PDA

View Full Version : Pay your way to a better slim server?



chrisla
2006-01-18, 22:05
In browsing around http://www.rentacoder.com by way of a slashdot
article I was wondering. Would people be interested in going in on
paying to get say the top voted X bugs or RFE's fixed/completed?

As someone with more money than coding skills, this seems like an
appealing idea to me. Anyone else interested?

(Granted slimdevices is a for-profit company, and thus consideration
should be given, however they also have already given a good deal to
the community as well.)

If there was some interest either some sort of escrow system would
have to be setup. Maybe trusting in sending your money to slim, who
would in turn bid on rentacoder on the community's behalf with someone
else managing the contractor.

-Chris

paulcolley
2006-01-19, 08:20
On 1/19/06, Chris Laplante <chrisla23 (AT) gmail (DOT) com> wrote:
>
> Would people be interested in going in on
> paying to get say the top voted X bugs or RFE's fixed/completed?
>
> As someone with more money than coding skills, this seems like an
> appealing idea to me. Anyone else interested?
>

This has already happened at least once --- On this mailing list, I offered
$US 20 or $CAN 30 for a software version of Squeezebox, and Richard started
work on Softsqueeze less than a week later, and met all my requirements in
less than two weeks from the time of my offer. Amazing!

I was very pleasantly surprised at the results, especially considering the
pittance I offered. I'm not sure Richard's wife will forgive me, but
that's a separate issue :-).

My original requirements said that for synchronization, "I'd be happy for
differences of 30 seconds". You can now all blame me for the
synchronization problems in Softsqueeze.

So... why don't you just make an offer on the developer list?

Mark Lanctot
2006-01-19, 09:18
I have even less money than coding skills right
now, but this doesn't feel right to me.

However a lot of open source software projects are
feeling cash crunches right now - DD-WRT Linksys
WRT54G router firmware, for example. These people
have to eat and have a right to the good things in
life for all their hard work.

Slim is a bit of a special case since it's
supported by its hardware sales and the software
is essential for the hardware, but if this "free"
software development starts to significantly
affect Slim's finances, something should be done.

Perhaps I'm looking at it from a socialized
medicine perspective ;-) but it seems like
queue-jumping to me, i.e. "your problem will only
get fixed if you pay us, and even then, if someone
outbids you on their problem, we'll work on theirs
first".

Maybe I shouldn't expect things to be free
forever, but for some of us a Squeezebox is a
large investment and pumping more money into it on
an ongoing basis is not very appealing. I don't
mind paying for players when/if I require them as
they are worth it though.

Chris Laplante wrote:
> In browsing around http://www.rentacoder.com by way
of a slashdot
> article I was wondering. Would people be interested
in going in on
> paying to get say the top voted X bugs or RFE's
fixed/completed?
>
> As someone with more money than coding skills, this
seems like an
> appealing idea to me. Anyone else interested?
>
> (Granted slimdevices is a for-profit company, and
thus consideration
> should be given, however they also have already
given a good deal to
> the community as well.)
>
> If there was some interest either some sort of
escrow system would
> have to be setup. Maybe trusting in sending your
money to slim, who
> would in turn bid on rentacoder on the community's
behalf with someone
> else managing the contractor.
>
> -Chris
>

Jeff Coffler
2006-01-19, 09:43
Not intending to put a value judgement on the "pay for a better SlimServer"
issue, but:

From: "Mark Lanctot" <marklanctot (AT) yahoo (DOT) ca>


>I have even less money than coding skills right
> now, but this doesn't feel right to me.
>
> ...
>
> Maybe I shouldn't expect things to be free
> forever, but for some of us a Squeezebox is a
> large investment and pumping more money into it on
> an ongoing basis is not very appealing. I don't
> mind paying for players when/if I require them as
> they are worth it though.

The other way of looking at this:

Name any other product (or any product at all, really) that offers unlimited
software upgrades (and new features from time to time, etc) for the life of
the product, where the life of the product is essentially "unlimited". I'm
failing to come up with anything.

I use a ton of software from a variety of vendors: Microsoft, Intuit, Cisco,
HP, Van Dyke, Palm, VMware, and others (anti-virus, etc). None of these
vendors offer unlimited upgrades (with new features) forever. The closest
is Palm, but they never offer "new features"; just the ability to continue
to run their desktop software on future versions of the O/S (if you want new
Palm features, you buy new hardware). A number of vendors (primarily
Microsoft and Cisco, I guess) will offer new versions (and new features) if
you pay for maintenance. But we SlimServer users don't do that either.

Now, I'm a heavy user of Fedora as well, and that's free. But that's
totally open source, and not funded at all by hardware sales. So it's not
really the same.

I like the SlimServer software being free. I've also contributed to the
SlimServer software (I worked with Fred on the SlimServerMod extension to
allow the AMX NetLinx automation system drive a SlimServer). This would
have been "trickier" if it weren't totally open source software.

That said: if Slim Devices were to do what most other companies do, they'd
charge for new major versions of their software. Doing this would change
the cost structure a lot, and likely fund a number of engineers hired by
Slim Devices. It might also kill the non-Slim contributors to the
SlimServer (unsure).

-- Jeff

Mark Lanctot
2006-01-19, 10:24
Jeff Coffler wrote:
> Not intending to put a value judgement on the "pay
for a better
> SlimServer" issue, but:
>
> From: "Mark Lanctot" <marklanctot (AT) yahoo (DOT) ca>
>
>
>> I have even less money than coding skills right
>> now, but this doesn't feel right to me.
>>
>> ...
>>
>> Maybe I shouldn't expect things to be free
>> forever, but for some of us a Squeezebox is a
>> large investment and pumping more money into it on
>> an ongoing basis is not very appealing. I don't
>> mind paying for players when/if I require them as
>> they are worth it though.
>
> The other way of looking at this:
>
> Name any other product (or any product at all,
really) that offers
> unlimited software upgrades (and new features from
time to time, etc)
> for the life of the product, where the life of the
product is
> essentially "unlimited". I'm failing to come up
with anything.

Perhaps it's that the Squeezebox is such an
unusual product in that the hardware is intensely
dependent on the software.

For regular consumer products, product updates and
bugfixes ARE free. Think of router firmware, for
example. No one charges for that. But on the
other hand, these aren't updated forever (a few
years at most), and few new features are added,
with new features being added to the next piece of
hardware and often finalized before release.

Slim's current policies seem similar. The SLIMP3
and Squeezebox1 are still supported, but the only
firmware changes coming will be bugfixes. The
hardware has essentially reached its limit and new
features are only being added to the newer hardware.

The closed firmware and open-source software form
a well-defined demarcation point for Slim.

The software is a grey area as new features can be
added to it which are quite usable with older
hardware, so it's kind of a mixed case. It's also
a special case here as software changes can
*radically* alter player performance and
functionality, more so than almost any other
device you can think of.

>
> That said: if Slim Devices were to do what most
other companies do,
> they'd charge for new major versions of their
software. Doing this
> would change the cost structure a lot, and likely
fund a number of
> engineers hired by Slim Devices. It might also kill
the non-Slim
> contributors to the SlimServer (unsure).
>

It is a tricky change. Slim could squash a lot of
bugs quickly this way but they may have to abandon
the open-source concept which would stifle or halt
3rd party development and may alienate users.

Open source can co-exist with a financial
structure, mostly in the form of donations at this
point.

I'm fortunate in that there are a few bugs that
affect me and none of them are "show stoppers".

But there's an absolutely massive bug list for
SlimServer right now. It's being well managed but
it must be incredibly hard to handle it and it's
far more work than Slim can handle. Some of the
enhancement requests seem frivolous - no offence
to anybody. However they are equally as important
to the people that created them as the ones I
created or voted on, and they were intended to
help everyone out by making the Squeezebox even
more useful and functional.
--
___________________________________


Mark Lanctot
___________________________________

Jeff Coffler
2006-01-19, 12:36
From: "Mark Lanctot":

> For regular consumer products, product updates and
> bugfixes ARE free. Think of router firmware, for
> example. No one charges for that. But on the
> other hand, these aren't updated forever (a few
> years at most), and few new features are added,
> with new features being added to the next piece of
> hardware and often finalized before release.

"Low-end" routers might have free updates (and few new features), but this
isn't the case with commerical routers. Commercial routers (like those from
Cisco, for example) are supported for many years, and generally get new
features at the same frequency as other new routers (unless the router is so
old that it doesn't have sufficient memory to store the new firmware).

To get these new features moving forward, you have to pay for it.

Security updates are free, but are also a "manual process" (i.e. more time
consuming for the customer).

My point is: There is a model today, at many companies, to add and update
features moving forward. But I know of no such company that does this for
free (other than Slim Devices, that is).

> Open source can co-exist with a financial
> structure, mostly in the form of donations at this
> point.

As I said, I like the model that Slim users now.

The fundamental problem with donations: It is difficult to predict what
donations will come in. This, I think, would make it difficult to justify
the expense of additional staff to fix stuff - you never know if future
donations would be able to fund new staffing levels. I suppose you could
hire short-term contractors with some of those funds, but then the company
may develop "spotty" service (i.e. sometimes bugs are fixed quickly,
sometimes not, depending on contractors - and funds - available at the
time).

For open source projects that take donations, they generally use donations
for things like new hardware and stuff. I don't think the donations that
come in are all that great (in my honest opinion). I certainly don't think
they'd fund employees (which, with salary, benefits and the like, cost a
whole lot of money).

-- Jeff

Listener
2006-01-19, 12:42
I've got my own list of enhancements for browsing/searching classical music (enhancement requests #2696-2701) but I think SlimDevices has to solve some overall problems:

1. Someone at SlimDevices needs to take full responsibility for making Slimserver a stable, robust, fully tested and fully documented software application. There is a bias toward adding features well outside the core function of playing music. There doesn't seem to be a commitment to adequate testing before release. That needs to be done ASAP rather than in some future release that involves adding functionality. SlimDevices needs to hire someone or make a current employee responsible. I don't mean a low-level QA person; what's needed is a project leader.

2. Slim Devices needs to hire someone to do serious work on Slimserver's UI design, functionality and external UIs. The current user base is heavily weighted toward techies who don't see the need for good UI design. They aren't likely to provide the insight or the implementation to make SB/SS useful to a wider audience.

3. Slim Devices needs to hire someone to provide thorough user level documentation. Slimserver server isn't carefully designed to be as transparent as a modern application like iTunes. Software like that needs more documentation than a relatively transparent application but the current documentation does not explain the concepts or provide step-by-step instructions for doing common tasks. Imporoving the documentation should be coordinated with the design work.

4. Slim Devices has to commit to providing stable, clearly documented external intewrfaces. And they need to document the SS source code. You won't get quality, fully implemented and fully tested extensions unless Slim Devices does its part.

5. SlimDevices needs to hire someone who can implement fast, reliable, flexible database components for Slimserver. The database of tag information is important for external UI software like Moose and TelCanto. The database is also the basis for improvements to the SS web UI. The current database looks like a dump of internal Perl objects and is not very flexible. And there seem to be problems about its permanence and the efficiency and reliability of the scanning process.

Bill

Mark Lanctot
2006-01-19, 12:55
Jeff Coffler wrote:
> From: "Mark Lanctot":
>
>> For regular consumer products, product updates and
>> bugfixes ARE free. Think of router firmware, for
>> example. No one charges for that. But on the
>> other hand, these aren't updated forever (a few
>> years at most), and few new features are added,
>> with new features being added to the next piece of
>> hardware and often finalized before release.
>
> "Low-end" routers might have free updates (and few
new features), but
> this isn't the case with commerical routers.
Commercial routers (like
> those from Cisco, for example) are supported for
many years, and
> generally get new features at the same frequency as
other new routers
> (unless the router is so old that it doesn't have
sufficient memory to
> store the new firmware).
>
> To get these new features moving forward, you have
to pay for it.
>
> Security updates are free, but are also a "manual
process" (i.e. more
> time consuming for the customer).

Ah, I've never worked with a Cisco router but I
see your point.

>> Open source can co-exist with a financial
>> structure, mostly in the form of donations at this
>> point.
>
> As I said, I like the model that Slim users now.
>
> The fundamental problem with donations: It is
difficult to predict what
> donations will come in. This, I think, would make
it difficult to
> justify the expense of additional staff to fix stuff
- you never know if
> future donations would be able to fund new staffing
levels. I suppose
> you could hire short-term contractors with some of
those funds, but then
> the company may develop "spotty" service (i.e.
sometimes bugs are fixed
> quickly, sometimes not, depending on contractors -
and funds - available
> at the time).
>
> For open source projects that take donations, they
generally use
> donations for things like new hardware and stuff. I
don't think the
> donations that come in are all that great (in my
honest opinion). I
> certainly don't think they'd fund employees (which,
with salary,
> benefits and the like, cost a whole lot of money).
>

Agreed.

I just don't want to see a mandatory recurring
payment scheme for all users. That would kill my
interest in the device and I'd probably sell both
of mine if that happened.

It's a bit of a vicious circle. As the Squeezebox
gets more and more popular, more and more people
are requesting more and more esoteric, specialized
features. The software is being changed to
accommodate some of this, but the cycle is
accelerating and the enhancement requests are
piling up.

As more users are battering the software from all
sides and as the software gets more and more
sophisticated due to the new features and
enhancements, more bugs are starting to appear.
Don't get me wrong, Dan's doing a fantastic job
and often doesn't get the credit he deserves, but
it's getting to be a lot for one man to handle and
it will only get worse.

--
___________________________________


Mark Lanctot
___________________________________

Kyle
2006-01-19, 13:11
I'll gladly pay $50 to anyone who will come to my house and get my artwork issues worked out.

Robin Bowes
2006-01-19, 13:33
Kyle said the following on 01/19/2006 08:11 PM:
> I'll gladly pay $50 to anyone who will come to my house and get my
> artwork issues worked out.

I'll do it remotely, if possible.

IM me:

AIM: robinbowes
Yahoo: robin_bowes
MSN: robin-.net <at> robinbowes <dot> com

R.

Kyle
2006-01-19, 14:02
I'll do it remotely, if possible.


See this thread.: http://forums.slimdevices.com/showthread.php?t=20204. I welcome any suggestions.

pfarrell
2006-01-19, 14:45
Listener wrote:
> 1. Someone at SlimDevices needs to take full responsibility for making
> Slimserver a stable, robust, fully tested and fully documented
> software application.

You do realize that almost all software engineers are only interested
in cool new stuff and hate boring stuff like documentation and
regression testing. In general, the good ones are rare and expensive.
The folks you can get are usually not good enough to do it well.

This is a hard problem, not just for Slim.

Its kinda like trying to get good developers for a Cobol
accounting system. There is a need, everyone sees it, but
managing it takes a huge amount of effort.

> 4. Slim Devices has to commit to providing stable, clearly documented
> external intewrfaces. And they need to document the SS source code.
> You won't get quality, fully implemented and fully tested extensions
> unless Slim Devices does its part.

Seems to me that Slim is mostly a hardware company. Focus is
critical for a small company. I don't think you are likely to
find that open source community and corporate standard source code
commenting is compatible.

More importantly, I'm not sure what you hope to get with this.
Strikes me as sort of a list nailed to the cathedral door.
I'm all for freedom of expression, but I like 99% of what
my slimserver has delivered for more than two years now.
I'm not at all sure I'd be happy if Slim did what you suggested
if the result was raising the cost of the SqueezeBox from $300 to
say $600. or worse, up in the price of packaged music server
hardware, which start at $1000 and go up fast.


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

Robin Bowes
2006-01-19, 15:38
Kyle said the following on 01/19/2006 09:02 PM:
> Robin Bowes Wrote:
>
>>I'll do it remotely, if possible.
>>
>
>
> See this thread.:
> http://forums.slimdevices.com/showthread.php?t=20204. I welcome any
> suggestions.

Seen it.

Let me in remotely and I'll take a look.

No fix, no fee.

R.

Listener
2006-01-19, 16:24
You do realize that almost all software engineers are only interested
in cool new stuff and hate boring stuff like documentation and
regression testing. In general, the good ones are rare and expensive.
The folks you can get are usually not good enough to do it well.
...

Seems to me that Slim is mostly a hardware company. Focus is
critical for a small company. I don't think you are likely to
find that open source community and corporate standard source code
commenting is compatible.

More importantly, I'm not sure what you hope to get with this.
Strikes me as sort of a list nailed to the cathedral door.
I'm all for freedom of expression, but I like 99% of what
my slimserver has delivered for more than two years now.
I'm not at all sure I'd be happy if Slim did what you suggested
if the result was raising the cost of the SqueezeBox from $300 to
say $600. or worse, up in the price of packaged music server
hardware, which start at $1000 and go up fast.

[/url]

I earned a living writing software until I retired. I understand the point. However, if you want to produce good software, you need a few first-rate people who lead. I used the word "responsible" in my post to convey the need for doing what's needed as opposed to what's fun.

It may be convenient for Slim Devices to regard itself as "mostly a hardware company" but the Squeezebox is not very useful without Slimserver. They need to make a bigger investment. Getting Slimserver done without cost to Slim Devices would be a neat trick if it didn't compromise the software.

> I'm all for freedom of expression, but I like 99% of what
> my slimserver has delivered for more than two years now.

Sounds as though you would rather not hear anything negative in this forum.

I understand that these forums are dominated by people in love with the SB/SS system. People who have problems post messages in hopes of getting some improvement. I'm in that category. I expressed my opinion. Maybe my way of describing problems offends you.

Bill

kdf
2006-01-19, 17:32
1. Someone at SlimDevices needs to take full responsibility for making Slimserver a stable, robust, fully tested and fully documented software application. There is a bias toward adding features well outside the core function of playing music. There doesn't seem to be a commitment to adequate testing before release. That needs to be done ASAP rather than in some future release that involves adding functionality. SlimDevices needs to hire someone or make a current employee responsible. I don't mean a low-level QA person; what's needed is a project leader.

That would be Dean, and he's pretty good at it. The bias you see it elevated because new features are more fun for users to talk about. The bug fixes do get done, and not just by the paid employees. They just don't talk about them in thread after thread. ASAP is what you are getting, but clearly not as soon as demanded. 6.2.2 is the place for these fixes, but there is also a large pile of architectural fixing to be dealth with in 6.5. As an experienced developer, I'm sure you know the kind of time investment required.

That being said, times change, and demand increases for more features.

currently, 567 bugs. 224 of these are actual bugs, while the remaining ones are categorised as enhancements. Of the 224, 156 are currently targetted for 6.5. Mostly this is by desire to fix, as many of them (you can take a look for yourself someime) are not yet reprducable by anyone willing/able to correct the problem in the code, and a fair number are not so much bugs in the code as akward support issues that merely need to dig into exactly what configuration flaw is the root cause. These targetted and reproducable bugs will be mostly gone by the time of release, with a few stragglers that just couldn't be fixed in time. After release, you get a flood of new bugs as always. These are due most often to configurations not yet experienced as it seems that each user is unique. I've lost entire evenings trying to replicate a bug and find I get only working results. That isn't to say that there aren't some big ones, but those aren't going to be solved by adding more managers.

Your points are valid, don't get me wrong. However, this is free music software for a $300 device. i've worked for companies that make devices worth 500 times as much who can't justify that management structure to the finance department. For what its worth, I'd love to see bugzilla locked off to enhancements, the developers list closed to user requests for enhancements and a nice delays on release. However, I also know I'd go mad from the resulting furor and probably never want to see the place again.

Not that long ago, I suggest in a request thread in the dev forum to leave us alone to work. The reply was less than understanding. I know you feel that this forum is anti-criticism, but its a forum. Strong reactions are standard, both ways.

For now, what can be done is done. What can be added, is considered. Long term, the server is getting split up into functional chunks, but it will probably still be lacking documentation. :)

My idea for the solution. Two button UI: "do what I want" and "open up a one line explanation that tells me exactly what I want and how to do it". (it's a joke....unless you have an idea how to do it)

Michaelwagner
2006-01-19, 17:36
1. Someone at SlimDevices needs to take full responsibility for making Slimserver a stable, robust, fully tested and fully documented software application.
There was a job posting recently in this area. I don't know if they hired anyone yet.

Michaelwagner
2006-01-19, 17:39
Its kinda like trying to get good developers for a Cobol accounting system.
Good G-d, Pat, you must be as old as I am :-)

Listener
2006-01-19, 18:26
Thanks, KDF for taking the time to give a long and considered reply. Now get back to work! <g>

> ... However, this is free music software for a $300 device.
> i've worked for companies that make devices worth 500 times
> as much who can't justify that management structure to the
> finance department.

A lot of those devices sink without a trace in the marketplace.

> I know you feel that this forum is anti-criticism, but
> its a forum. Strong reactions are standard, both ways.

There is a difference between holding a different opinion and expressing it and trying to suppress any negative comment. Things have gotten much better since I started looking at the mailing list digest just after SB1 appeared but there are still people who immediately react to defend their loved one (SB/SS.)

Thanks again.

Bill

pfarrell
2006-01-19, 18:54
Listener wrote:
> pfarrell Wrote:
>>More importantly, I'm not sure what you hope to get with this.
>
> I earned a living writing software until I retired. I understand the
> point.

Then as a pro, I really don't understand what you hope by
posting a list of demands.


> It may be convenient for Slim Devices to regard itself as "mostly a
> hardware company" but the Squeezebox is not very useful without
> Slimserver.

I do not work for Slim, I do not speak for them. I am just a happy
customer.

But if you notice, a lot of their recent firmware support,
in addition to doing stuff that SlimServer users like me want, is to
enable to SqueezeBox to work without a PC, without a SlimServer.
SqueezeNetwork and all that.

> They need to make a bigger investment.

Perhaps, you think so. I ask where the investment is going from,
and how is a business case made for it without raising the price
of the hardware from where it is. I don't want the stuff you
demand. I would gladly like it if it was free. But we pros
know that TINSTAAFL. So where does the investment come from?


>>I'm all for freedom of expression, but I like 99% of what
>>my slimserver has delivered for more than two years now.
> Maybe my way of describing problems offends you.

You think maybe?

Developers are always welcome, check out the developers list.
See how overworked, underpaid they are. Re-read your demands
and see if you really think honey is ineffective at attracting bees.


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

kdf
2006-01-19, 20:14
Listener wrote:
> Thanks, KDF for taking the time to give a long and considered reply.
> Now get back to work! <g>
>
>
and you say you want to reduce bugs? ;)
>> ... However, this is free music software for a $300 device.
>> i've worked for companies that make devices worth 500 times
>> as much who can't justify that management structure to the
>> finance department.
>>
>
> A lot of those devices sink without a trace in the marketplace.
>
>
You'd be surprised. They tend to survive based on small numbers and
users who aren't allowed to customise much without a service technician
to do it for them. By contrast, you have companies like Nortel
Networks. Top heavy with management, procedures and documentation to
drain a small forest, and they can nearly collapse when their plans for
the market don't manage to control the market the way they hoped.

My current company, in its past would be what you would call
seat-of-the-pants engineering. We are now part of a much larger body,
very much into documenting and planning. We are to be the entrance for
their new market, and they recognise that their ways are too slow to get
caught up in the time they want to be players in a rapidly changing new
market. We realise it will be a big risk to make a product with their
kind of wide release using our old methods. Both sides are working to
find a happy middle, and we don't have the clamour of a huge number of
users adding in their pet desires in the middle of it all. Stripping
down slimserver will probably do a lot of good, but it won't happen
quickly. Let the api's that are being reworked now be the platform for
all of the features that users demand. Features drive the sales, but
reliability drives repeat sales. With a structure that currently only
supports one product at a time, features are a necessary evil. Bugs are
a huge nuisance, but it is hard to give up that cash flow when the bugs
aren't really affecting the majority.

SD now has AndyG to take the con for squeezenetwork. The postings went
out for a test engineer and more, so it should all be taking shape.

and now, I will be back to my task.
-k