PDA

View Full Version : Fork it!



tigersim
2012-09-26, 14:10
Just read "The News"... very sad, but not surprising :( Well there's not much I can add to the 47 page thread I've just read. Instead, I would like to think about constructive ways forward. A few observations:

1) The Squeezebox ecosystem and community is strong, active, knowledgeable and engaged. Yes there are gripes about bugs and quirks, but that is the same with any sufficiently advanced technology.

2) There are many installed system out there working perfectly, day in day out for their owners.

3) In many cases (particularly multi-device setups), these installed systems represent a significant investment in time, money and 'tweaking effort'. Replacing them with an alternative, no matter how good, is non-trivial. Consider the network, wireless, firewall config, audio ripping, organising your library etc etc. It's not like replacing your TV. This assumes there is even an acceptable alternative. I'm personally not convinced there is, for my purposes at least.

4) Despite the corporate gloss imparted my Logitech, the system has its roots in a small start-up / cottage-industry company. The early devices at least were fairly simple, but nicely executed.

5) The server software has always been open.

6) No SB device (to my knowledge) has ever become unsupported by the server, which to me implies the underlying client-server technology must be well proven, stable and - given that it has been in use for many years on old devices - relatively simple.


This to me suggests there is a good possibility to sustain this ecosystem just in the same way that other open source projects live on. Maybe in time there's even a possibility to come up with basic 'thin' players (think SB Receiver) based on a Raspberry Pi or whatever.

So I suppose my question is, what are the legal and technical barriers, if any, of forking the LMS development and moving away from reliance on Logitech? A new forum too, aimed at The New World (Free/Community, not UE). Possibly some kind of hum for open source developers? I don't know, it would just seem wise to me to start considering contingencies before we wake up one day and find the lights have gone out.

Thoughts?

sim

castalla
2012-09-26, 14:19
An excellent post.

pallfreeman
2012-09-26, 14:28
start considering contingencies before we wake up one day and find the lights have gone out.

Thoughts?


I'm already awake, but it's pretty dark in here. So let's take a stab in this 'ere dark: I think the fact that the devs are putting LMS into git speaks volumes.

Perhaps we should move this to the developers forum.

tigersim
2012-09-26, 14:31
Just to add... I would fully accept that such a move would be aimed at satisfying the geekier/hobbyist end of the spectrum. In a way that seems to me to have always been the Squeezebox's true calling, and maybe that's why Logitech struggled to take it to the masses. There's nothing wrong with the non-savvy masses (grits teeth... ;)) and perhaps it makes sense for Logitech to turn their attention to answering their needs because that's where the $$$ is, but that does NOT mean there's no call for something a bit more open, flexible and capable amongst those with the aptitude and willingness to pursue it.

tigersim
2012-09-26, 14:34
I'm already awake, but it's pretty dark in here. So let's take a stab in this 'ere dark: I think the fact that the devs are putting LMS into git speaks volumes.

Perhaps we should move this to the developers forum.

Into git? Interesting!

Maybe you're right, the developer forum might be a better bet. I suppose I wanted to get some feedback from technically minded but non-dev owners, just to gauge interest.

bluegaspode
2012-09-26, 14:55
So I suppose my question is, what are the legal and technical barriers, if any, of forking the LMS development and moving away from reliance on Logitech?

Without written permission by Logitech I suppose that:

- all images contained in the repository are under copyright
- binary firmware is not allowed to be redistributed
- the name "Squeezebox" is trademarked and should be removed

Not that I would expect Logitech to care if anyone forked indeed ;) ...

tigersim
2012-09-26, 15:01
Without written permission by Logitech I suppose that:

- all images contained in the repository are under copyright
- binary firmware is not allowed to be redistributed
- the name "Squeezebox" is trademarked and should be removed

Not that I would expect Logitech to care if anyone forked indeed ;) ...

Images and name can be "re-imagined"... :)
The firmware sounds like more of an issue :(

erland
2012-09-26, 15:18
This to me suggests there is a good possibility to sustain this ecosystem just in the same way that other open source projects live on. Maybe in time there's even a possibility to come up with basic 'thin' players (think SB Receiver) based on a Raspberry Pi or whatever.

So I suppose my question is, what are the legal and technical barriers, if any, of forking the LMS development and moving away from reliance on Logitech? A new forum too, aimed at The New World (Free/Community, not UE). Possibly some kind of hum for open source developers? I don't know, it would just seem wise to me to start considering contingencies before we wake up one day and find the lights have gone out.

Thoughts?

Those of us that can do it has already thought about doing it several times during the last years with decreased Logitech development resources, but we have all realized that even if Logitech only puts one or two full time developers into maintaining it, that's still better than having 2-3 spare time developers which already is fully occupied with supporting their third party plugins/applets/apps on their spare time today. At least as long as Logitech doesn't spend their time to intentionally making it harder to extend their platform or remove features we want to remain.

I'm not saying it won't happen, because I do believe something will happen eventually, I'm just saying that it's not urgent as long as Logitech still spends some resources to keep the system working.

I'm not saying that this is the case for you, but sometimes it's kind of funny that people who haven't showed that they are prepared to spend their own spare time to make other people happy without any economical compensation ask other people to do it for themselves.

Generally I don't think you or anybody else should ask other people to create an open source project and offer it for free to the world unless you have tried to do it yourself, without the experience you really don't know what you are asking people to do and what kind of effort it requires. It's one thing to offer a new geeky/interesting feature/add-on for free, it's a completely different thing to spend your spare time to support/maintain basic features year after year which people just expect to work.

My personal feeling is that for long time survival, we really need something with a business perspective to back up an effort as the one you suggest, as a spare time open source project available for free with no new non geek player hardware it's going to die over time.

The future will tell who will fill the hole Logitech has left, might be someone familiar or might be someone new, but based on what I've seen during last 1-2 years I don't expect Logitech to do it themselves. Independent who will do it, it wouldn't surprise me if the new journey in some way starts here in this community, so sticking around in the community and keep using Squeezeboxes might be a good idea even if you don't believe Logitech is focusing on the right things at the moment.

To anyone that really want to fork LMS today and have the development skills to do it, first make an interesting third party plugin and try to maintain/support it for at least 3-6 months, so you at least know what your are doing before starting to fork LMS (which is a lot bigger effort).

erland
2012-09-26, 15:43
Images and name can be "re-imagined"... :)

If someone with the necessary drawing skills is interested to spend some time drawing completely new icons/logos for whole LMS and Touch/Radio firmware, that would be a great start if some developer wants to make what you suggested happen, most developers I know (including myself) really don't know how to draw graphics. So if someone feels this would be an interesting challenge and would like to help, please let us know, or just do it and post the result in the "Developers" section of the forum.



The firmware sounds like more of an issue :(

Redistributing custom firmware for Classic/Boom/Transporter/Receiver is going to be hard, but I think it's unlikely that Logitech will remove it from their servers anytime soon and the Classic/Boom/Transporter/Receiver hasn't really got any firmware updates lately so what we have today might be good enough also for the future.

Custom firmware for Radio/Touch/Controller is less of a problem as most of it is available under BSD license which allows redistribution and even commercial usage if needed. There are some firmware parts which isn't redistributable, if I've understood correctly it's related to some drivers, some codecs and some online streaming service integrations, but re-implementing the most important parts of that is probably not a big issue for someone with the right knowledge.

Except for the already mentioned parts, there are some other parts which also isn't redistributable:
- Windows exe binaries (The compiled exe versions of LMS is only redistributable by Logitech)
- Windows Installer (Only redistributable by Logitech)
- I think MP3 libraries also might have some redistribution restriction, but I'm not completely sure
- The Logitech Android/iOS apps aren't even available as source code, but it would be possible to refer people to the existing commercial third party apps instead, so this isn't critical.

Finally, most online streaming service integrations is implemented on mysqueezebox.com which due to this isn't available as source code, so to make Rhapsody, Pandora, LastFM, Deezer, TuneIn and similar services work someone would have to get an agreement with the streaming providers and implement a plugin for it. Getting an agreement with some of these providers as an individual spare time developer won't work as some of them refuse to even start a discussion unless you are backed up by a company.

tigersim
2012-09-26, 15:52
Those of us that can do it has already thought about doing it several times during the last years with decreased Logitech development resources, but we have all realized that even if Logitech only puts one or two full time developers into maintaining it, that's still better than having 2-3 spare time developers which already is fully occupied with supporting their third party plugins/applets/apps on their spare time today. At least as long as Logitech doesn't spend their time to intentionally making it harder to extend their platform or remove features we want to remain.

I'm not saying it won't happen, because I do believe something will happen eventually, I'm just saying that it's not urgent as long as Logitech still spends some resources to keep the system working.

I'm not saying that this is the case for you, but sometimes it's kind of funny that people who haven't showed that they are prepared to spend their own spare time to make other people happy without any economical compensation ask other people to do it for themselves.

Generally I don't think you or anybody else should ask other people to create an open source project and offer it for free to the world unless you have tried to do it yourself, without the experience you really don't know what you are asking people to do and what kind of effort it requires. It's one thing to offer a new geeky/interesting feature/add-on for free, it's a completely different thing to spend your spare time to support/maintain basic features year after year which people just expect to work.

My personal feeling is that for long time survival, we really need something with a business perspective to back up an effort as the one you suggest, as a spare time open source project available for free with no new non geek player hardware it's going to die over time.

The future will tell who will fill the hole Logitech has left, might be someone familiar or might be someone new, but based on what I've seen during last 1-2 years I don't expect Logitech to do it themselves. Independent who will do it, it wouldn't surprise me if the new journey in some way starts here in this community, so sticking around in the community and keep using Squeezeboxes might be a good idea even if you don't believe Logitech is focusing on the right things at the moment.

To anyone that really want to fork LMS today and have the development skills to do it, first make an interesting third party plugin and try to maintain/support it for at least 3-6 months, so you at least know what your are doing before starting to fork LMS (which is a lot bigger effort).

Hi Erland

Thanks for your considered post - your points are very well grounded, and well made. I fully take the point about the level of commitment required to sustain any development project. I'm a developer myself, albeit working in a different arena, so I do appreciate the amount of time and effort it takes develop - and critically to support - any technical project, especially one targeting the general public. I don't personally "expect" anyone to do take it on, but there have been (approximate) parallels in other projects when the primary stakeholder has taken it down a direction which deviates from that of the established community. So I suppose my question is more one of whether, in the case of the Squeezebox system as a whole, this even exists as a theoretical possibility, from a legal and technical standpoint.

As for a new business entity taking it on, well I'm sure most people here would wish any such venture the very best, but with my objective hat on I'm not sure I can yet see a strong enough commercial case - one that would stand on pure business merits, independent of 'blind' enthusiasm for the product (I've been at the sharp end of running a tech start-up and learned some hard but valuable lessons on that front... ;)) I would love to be proved wrong however.

With that same hat on, blasphemous as it may sound, I think I would have to say that Logitech is probably doing the right thing for its shareholders. It's a business, its sole reason to exist is to sell 'product' and extract money from customers. They already have my money, job done. Now to continue that they need to widen their appeal, which almost inevitably means disappointing one niche in favour of a much larger one. Now, I say they are probably doing the right thing - as far as intent is concerned. Whether their implementation of that intent hits the mark commercially - the products, support and marketing they come up with - remains to be seen. They will be exchanging one problem (geeky niche customer base) for another (saturated, highly competitive mainstream). Whether they can make that fly will be up to them. That new direction may not be personally interesting to me, but Logitech doesn't exist for my benefit. :)

tigersim
2012-09-26, 16:16
If someone with the necessary drawing skills is interested to spend some time drawing completely new icons/logos for whole LMS and Touch/Radio firmware, that would be a great start if some developer wants to make what you suggested happen, most developers I know (including myself) really don't know how to draw graphics. So if someone feels this would be an interesting challenge and would like to help, please let us know, or just do it and post the result in the "Developers" section of the forum.


Redistributing custom firmware for Classic/Boom/Transporter/Receiver is going to be hard, but I think it's unlikely that Logitech will remove it from their servers anytime soon and the Classic/Boom/Transporter/Receiver hasn't really got any firmware updates lately so what we have today might be good enough also for the future.

Custom firmware for Radio/Touch/Controller is less of a problem as most of it is available under BSD license which allows redistribution and even commercial usage if needed. There are some firmware parts which isn't redistributable, if I've understood correctly it's related to some drivers, some codecs and some online streaming service integrations, but re-implementing the most important parts of that is probably not a big issue for someone with the right knowledge.

Except for the already mentioned parts, there are some other parts which also isn't redistributable:
- Windows exe binaries (The compiled exe versions of LMS is only redistributable by Logitech)
- Windows Installer (Only redistributable by Logitech)
- I think MP3 libraries also might have some redistribution restriction, but I'm not completely sure
- The Logitech Android/iOS apps aren't even available as source code, but it would be possible to refer people to the existing commercial third party apps instead, so this isn't critical.

Finally, most online streaming service integrations is implemented on mysqueezebox.com which due to this isn't available as source code, so to make Rhapsody, Pandora, LastFM, Deezer, TuneIn and similar services work someone would have to get an agreement with the streaming providers and implement a plugin for it. Getting an agreement with some of these providers as an individual spare time developer won't work as some of them refuse to even start a discussion unless you are backed up by a company.


Interesting that the various firmwares are under different licenses, if I've understood correctly.

I've never seen the Windows version - I didn't realise it was wrapped up as an exe. I suppose it had to be done like that just to avoid user issues. I suppose this is where the distinction between a commercial venture and a "classic" open source project comes in. In the latter case, the Windows issue (or any other issue) would only be addressed if someone with the appropriate dev skills, time and inclination has that particular itch which needs scratching (the same reason why it's not reasonable for users in any free/open-source project to 'demand' action on a particular point), whereas if you're going down the commercial angle then you address the needs which will satisfy the most customers - in which case things like a nice Windows installer (and Windows support itself) are a necessity.

I hadn't forgotten the mysqueezebox.com issue, I just didn't want to cloud things further :) The only way I can possibly see that continuing is under a commercial arrangement, for the very reasons you enumerate. That either means Logitech continuing it out of some kind of obligation or good-will, or another company taking it on. I don't know the ins and outs but I could conceive of the possibility that there could be a potential business in running that service alone, distinct from any hardware & software, and taking a cut of revenue-share / affiliation fees or whatever from the likes of Pandora et al.

erland
2012-09-26, 16:38
As for a new business entity taking it on, well I'm sure most people here would wish any such venture the very best, but with my objective hat on I'm not sure I can yet see a strong enough commercial case - one that would stand on pure business merits, independent of 'blind' enthusiasm for the product (I've been at the sharp end of running a tech start-up and learned some hard but valuable lessons on that front... ;)) I would love to be proved wrong however.

If anyone wants to fill the hole they would obviously have to find a different business model/perspective compared to what Logitech have used. Logitech has already shown that their way of doing it didn't work, that's why they shut it down.


Now, I say they are probably doing the right thing - as far as intent is concerned. Whether their implementation of that intent hits the mark commercially - the products, support and marketing they come up with - remains to be seen. They will be exchanging one problem (geeky niche customer base) for another (saturated, highly competitive mainstream). Whether they can make that fly will be up to them.

If their intent is to try to reach the masses and make the device more appealing to more users, that's definitely a move in the right direction.

However, having tried the UE Smart Radio solution, I really got to say that even though the way they have removed the need to switch between different servers is something I really love, there are other aspects that really makes it a harder and less enjoyable to use than a traditional Squeezebox even for non geeks. If your intention is to use it to listen to a song now and then or use it for continuous background internet radio listening, it might work great, but for that kind of usage I'm skeptical if people is prepared to buy a $180 device. Since they decided to include the battery it's even more expensive than the Squeezebox Radio was, which feels kind of strange if they want to target the masses.

So I'm a bit skeptical at the moment, but as long as the UE Smart Radio effort results in that they will spend some resources to also support LMS/mysqueezebox.com since the software is almost identical, my hope is that I'm wrong and the UE Smart Radio will be a great success for them.

agillis
2012-09-26, 18:19
We already have a fork that we are running as part of the VortexBox distro and we already have a player that can replace Logitech hardware. Now that all the logitech code is on Git Hub it should be really easy to do a proper fork if we want.

dasmueller
2012-09-26, 18:25
Erland while I do not use your plugins I have great respect for the efforts you have put in to accomodate peoples wishes on how the SBS/LMS system could be improved to fit needs they have expressed. Your posts in this thread I believe show a level of perspective that is not often seen these days on these forums. Kudos to you on your efforts.

My concern w the whole Squeezebox system going forward is that by attempting to appeal to more of a mass market that Logitech is shooting itself in the foot. The market they are looking at is pretty much satisfied w the Apple approach in my opinion. Maybe it's because I am not a tech/what's the latest type of person, but Logitech's marketing is abysmal. The only way one knows that these products even exist is to know someone who has them or to go searching for a solution to what you want to acheive. Apple products flood every conceivable media type and Logitech is not prepared to compete. Hopefully someone at some point will pick up the flag and produce products that are compatible w Squeezeboxes but I am not optimistic.

erland
2012-09-26, 22:58
My concern w the whole Squeezebox system going forward is that by attempting to appeal to more of a mass market that Logitech is shooting itself in the foot.

Logitech needs to do mass market, that's their strength, as a company they don't have the experience and knowledge to handle a smaller niche market and make it profitable. Another company might be able to do it but not Logitech, they have clearly showed that during the last 5 years, Logitech needs the volumes because the worldwide distribution is one of their biggest strengths.

guidof
2012-09-27, 15:09
We already have a fork that we are running as part of the VortexBox distro and we already have a player that can replace Logitech hardware. Now that all the logitech code is on Git Hub it should be really easy to do a proper fork if we want.

I'm a satisfied but naively ungeeky user of the VortexBox Appliance. I appreciate your statement (and Andrew's on the VB Forum) and I hope that you will continue to support those of us who use both SBs and VB.

However I am perplexed by your assertion that you already have a player that can replace Logitech hardware. Is VB Player not a software player that makes it possible to use LMS to produce audio at the output of a VB sound card? If so, how do you see that this can replace e.g. a SBTouch?

Regards,

Guido F.

garym
2012-09-27, 15:13
I'm a satisfied but naively ungeek user of the VortexBox Appliance. I appreciate your statement (and Andrew's on the VB Forum) and I hope that you will continue to support those of us who use both SBs and VB.

However I am perplexed by your assertion that you already have a player that can replace Logitech hardware. Is VB Player not a software player that makes it possible to use LMS to produce audio at the output of a VB sound card? If so, how do you see that this can replace e.g. a SBTouch?

Regards,

Guido F.

I'm 99% sure he means the VB Player. And you are correct, although a very nice software player, it does not take the place of a TOUCH (that can sit in a different room from the VB server/computer and be connected via WIFI or ethernet back to the server).

guidof
2012-09-27, 17:01
I'm 99% sure he means the VB Player. And you are correct, although a very nice software player, it does not take the place of a TOUCH (that can sit in a different room from the VB server/computer and be connected via WIFI or ethernet back to the server).

I was sort of hoping that he may foresee the possibility of a device that can use the VB player and replace my SBTouches when they eventually wear out. I'd be interested in his thinking . . .

Guido F.

oktup
2012-09-28, 02:09
It would be really interesting to start an open source player hardware project, if anyone had the necessary skills etc. Must be possible given the success of Raspberry Pi, Arduino, Kickstarter, 'maker' culture etc - 'print' your own Squeezebox at home even? ;) It would also bring it full circle, given how the original SliMP started...

esharpmajor
2012-09-28, 22:40
I just wanted to weigh in with my two cents here. I'm new to the SqueezeBox platform (and to this forum), having got a SB Duet after Logitech discontinued the whole SB line. However, I had been experimenting with the server and software client for a while after embarking on a project to rip all my CDs (a very eclectic mix from classical to modern music) and finding that SB was the only system available that could sensible handle everything with correct metadata. (And Erland's plugins would make this even better, when I have time to set them up) I've also done a fair amount of development work, both open source and research coding.


Those of us that can do it has already thought about doing it several times during the last years with decreased Logitech development resources, but we have all realized that even if Logitech only puts one or two full time developers into maintaining it, that's still better than having 2-3 spare time developers which already is fully occupied with supporting their third party plugins/applets/apps on their spare time today. At least as long as Logitech doesn't spend their time to intentionally making it harder to extend their platform or remove features we want to remain.

I'm not saying it won't happen, because I do believe something will happen eventually, I'm just saying that it's not urgent as long as Logitech still spends some resources to keep the system working.


Granted, although without a community and perhaps a few spare time developers keeping the system current, it's likely to continue in a downward spiral. Conversely, if the user and development communities stay alive, the chances become higher than someone with the funding and access to the necessary manufacturing and engineering expertise will consider the system worth investing in.

As I see it (and this may be old news to some), there are two possibilities for further commercial viability:
- the whole system, gets bought and/or licensed from Logitech by some other company (probably an existing, large-ish, audio company)
- the platform gains momentum as an open-source system which becomes attractive to a number of smaller companies looking to springboard their product on the market. An example of this is the VortexBox appliance. Ideally, some device manufacturers would join in and make players (this is only likely to happen if they believe the backend is not "owned" by a competitor).



My personal feeling is that for long time survival, we really need something with a business perspective to back up an effort as the one you suggest, as a spare time open source project available for free with no new non geek player hardware it's going to die over time.


Absolutely, and this is where the support of a company would be needed to really make a mass or even niche market product. Both the SB Touch and Duet products could easily be cloned by someone with electronics design experience. The basic Duet hardware is a single board computer with ethernet, wireless and a nice DAC (there heaps of reference designs out there with everything but the DAC, and the latter is normally easy to bolt on). Similarly, the SB Touch would share the same design plus touchscreen controller (it's basically a tablet in a different chassis). BUT, it's certainly not a hobby undertaking (speaking as a mechatronics engineer) as the tooling required to even make one device is hugely expensive (with BGA chips etc).



To anyone that really want to fork LMS today and have the development skills to do it, first make an interesting third party plugin and try to maintain/support it for at least 3-6 months, so you at least know what your are doing before starting to fork LMS (which is a lot bigger effort).
While the SB server code is available on Github, there's no danger of it disappearing (and mirrors of the code will likely appear all over the place). So, in that sense there is no urgency in 'forking' the code. However, working towards a "clean fork" would be a nice goal that could be approached slowly by a number of people. By this I mean, stripping all Logitech branding (SB name, Logitech logos, non-redistributable firmware) out of the server code. The two immediate benefits to this would be: better acceptance on Linux distributions and BSDs, which may ease installation by technical but non-geek types looking to build a server (as well as making it easier to distribute packages for popular NASes); secondly, it would position the system as an open (market definition of "open" - ie manufacturer-agnostic) and open-source system (even it is functionally and otherwise identical to Logitech SB).

If there is community support, I am interested in doing what I can in donating my skills (and currently very limited time, although this will change) to work on SB. I have not used the system for long enough to want to volunteer to write a plugin (so far I am happy, and Erland's plugins will fill the current gaps) - however, I do have a few other ideas (eg a nice desktop client as well as the clean fork) that I would be interested in getting involved in.