Is an Open-Source Core Really That Big a Deal?

[speechles 1917]

Number 2 is what I dig about emby plugins.

 

I can add behavior to the server.

 

However for most backend server apps I will build my own API using something like OWIN on windows and create my own services.

 

I really don't see the core team selling out, to remove plugins like our friends @@Plex #Noplugins.

 

Dispite audits..

 

Speaking of plugins..

 

https://forums.roku.com/viewtopic.php?t=120117&start=15#p603318

 

See my post above on Roku forums. Read it for fun. Days after my post on Roku forums Plex kills off plugins. Coincidence or consequence? Too much, too soon? Sorry.... just found it rather peculiar. Did Roku go to them and say something? Very very likely they did. The timing is just too perfect.

Edited January 1, 2019 by speechles

[Chyron 221]

One bad onion spoils the soup.

 

The idiom I've always heard was "one bad apple spoils the bunch."

 

Users on both sides are going to say dumb shit, that is the nature of passionate users that are loyal to a project or product, they will defend and attach.

Except, as a developer, professionalism should be important. I wouldn't think as highly of Luke or ebr if they made a habit of swearing like sailors. It's not professional behavior. If you want people to respect your project, you should behave respectably.

Edited January 2, 2019 by chyron8472

[legallink 187]

Can we just let this thread die already?

[Chyron 221]

@,

 

Oh, I thought this:

 

Users on both sides are going to say dumb shit, that is the nature of passionate users that are loyal to a project or product, they will defend and attach.

Was part of the discussion about this:

 

Not just their followers... their admin seems the one with the foulest mouth and strongest opinions: https://www.reddit.com/r/jellyfin/comments/a9ftd1/jellyfin_statement_on_recent_code_changes/

 

Can't say that looking at that Reddit is improving my opinion of the project, nor its main developers... nothing personally against the gentleman trying to explain things in this thread as I'm sure at least his intentions are pure.

[Perplexed 25]

Are null_sum and null_pointer the same person, or is this just a coincidence? Not that it matters... just asking.

[Luke 37060]

Are null_sum and null_pointer the same person, or is this just a coincidence? Not that it matters... just asking.

No, different people.

[Perplexed 25]

No, different people.

 

Thanks Honestly it wouldn't have made a difference (to me), I was just wondering

[Jdiesel 1114]

Emby is really going to need to step it's API game up as @ alluded to. Emby has a history of introducing breaking changes with little to no warning to third party plugin authors. In the time I've been around I've seen many interesting plugin projects come and go which I suspect is because of the continuous changes that break how the plugins work. Nobody wants to rewrite their app and go compatibility updates every single time a new version of Emby is released. Not to mention the 2 years with out of date Swagger documentations. 

 

I'm not saying that it will not work, just that more effort has to be put into making Emby an attractive place for third party developers. Just browse the Plugins section of the forum to see all the great but abandoned projects.

[Luke 37060]

Emby is really going to need to step it's API game up as @ alluded to. Emby has a history of introducing breaking changes with little to no warning to third party plugin authors. In the time I've been around I've seen many interesting plugin projects come and go which I suspect is because of the continuous changes that break how the plugins work. Nobody wants to rewrite their app and go compatibility updates every single time a new version of Emby is released. Not to mention the 2 years with out of date Swagger documentations. 

 

I'm not saying that it will not work, just that more effort has to be put into making Emby an attractive place for third party developers. Just browse the Plugins section of the forum to see all the great but abandoned projects.

 

We are working on doing just that. Swagger works very nicely now, we've had Softworkz helping on documentation, and we've been implementing API-related feature requests that have been filed. I agree that some of our past releases were too aggressive in regards to breaking changes.

[Perplexed 25]

who the hell is null_sum? on this forum or on the internet?

you know the internet is a big place right? lots of people.

 

This reminds of that time I was traveling interstate, at the time I was living in a 1M plus city and was visiting this little town, this guy I just met said "Hey do you know Bob, he lives in your city"

 

Awwww, let's keep our tone civil shall we? And the sarcasm at bay, I know we can have a great conversation without all that  

 

The person that uses the handle "nullsum" (sorry not null_sum) is the main dev of the Jellyfin project and I just wondered if you two were related. Again, it wouldn't make any difference whatsoever, but what can I say... I'm a girl, curiosity runs in my veins. 

 

Now that we both know you're not, the conversation can carry on as before... nothing to see, move along, move along.

[Jdiesel 1114]

I agree the Swagger seems to be a good option when its working. If I can figure out how to use it anyone can. Reverse engineering using the Chrome debug console has always worked well for me too. I guess if my comment seemed overly critical it is because I remember a time not long ago where plugin developers where quite frustrated with frequent breaking changes. To be fair one could easily just target the stable released rather than trying to offer support for the beta releases. 

[Carlo 4330]

Personally what I would have seen as "perfect" would have been to keep the basic Emby framework open source with full scripts/projects for compiling.

Make all clients free to use for basic navigation and playback.

 

Then add things like GraceNote as plugins that aren't open source.  I had suggested this back on gitHub issues.

 

Mobile Sync, cloud sync, and/or any other premier functionality would be plugins that run on top of the open source version and cost money. Buy a "Premier" from Emby and get these 20 plugins for your use type thing.

 

Devs could then continue to contribute to the open source/framework or they could make their own paid plugins for additional functionality.  Of course these could be limited due to the clients being closed source, but server functionality could certainly be extended.

 

I know I was going to jump in with development on the HW transcoding stuff but it started going closed source and you couldn't build the project without jumping through hoops for NET.  That sort of turned me off to doing it.  Luckily Softworkz has been doing a bang up job on this front recently.

 

We used to have the ability to check the status of all tuners in the system including IPTV.  I loved that but it was removed.  With an open source framework I could add this back in.  There are probably a lot of things I could/would do but I'm very hesitant to even try right now.  I see the pain null_pointer is having in the forums just trying to use the APIs and know what they return (or have changed from version to version).

 

I'm not mad about the change.  I like where Emby is going functionality wise.  The difference is if I contribute or not code wise.

 

BTW, when I say "plugin" I'm using the term loosely.  You could have an open source version of a DLL and an Emby Premier Plugin that just replaces said DLLs with enhanced versions (ie GraceNote).  Anyone could compile the base code and use but wouldn't get enhanced functionality.

 

I see the real "pickle" of closed source besides things like GraceNote as the WANT of the devs to limit how many users can use the software device wise (ie. 15 premier devices).  I've said it before and will say it again, this needs to be revised as it scares many Plex users thinking of migrating.  Every device should get basic playback abilities.  Things like Live TV/DVR could be options/premier in the client.

 

The only problem with this is combating the use of Emby servers for illegal distribution of content (people selling content), but that can be handled easily in other ways.

 

Carlo

[Charlie117 56]

So, it seems Emby 4.0 stable has been released with dozens of items in the changelog, numerous notes and comments included with its release but not a single mention of Emby going closed source from this version on. Additionally, the GitHub issue that includes literally the only dev response on the subject that could even be remotely considered an official announcement, has been deleted: https://web.archive.org/web/20181212100152/https://github.com/MediaBrowser/Emby/issues/3479

 

This makes it painfully clear that the decision to go closed source was nothing more than a business decision to benefit the bottom line. The communication plan appears to be to deny that Emby was ever open source in the first place.

 

I am likely sounding like an entitled asshole, but some form of communication about this decision would have been desirable instead of ignoring it happened. Also, more information on the improved API system that is supposed to be an alternative for people that want to contribute new features is also not mentioned anywhere.

Edited January 17, 2019 by Charlie117

[Guest]

why? Why does it matter? Open-source is not automatically better.

 

I'm not a FOSS guru per se, but there are many arguments and/or ideologies for FOSS out there. The Free Software Definition being one of the more famous ideologies.

 

For some it has to do with reliability. Some don't like to rely on software that can (and will) at unannounced times close off previously freely available parts behind a paywall. Emby has a lot people with this opinion, because back when Media Browser (later Emby) was open, users started embracing it specifically for that reason, as a FOSS alternative to Plex Media Server.

 

What's wrong with a paywall? It is not necessarily bad to pay for good software. But when you have Netflix and HBO, and only occasionally use Emby, half the price of Netflix is too much, otherwise you could have stayed with Plex, which has approximately the same pricing as Emby. This is similar to the old Adobe Collections. Freelancers that only use their products 10% of time in the office were happy to pay for single releases. But since it became a mandatory subscription model, while the professionals that use it for the majority of their workday would happily subscribe, freelancers would seek reasonably priced alternatives such as Affinity products, or went FOSS using Krita and Lightworks.

 

As for the low amount of commits; I think there is clearly interest for developers to work on this type of software. Hence projects like Streama, Jellyfin, and Kodi itself. But volunteers are not motivated to commit to projects that are not (fully) available in a free/libre capacity. If they volunteer code, they are motivated by the idea that everyone can fully enjoy their work for free, including by extension the apps.

 

Personally, I use the free options of Emby. It was enough for me. I said in the past that I was willing to pay a dollar per month, which seems fair to me since for 8 dollars per month I get a player with tons of content (Netflix). However now that I was disconcerted to learn after upgrading to 4.0 hardware accelerated transcoding has too moved from free to behind the paywall, this for me is the moment where I'm on the fence. HD no longer transcodes fluently. I'm on the fence between upgrading to Plex, downgrading to Jellyfin, or hoping for a big discount for Emby Premiere. I am undecided.

 

This is my opinion, and my idea of what the (monetary) value of something is. People can disagree. But much like love, it's a bit subjective and personal.

[Spaceboy 2493]

I'm not a FOSS guru per se, but there are many arguments and/or ideologies for FOSS out there. The Free Software Definition being one of the more famous ideologies.

 

For some it has to do with reliability. Some don't like to rely on software that can (and will) at unannounced times close off previously freely available parts behind a paywall. Emby has a lot people with this opinion, because back when Media Browser (later Emby) was open, users started embracing it specifically for that reason, as a FOSS alternative to Plex Media Server.

 

What's wrong with a paywall? It is not necessarily bad to pay for good software. But when you have Netflix and HBO, and only occasionally use Emby, half the price of Netflix is too much, otherwise you could have stayed with Plex, which has approximately the same pricing as Emby. This is similar to the old Adobe Collections. Freelancers that only use their products 10% of time in the office were happy to pay for single releases. But since it became a mandatory subscription model, while the professionals that use it for the majority of their workday would happily subscribe, freelancers would seek reasonably priced alternatives such as Affinity products, or went FOSS using Krita and Lightworks.

 

As for the low amount of commits; I think there is clearly interest for developers to work on this type of software. Hence projects like Streama, Jellyfin, and Kodi itself. But volunteers are not motivated to commit to projects that are not (fully) available in a free/libre capacity. If they volunteer code, they are motivated by the idea that everyone can fully enjoy their work for free, including by extension the apps.

 

Personally, I use the free options of Emby. It was enough for me. I said in the past that I was willing to pay a dollar per month, which seems fair to me since for 8 dollars per month I get a player with tons of content (Netflix). However now that I was disconcerted to learn after upgrading to 4.0 hardware accelerated transcoding has too moved from free to behind the paywall, this for me is the moment where I'm on the fence. HD no longer transcodes fluently. I'm on the fence between upgrading to Plex, downgrading to Jellyfin, or hoping for a big discount for Emby Premiere. I am undecided.

 

This is my opinion, and my idea of what the (monetary) value of something is. People can disagree. But much like love, it's a bit subjective and personal.

open source is one thing but your issue is about cost and the software being free. I have little time for that argument

[Chyron 221]

open source is one thing but your issue is about cost and the software being free. I have little time for that argument

 

Also, if you consider the price of Emby Premiere Lifetime, technically after two years Premiere is free. So if you're going to use Emby for two years, Lifetime is the obvious choice.

 

As far as I'm concerned, Free and Open Source Software can only take a dev so far. To some degree it will always feel like a project instead of a product. There are several examples I can give of this, where a software package might be good but the FOSS aspect really shows in how slapdash or convoluted the UI is. KeePass, Audacity, and Calibre being just a few of those examples.

Edited January 17, 2019 by chyron8472

[Guest]

open source is one thing but your issue is about cost and the software being free. I have little time for that argument

 

Too late, looks like you already read it. 

 

I was answering TS' question which I believe was not a rhetorical one. Also, I think you misunderstood. Issue one was about the software being overpriced (which is subjective, just like some people canceled Netflix after the 2nd price increase, while others didn't) compared to options within the same realm. Issue two was about the argument that little to no one commits to the source code were done by 'the community', which has to do with the software not being free (as in speech).

 

 

As far as I'm concerned, Free and Open Source Software can only take a dev so far. To some degree it will always feel like a project instead of a product. There are several examples I can give of this, where a software package might be good but the FOSS aspect really shows in how slapdash or convoluted the UI is. KeePass, Audacity, and Calibre being just a few of those examples.

 

You can find success stories and bad stories on both sides of the fence. There are some great FOSS products. Blender, sponsored by people who buy/sponsor their open movies. Ardour, funded by bounties. Red Hat Enterprise Linux, funded by a commercial support model. Webkit, volunteer project later acquired by Apple. CUPS, originally funded by commercial support, later acquired by Apple. Firefox. Wordpress. GCC. Wikipedia. Git. GitLab. Popcorn Time. Etchy. Bitcoin.

 

I agree that many FOSS projects are ugly. It's because there are zero designers. So the UI is "written, not drawn". But some projects are pretty. Emby looked equally good before the server was source was closed.

 

Just my 4 cents. Feel free to make your own deliberations.

[BAlGaInTl 279]

As far as I'm concerned, Free and Open Source Software can only take a dev so far. To some degree it will always feel like a project instead of a product. There are several examples I can give of this, where a software package might be good but the FOSS aspect really shows in how slapdash or convoluted the UI is. KeePass, Audacity, and Calibre being just a few of those examples.

OMG yes.

 

I really WANT to use KeePass, but the interface is so slow and clunky. I have to have a solution that is also "spouse-proof". I'm willing to work with a crappy interface for some FOSS that I use occasionally, but I use my password manager multiple times a day. So I use Enpass so I can control where my data goes. It's closed source, but the encryption is based on a FOSS standard and encrypted before it goes to cloud storage.

[Perplexed 25]

Offtopic as to the main subject, but I use Bitwarden as my password manager. It's fully open source, looks pretty (enough) and works like a charm

Edited January 17, 2019 by Perplexed

[adamstewiegreen 147]

 

What's wrong with a paywall? It is not necessarily bad to pay for good software. But when you have Netflix and HBO, and only occasionally use Emby, half the price of Netflix is too much, otherwise you could have stayed with Plex, which has approximately the same pricing as Emby.

Except Netflix and HBO are not comparable to Emby or Plex. All they do is provide content, emby organises content and serves it to clients.

[metsuke 27]

What you are referring to here is how useful something is to you, how much value it has for you and yes that is all up to the individual but it has nothing to do with open source.

 

People often like to talk about $ value when referencing open source. They should stop, that is not what the FREE in free open source software means.

 

I am a big supporter of open source, I am also a big supporter of paid software, Free$ != Open, it is very hard to get the balance correct sometimes when you have a project that is OSS and you have a payment model.

 

I think Emby should have gone closed over a year ago, it was becoming clear the product was targeting a business model that was becoming harder to maintain with the OSS approach it had. While I am a big supporter of open source I also am a realist and sometimes for a product to survive hard decisions need to be made.

 

My hope is the Emby team can concentrate on producing the best product they can going forward so all Emby users benefit.

 

I was under the impression based on this post that the reason for closing the source was not due to the economic effects of remaining open source, but rather because modularizing code that needed to be obfuscated would take too much time and delay the 4.0 release.  Is this not the case?

 

If it is purely economics that is driving this "source-closing" mentality then I'd ask how it increases revenue, since at least in my case, I'd pay for the lifetime subscription either way.

[mlcarson 13]

I love the concept of open source but if you're going to be providing things like a TV Guide provider which cost money then it doesn't make sense.  What's more important to me is the ability to get Emby on all of the various platforms -- particularly the ones that are less popular like FreeBSD. 

[metsuke 27]

I love the concept of open source but if you're going to be providing things like a TV Guide provider which cost money then it doesn't make sense.  What's more important to me is the ability to get Emby on all of the various platforms -- particularly the ones that are less popular like FreeBSD. 

Can you explain why providing services that cost money prohibits an open source?

[Luke 37060]

It doesn't have to. It just takes time and effort to make the necessary separations and we haven't done that yet.

[metsuke 27]

It doesn't have to. It just takes time and effort to make the necessary separations and we haven't done that yet.

Thanks Luke.  I actually fully understand and accept your position on this issue.  I was only asking mlcarson about his own reasoning so as to move the community to the same understanding or realign my own.  It is important to me to identify the real pain points of open source software, but also to remove any false stigma.