be expensive, but so it is with high-quality: it costs!

I bought one of the best hi-fi players on the market and I expect similar quality from the software as well: "Reliability and easy of use".

My 2cc.

Alessandro

Browny
2007-02-20, 04:31
Personally I started with Slimserver when it lived at 6.2 - it was usable, buy did not do what I wanted...

Simply because it was open source I was able to communicate with the development team and get MY Slimserver to act as I wanted (I wanted to stop having to scan your file system and have a method to add albums to the database as I ripped them). Dan (the lead developer) made the effort to assist me with this.

I now have a customised Slimserver 6.5.1 that does exactly what I need - and what is more my Squeezebox/Slimserver is now not only the the coolest gadget I ever bought, but is also now also performing exactly as I would want it to.

There are issues with Slimserver 6.5 that I would like to see dealt with better, but that comes down to personal preference...

As my daytime job I am also a software developer and have to filter the personal preference from the 'enhancement to the system' that gets worked on - but my system is 'closed source' - mainly because my user base is the Business (who just expect IT to work). Ultimately at lot of what I do is a measure of benefit vs. cost - Sooner or later, whether this is closed or open source, you need SOMEONE to make that judgment - to be fair to SD this is something they do very well....I can see that my requirements are not those of the masses.

BUT...because it is open source I can change it for me if I want to.

Someone previously mentioned M$ - Try doing what I have done with a closed source model - it would be impossible.

As a final signoff - Slimserver is a FANTASTIC piece of software...please remember that before you have a go at it...

If you don't like it, you can join in and make it better (have a look at the Nokia 770 Skin, which is a superb example of someone wanting something and then making it available to the masses) - this is something that can never happen under a closed source model.

Peace!

Browny

mudlark
2007-02-20, 07:04
If anyone from Logitech is reading this please note that if Logitech pull the plug on the open source side of Slimserver then all my Logitech stuff will go in the bin.

A game controller, a keyboard and mouse, a joystick, a steering wheel and a squeezebox.

I will never buy anything from Logitech ever again and I will use all opportunity to encourage people to stay open source.

I feel very strongly about this stupid and dangerous comment from the first poster. If you think your views are reasonable go buy a closed source device and keep your comments off this forum.
M.
PS
I've a good experience with a PVR the Humax Duo visio 9200. This has been very buggy and is closed source. The updates are slow to arrive and the experience isn't much fun. Slimserver is just brilliant by comparison.

totoro
2007-02-20, 08:30
If anyone from Logitech is reading this please note that if Logitech pull the plug on the open source side of Slimserver then all my Logitech stuff will go in the bin.

A game controller, a keyboard and mouse, a joystick, a steering wheel and a squeezebox.

I will never buy anything from Logitech ever again and I will use all opportunity to encourage people to stay open source.

I feel very strongly about this stupid and dangerous comment from the first poster. If you think your views are reasonable go buy a closed source device and keep your comments off this forum.
M.
PS
I've a good experience with an PVR the Humax Duo visio 9200. This has been very buggy and is closed source. The updates are slow to arrive and the experience isn't much fun. Slimserver is just brilliant by comparison.

Chill out. Other people have other needs, and didn't necessarily buy the sb because it was open source, but because it was the best thing out there. Your insistence that they should shut up or go away adds nothing of any value to the discussion.

Open sourcing a project is neither a panacea nor a guarantee of a disaster. The issue of bringing in people from sd to do things like integrate a unit test harness, make sure that there is good test coverage, blah blah blah (all the stuff many developers don't like doing) is pretty pertinent, and imho, a lot of people are actually thinking about things like this when talking about going closed source (incorrectly conflating some issues, as I've pointed out earlier).

Have you yourself ever actually written any significant software, so that you have _any_ basis at all for talking about how "stupid and dangerous" this discussion is? I have, and find your position completely insupportable.

nicketynick
2007-02-20, 09:36
Makes me wonder...if Slimserver had never developed web interfaces at
all would people view the situation differently? That might actually
have made it easier to move into the mainstream.

That's a good point - hard to complain about it if it doesn't exist...... but that would make it a Roku with a better player user interface, wouldn't it? I've used both, and SB/SlimServer is far better IMO.
And let's keep things in perspective here - Slim Devices never got big by itself - we're still in 'early-adopter/niche market' days here. I expect Logitech will be introducing some products that are barely recognizable as brethren of the SB3 we know and love in order to try to penetrate the mass-market - heck - it will probably be marketed as an 'mp3 player' (egads!)

mudlark
2007-02-20, 09:38
Chill out. Other people have other needs, and didn't necessarily buy the sb because it was open source, but because it was the best thing out there. Your insistence that they should shut up or go away adds nothing of any value to the discussion.

Open sourcing a project is neither a panacea nor a guarantee of a disaster. The issue of bringing in people from sd to do things like integrate a unit test harness, make sure that there is good test coverage, blah blah blah (all the stuff many developers don't like doing) is pretty pertinent, and imho, a lot of people are actually thinking about things like this when talking about going closed source (incorrectly conflating some issues, as I've pointed out earlier).

Have you yourself ever actually written any significant software, so that you have _any_ basis at all for talking about how "stupid and dangerous" this discussion is? I have, and find your position completely insupportable.

Hi totoro,

Thanks for your comments. I am a bit sad because i have little compassion for some folks who want a world safe from any slight complications.

i bought a squeezebox because of it's ability do a simple job for me and because it got a good write up from the reviewers. The community looked good also. I didn't buy it because it was open source. Since getting the Sb I have been encouraged to go Linux. This has been very helpful for me as it has given me hours of enjoyment and contact with a community which seems integrated with the developers and designers of the hardware. Before this experience I had some issues with conflicts in Windows XP. By experiment and ideas off the internet i sorted my problems out myself. I posted my experience and I hope that has helped others. This experience involved Logitech who were brilliant and helpful even when, in the end, they were not the guilty party. I got chuff all help from the software side. They wanted money with no guarantee of success. Logitech's attitude was great and I bought other stuff from them because of this experience. My conclusions are that the open source system seems to integrate users, developers and manufacturers closer together which helps me. I can't speak for others. I know the open source system is better for me.

I haven't got any experience of writing any code or "significant software", the reason why I am worried about the SB project going closed shop is that it will reduce the very great pleasure I get from the situation as it is.

I see too much of the desperate drive to come up with a device to save the world (ipod style) and make lots of money. The sb model will never become run of the mill in the short term. All i am hoping is that this nerdy type device remains so accessible and enjoyable and does not become corrupted. I respect ipod and apple for what they have done, but PLEASE don't bugger up the sb model by wanting to make it the answer to Mrs Smith's cravings for a plug and play model.
M.

ChrisOwens
2007-02-20, 10:06
Well, there are things going on at Slim Devices to which *I'm* not privy, but I'll go out on a limb and say a few things. At any other company, I would run such a post by my manager, and probably his manager, and maybe the PR director, but things are still "open" enough here that I think my own judgment is good enough for this post.

The changes going on at Slim Devices broadly fall into two groups: Logitech wants to capitalize on its investment after the acquisition, and Slim Devices wants to use the financial backing of Logitech to organize itself into a well-run business unit instead of a startup running on a shoestring. Add into the mix that all of us here are very mindful that we don't want to lose any of the Slim Devices "magic", whether that be engineering, quality, customer service, or "community".

Within a few months, you're going to see Squeezeboxes in major retail outlets for which the old Slim Devices could have never manufactured enough units, and advertising in places the old Slim Devices could have never afforded. Logitech has manufacturing and marketing muscle, and it's not something that's just available to us, it's part of Logitech's plan to sell a lot of Slim Devices products. And that's a good thing, because...

We can now afford to take a few risks on new products. I'm still going to follow Sean's lead on not announcing anything before its time, but there are projects that Sean and Dean have wanted to do for a long time, as well as some things that might work well with Logitech's more traditional customer base that are now gathering steam.

These two areas are where we're spending our energy.

Throwing out the Slim Devices way of doing business is not part of the plan. I have been in meetings at Logitech now where we discuss products that have been flops (or worse, are in the process of flopping). I don't want to insult my new coworkers, but many of these flops are due to very strange thinking from a Slim Devices point of view. You can spend a lot of time doing market research and interface design and writing specifications and developing hardware and software, release it, fix some bugs, and start the whole cycle over again...

Or you can make it powerful enough and flexible enough and open enough that the *customers* can *turn it into* the product that *they want to use*. Naturally there still has to be support and direction on the Slim Devices side, but there are definitely real business benefits to the Open Source approach.

Now, looking back at what I've written, can you see where in these plans and priorities killing Slimserver would fit in? Do see where a plan to spend a lot of legal and development energy reinventing the wheel and ending up with a proprietary software package would fit in?

Me neither.

bklaas
2007-02-20, 10:21
Not wanting to start a flame war here, but seriously is the current open source approach to developing Slimserver viable?

Yes.

#!/ben

nicketynick
2007-02-20, 10:22
Wow, that's more than we've heard for you guys in a long time! Thanks Chris!
And you thought that every word that the Fed Chairman (Bernanke, and Greenspan before him) said was carefully dissected - watch this space!
Have to admit I'm on pins and needles now - I was starting to get discouraged, but now I can't wait to see what comes next - and April 1st is coming quick!

amcluesent
2007-02-20, 10:39
Good to hear that plans in-place to match the great capabilities of the Squeezebox/Slimserver (when it running well) with a mature, disciplined s/w development process.

As was pointed out, I was using 'open source' to conflate numerous dimensions on which to place slimserver, such as -

Open source - closed source as a licensing model
Community - in-house team for s/w build and testing
multi-platform - Wintel as target
Binary - perl as technology stack

On reflection, my original assertion was that Slimserver was currently placed at the atypical ends of many of these dimensions which created risks as well as opportunities and that the risk outweighed the opportunity.

The mass-market for Squeezeboxes would be people who have already got a ripped music library, which would overwhelmingly mean iPod owners using Windows XP - that's just a fact.

So would the current slimserver "delight" that marketplace? IMHO, no it wouldn't as of 6.5.1 which would more likely create "shock and awe". And I would wager that a non-trivial number of Squeezebox buyers aren't even aware they need to use slimserver and find the current process of getting it running to their satisfaction quite 'challenging'.

Looking forward to slimserver 7. :-)

peter
2007-02-20, 11:02
ChrisOwens wrote:
> Well, there are things going on at Slim Devices to which *I'm* not
> privy, but I'll go out on a limb and say a few things. At any other
> company, I would run such a post by my manager, and probably his
> manager, and maybe the PR director, but things are still "open" enough
> here that I think my own judgment is good enough for this post.
>
> The changes going on at Slim Devices broadly fall into two groups:
> Logitech wants to capitalize on its investment after the acquisition,
> and Slim Devices wants to use the financial backing of Logitech to
> organize itself into a well-run business unit instead of a startup
> running on a shoestring. Add into the mix that all of us here are very
> mindful that we don't want to lose any of the Slim Devices "magic",
> whether that be engineering, quality, customer service, or "community".
>
>
> Within a few months, you're going to see Squeezeboxes in major retail
> outlets for which the old Slim Devices could have never manufactured
> enough units, and advertising in places the old Slim Devices could have
> never afforded. Logitech has manufacturing and marketing muscle, and
> it's not something that's just available to us, it's part of Logitech's
> plan to sell a lot of Slim Devices products. And that's a good thing,
> because...
>
> We can now afford to take a few risks on new products. I'm still going
> to follow Sean's lead on not announcing anything before its time, but
> there are projects that Sean and Dean have wanted to do for a long
> time, as well as some things that might work well with Logitech's more
> traditional customer base that are now gathering steam.
>
> These two areas are where we're spending our energy.
>
> Throwing out the Slim Devices way of doing business is not part of the
> plan. I have been in meetings at Logitech now where we discuss
> products that have been flops (or worse, are in the process of
> flopping). I don't want to insult my new coworkers, but many of these
> flops are due to very strange thinking from a Slim Devices point of
> view. You can spend a lot of time doing market research and interface
> design and writing specifications and developing hardware and software,
> release it, fix some bugs, and start the whole cycle over again...
>
> Or you can make it powerful enough and flexible enough and open enough
> that the *customers* can *turn it into* the product that *they want to
> use*. Naturally there still has to be support and direction on the
> Slim Devices side, but there are definitely real business benefits to
> the Open Source approach.
>
> Now, looking back at what I've written, can you see where in these
> plans and priorities killing Slimserver would fit in? Do see where a
> plan to spend a lot of legal and development energy reinventing the
> wheel and ending up with a proprietary software package would fit in?
>
> Me neither.
>

Thanks for the reassuring words, Chris. We all hope the SD of making and
supporting will take over the world even if it will mean we'll be a
little less l33t.

Regards,
Peter

bdelp
2007-02-20, 11:09
Thanks Chris for that very informative post. This really eases my fear of the whole Logitech deal.
Vive La Open Source!
BD

mudlark
2007-02-20, 12:13
Well, there are things going on at Slim Devices to which *I'm* not privy, but I'll go out on a limb and say a few things. At any other company, I would run such a post by my manager, and probably his manager, and maybe the PR director, but things are still "open" enough here that I think my own judgment is good enough for this post.

The changes going on at Slim Devices broadly fall into two groups: Logitech wants to capitalize on its investment after the acquisition, and Slim Devices wants to use the financial backing of Logitech to organize itself into a well-run business unit instead of a startup running on a shoestring. Add into the mix that all of us here are very mindful that we don't want to lose any of the Slim Devices "magic", whether that be engineering, quality, customer service, or "community".

Within a few months, you're going to see Squeezeboxes in major retail outlets for which the old Slim Devices could have never manufactured enough units, and advertising in places the old Slim Devices could have never afforded. Logitech has manufacturing and marketing muscle, and it's not something that's just available to us, it's part of Logitech's plan to sell a lot of Slim Devices products. And that's a good thing, because...

We can now afford to take a few risks on new products. I'm still going to follow Sean's lead on not announcing anything before its time, but there are projects that Sean and Dean have wanted to do for a long time, as well as some things that might work well with Logitech's more traditional customer base that are now gathering steam.

These two areas are where we're spending our energy.

Throwing out the Slim Devices way of doing business is not part of the plan. I have been in meetings at Logitech now where we discuss products that have been flops (or worse, are in the process of flopping). I don't want to insult my new coworkers, but many of these flops are due to very strange thinking from a Slim Devices point of view. You can spend a lot of time doing market research and interface design and writing specifications and developing hardware and software, release it, fix some bugs, and start the whole cycle over again...

Or you can make it powerful enough and flexible enough and open enough that the *customers* can *turn it into* the product that *they want to use*. Naturally there still has to be support and direction on the Slim Devices side, but there are definitely real business benefits to the Open Source approach.

Now, looking back at what I've written, can you see where in these plans and priorities killing Slimserver would fit in? Do see where a plan to spend a lot of legal and development energy reinventing the wheel and ending up with a proprietary software package would fit in?

Me neither.

Shit hot, thanks heavens for that, YES, etc

Chris, Thanks for taking the risk with such a post. I'm an insignificant git from cumbria in the UK. I got treated well by a nice Logitech chap from Switzerland and now I hope I get the drift about a company in the USA. Please don't let us down!

All the best,

Mike.

Marc Sherman
2007-02-20, 12:19
amcluesent wrote:
>
> The mass-market for Squeezeboxes would be people who have already got a
> ripped music library, which would overwhelmingly mean iPod owners using
> Windows XP - that's just a fact.
>
> So would the current slimserver delight that marketplace? IMHO, no it
> wouldn't as of 6.5.1 And I would wager that a non-trvial number of
> Squeezebox buyers aren't even aware they need to run slimserver.

So the smart thing for Logitech to do isn't to close the source, but to
concentrate their investment on QA and Dev staff to shore up that
(admittedly poorly served) use case. The existing community can keep
testing/improving the code for the fringe platforms we all run, just fine.

- Marc

ezkcdude
2007-02-20, 12:41
SS is great as is, my GF even told me this after looking at it for the first time. She goes, "If I had this, I'd be at my computer for hours at a time." No kidding, I told her. Why change just for the sake of change. The Open Source model obviously is working. How do I know? Because I'm using the software, and I'm not one of the developers. To me that is the definition of "it just works". Sure, other people are devoting resources and time to make it work, but as long as that remains the case, why close the source? If the open source community dries up, then SD/Logitech could take over. Until then, why mess with a good thing? JMTCW.

Nostromo
2007-02-20, 13:59
Throwing out the Slim Devices way of doing business is not part of the plan. I have been in meetings at Logitech now where we discuss products that have been flops (or worse, are in the process of flopping). I don't want to insult my new coworkers, but many of these flops are due to very strange thinking from a Slim Devices point of view. You can spend a lot of time doing market research and interface design and writing specifications and developing hardware and software, release it, fix some bugs, and start the whole cycle over again...

Or you can make it powerful enough and flexible enough and open enough that the *customers* can *turn it into* the product that *they want to use*. Naturally there still has to be support and direction on the Slim Devices side, but there are definitely real business benefits to the Open Source approach.


I think that going open source was brilliant move by Slimdevices. But a counter-example would be Apple. The iPod is hugely popular, but its barely customizable, compared to the Squeezebox. I'm not saying that customizability doesn't explain, in part, the Squeezebox's success. It does, IMO. Its one reason why I love it. I'm just saying it doesn't necessarily explain why the Logitech product in question failed.

JJZolx
2007-02-20, 14:15
Wow, that's more than we've heard for you guys in a long time! Thanks Chris!

No, it's pretty much the *only* thing we've heard in a long time.

snarlydwarf
2007-02-20, 14:17
Wow, that's more than we've heard for you guys in a long time! Thanks Chris!
And you thought that every word that the Fed Chairman (Bernanke, and Greenspan before him) said was carefully dissected - watch this space!
Have to admit I'm on pins and needles now - I was starting to get discouraged, but now I can't wait to see what comes next - and April 1st is coming quick!

I will settle for getting the KRONOS shipping...

http://www.slimdevices.com/au_press_kronos.html

I can almost guarantee there will be a new product announced on April 1st.

mudlark
2007-02-20, 14:33
nearly LOL, brilliant.

every company has someone with the foresight to be really at the cutting edge.

promote that man/woman immediately.

kdf
2007-02-20, 14:45
Quoting JJZolx <JJZolx.2mbobb1172006401 (AT) no-mx (DOT) forums.slimdevices.com>:

>
> nicketynick;181952 Wrote:
>> Wow, that's more than we've heard for you guys in a long time! Thanks
>> Chris!
>
> No, it's pretty much the *only* thing we've heard in a long time.

given how unpleasant the responses often get, no shock there.
-kdf

peter
2007-02-20, 14:51
Nostromo wrote:
> ChrisOwens;181947 Wrote:
>
>> Throwing out the Slim Devices way of doing business is not part of the
>> plan. I have been in meetings at Logitech now where we discuss
>> products that have been flops (or worse, are in the process of
>> flopping). I don't want to insult my new coworkers, but many of these
>> flops are due to very strange thinking from a Slim Devices point of
>> view. You can spend a lot of time doing market research and interface
>> design and writing specifications and developing hardware and software,
>> release it, fix some bugs, and start the whole cycle over again...
>>
>> Or you can make it powerful enough and flexible enough and open enough
>> that the *customers* can *turn it into* the product that *they want to
>> use*. Naturally there still has to be support and direction on the
>> Slim Devices side, but there are definitely real business benefits to
>> the Open Source approach.
>>
>>
>
> I'm not disagreeing with you. I think that going open source was
> brilliant move by Slimdevices. But a counter-example would be Apple.
> The iPod is hugely popular, but its barely customizable, compared to
> the Squeezebox.
>

Apple has huge funds for marketing. SD hadn't. They targeted a specific
group of tech savvy people who would want such a product and consider
the OS nature of the system an advantage as well as insurance against
the company going under. I was one of those who bought the first SliMP3
and it was a bit of a leap of faith to buy something from an unknown
company that had only just obtained a back cover. But the system was
exactly what I wanted, so I got one anyway. They weren't cheap back then
either.

Regards,
Peter

JJZolx
2007-02-20, 15:25
> No, it's pretty much the *only* thing we've heard in a long
> time.

given how unpleasant the responses often get, no shock there.

What do you think it's going to be like once WalMart and Target start selling these things? I suppose when that happens we can expect absolutely no communication from Slim Devices in these forums.

Maybe it's just me, but doesn't "open" source imply some level of communication, at least among developers? Or is it felt that making the source code public is the only requirement? As I've noted elsewhere, SlimServer development has all but stopped in its tracks, I guess while other projects take precedence. To me, this is the most important indicator of how well the current software development approach - call it what you like - is doing. When the entire company and all of its develppers decide to focus on something else and not even major bugs are being addressed, then what difference does it make?

mherger
2007-02-20, 15:35
> Maybe it's just me, but doesn't "open" source imply some level of
> communication, at least among developers?

Don't worry, there is.

--

Michael

-----------------------------------------------------------------
http://www.herger.net/SlimCD - your SlimServer on a CD
http://www.herger.net/slim - AlbumReview, Biography, MusicInfoSCR

kdf
2007-02-20, 15:45
Quoting JJZolx <JJZolx.2mbrjz1172010601 (AT) no-mx (DOT) forums.slimdevices.com>:

>
> kdf;182057 Wrote:
>>
>> > No, it's pretty much the *only* thing we've heard in a long
>> > time.
>>
>> given how unpleasant the responses often get, no shock there.
>
> What do you think it's going to be like once WalMart and Target start
> selling these things?

That's irrelevant.

> I suppose when that happens we can expect
> absolutely no communication from Slim Devices in these forums.

incorrect. plenty from andy and ross.

> Maybe it's just me, but doesn't "open" source imply some level of
> communication, at least among developers?

there is plenty of communication.

> Or is it felt that making
> the source code public is the only requirement?

I'll leave the reading of the GPL for you.

> As I've noted
> elsewhere, SlimServer development has all but stopped in its tracks

And it's been noted that you are way off base on that one. Totally incorrect.
Just because YOUR bugs aren't getting 4 posts a day and one fixed
every couple of days doesn't mean it's stopped. It just means no one
needs any information from you.

svn is fully viewable. Anyone can see just how much is going on.

plain FUD.

-kdf

Ben Sandee
2007-02-20, 20:56
On 2/20/07, kdf <slim-mail (AT) deane-freeman (DOT) com> wrote:
>
>
> > As I've noted
> > elsewhere, SlimServer development has all but stopped in its tracks
>
> And it's been noted that you are way off base on that one. Totally
> incorrect.
> Just because YOUR bugs aren't getting 4 posts a day and one fixed
> every couple of days doesn't mean it's stopped. It just means no one
> needs any information from you.


Also, as you know kdf, SD will often go 'offline' for non-public projects
and then a massive svn merge happens once the info is public. I think they
may be targeting more for 7.0 than is currently on the roadmap. Some people
just need to complain, don't let them get you down.

Ben

JJZolx
2007-02-20, 21:14
SD will often go 'offline' for non-public projects
and then a massive svn merge happens once the info is public. I think they
may be targeting more for 7.0 than is currently on the roadmap. Some people
just need to complain, don't let them get you down.


Not a complaint, just an observation about the lack of substantial visible work on SlimServer over the past four months. If there's a complaint, it's only about the complete lack of communication that was so commonplace before the Logitech buyout.

Ah well, things change. We have Squeezebox magazine ads and Logitech branded SBs at Sam's Club to look forward to. REAL progress...

Skunk
2007-02-20, 22:51
So if it is sustainable, and these will be on shelves, this is all good news. People learn about open source software, while freeing their music. Both worthy causes, and if represented on the shelves of wal mart then all the better.

Wal mart people need a wizard and probably app too, though. Sb3 being on the shelves now would be a bit like making the horse push the cart IMHO.

FWIW sean still posts here a lot.

seanadams
2007-02-20, 22:52
Not a complaint, just an observation about the lack of substantial visible work on SlimServer over the past four months.

I would really like to understand where you're getting that impression, because I get a steady stream of code checkins and bug updates all day long which suggest quite the opposite. That's not to mention other new non-SS releases such as Rhapsody, which take many man-months to complete... perhaps you don't care, but others certainly do!

What you're saying is demonstrably 100% false, so I can only guess that there some specific thing you want to see from us that is not happening?

werock
2007-02-21, 02:12
It is easy to get downhearted when the complainers seem to have a louder voice than anyone else, but take this post that I spotted over on the Topfield forums as inspiration:

>>Good grief. I've just got my Squeezebox. I had installed the server t'other day. I plugged it in and after 10mins was merrily playing. As opposed to the 2 weeks for my MP101 and it's flakey software.

>>It's, quite simply, the most awe-inspiring piece of home audio I've used. I dread to think how good a Transporter must be.

>>If you're in the market for a network music player, buy this one. And have big grin on face!

mudlark
2007-02-21, 02:48
Closed or open source, doesn't matter a wit to me. If making it closed source would somehow hasten SlimServer's development into something more stable and usable, I'd be all for it. SlimServer still feels pretty primitive to me and its progress into something more advanced is progressing at a snails pace. I'm not exatly sure what to attribute that to. One, there are very few developers on the project, despite the open source approach. Two, they seem to spend a great deal of their time fixing bugs that keep reappearing. Also, the cross-platform compatibility issues also appear to be eating them up lately.

This comment seems to me to be from a person who has made a smart comment which is completely out with my experience as a typical user. I don't look for problems. My slimserver/SB works well and is better than some pieces of hardware/software on XP. On Linux which is now my preferred OS the experience was involving and I can't see where the large number so called bugs are. I saw a slight problem with the top part of the web page on Konquerer, but the Firefox experience is nothing short of excellent. The system works superbly at playing music. I regard the web interface as a fantastic and well designed way to use the system. I find the design better than any other music playing software and I have tried many including WMP, nero, winamp etc al.

Sean your device is great and, believe it or not, has helped convince me of hope for the future in that there are helpful people all over the world even in the state of Arne.

Mike.

moley6knipe
2007-02-21, 06:26
It is easy to get downhearted when the complainers seem to have a louder voice than anyone else, but take this post that I spotted over on the Topfield forums as inspiration:

>>Good grief. I've just got my Squeezebox. I had installed the server t'other day. I plugged it in and after 10mins was merrily playing. As opposed to the 2 weeks for my MP101 and it's flakey software.

>>It's, quite simply, the most awe-inspiring piece of home audio I've used. I dread to think how good a Transporter must be.

>>If you're in the market for a network music player, buy this one. And have big grin on face!

That was me! To my mind the SB3 is the Topfield of the network music player world, and there really is no higher compliment as far as I'm concerned.

I've come from using a Netgear MP101. Netgear = big company. Server bundled with MP101 = doesn't work. That's not an opinion, it's a fact; it would NEVER find any albums past the letter D. So I used Twonky, but the bugs with that led me to look at my equipment again, i.e. various artist support=broken, iTunes playlists not found for no valid reason etc. I know Twonky's got more platforms and clients to support (not to mention video and pics).

My personal experience. I read up on the SB3. I downloaded the user guide. It said install SS first. I did so. It found everything first time, even with McAfee running. I bought SB3. I plugged it in. I tweaked my preferences. I was playing music within 10 minutes.

I would say I'm an average user. Win XP Home, a wireless router, iTunes, and iPod. I know the basics of how systems like SS/SB3 work, but nothing in depth. In my opinion, SS 6.5.1 is configured perfectly for Mr Average straight "out of the box". It just works. The sound quality is superb. I've by no means got an audiophile set-up, but what I've got it's made sound 10x better.

The great thing about SS, in my short use, is that it works out of the box, but if you want to take it further, the possibilities are endless. It strikes me these sort of extras can only be done if SS is open source. A good parallel is m other fave gear, the Topfield 5800. It works straight out of the box, albeit the user interface isn't massively intuitive - from reading this post some would say the same about SS (not me, coz I've only got experience of 3 servers!). However if you really want to make the Toppy fly, install TAPs (extra software packages) - again, this is only possible because Topfield have designed the box to be tweaked and changeable.

I think if Logitech start getting this product on to the shelves in bigger retail outlets it'll do fine - as long as the retailers don't sit it on the shelves near the mp3 players! The challenge comes with the extra documentation needed, I guess. Average PC World/Currys customer won't want to be directed to a forum if they have problems!

Anyway, no problems yet, but I'm sure I'll find the answers here if I have any!

superbad
2007-02-21, 18:58
I don't read these forums much anymore, because SS has improved so much over the last few years that I don't need to. My SB1 just works now, every day, all day. I can't remember the last time it gave me any trouble at all. Thanks Sean and everyone else who has helped make that happen. That's all I wanted to say. I look forward to seeing what new products are on the way.

Balthazar_B
2007-02-22, 13:30
Going back to the first message in this thread (and others similar to it), I think the root issue underlying whatever frustration users/customers are feeling is not really open vs. closed source. Rather, it generally comes down to performance (i.e., perceived lack of) and/or bugs (usually real, rarely imagined). As many have mentioned, better development processes/control (incl. regression testing, etc.) would help address the latter. The former is more difficult, and perhaps a large portion of performance issues come down to the code base. There may be certain operations that PERL, PHP, etc., may never be able to perform as efficiently as C dialects or other compiled programming languages.

I have no idea what's towards the top of Logitech/Slim's priority list, but I'm wondering whether they aren't stepping back to take a look at the whole server package, and figuring out whether some parts of it shouldn't be modularized and then written using a code base that will address sluggishness and other performance symptoms. Of course this wouldn't be easy, but it would probably make it easier to release a server-in-a-box product (if that's what they have in mind) that would not get savaged by reviewers testing with large and/or complex media libraries.

kdf
2007-02-22, 13:42
Quoting Balthazar_B
<Balthazar_B.2mfbkb1172176501 (AT) no-mx (DOT) forums.slimdevices.com>:

>
> and/or bugs (usually real, rarely imagined).

I think that is a statement that can easily lead to long hours of debate.

This is starting to sound like yet another excuse to be condescending
towards people who might actually know what they are doing. It seems
"open source" also gains the perception that only untrained hacks take
part.

Then again, maybe I've just been imagining the increasing QA team.
-kdf

peter
2007-02-23, 00:33
Balthazar_B wrote:
> Going back to the first message in this thread (and others similar to
> it), I think the root issue underlying whatever frustration
> users/customers are feeling is not really open vs. closed source.
> Rather, it generally comes down to performance (i.e., perceived lack
> of) and/or bugs (usually real, rarely imagined). As many have
> mentioned, better development processes/control (incl. regression
> testing, etc.) would help address the latter. The former is more
> difficult, and perhaps a large portion of performance issues come down
> to the code base. There may be certain operations that PERL, PHP,
> etc., may never be able to perform as efficiently as C dialects or
> other compiled programming languages.
>

Perl is a compiled language. It's compiled when you start the program.
There's no reason a Perl program can't be lightning fast. I have Perl
programs processing thousands of users at a very high speed, it's all in
the coding, of course. I have the feeling slimserver could be optimized
to make certain operations that cause the slowness perception (like
recursively adding a directory to a playlist) much faster. The web
interface is responsible for a lot of the perceived slowness. I think it
should be complemented with a slick GUI app that reacts instantaneously.

Regards,
Peter