（评论）

（评论）
(comments)

原始链接: https://news.ycombinator.com/item?id=41031998

Jellyfin 团队表示，向他们的项目捐款可能会产生一些问题，例如分心、分裂以及对特定项目或个人的偏袒感。因此，他们建议不要使用捐款来扩大项目。相反，他们鼓励捐助者为他们认为重要的其他项目或事业做出贡献。 Jellyfin 团队感谢他们收到的捐赠，并计划根据基础设施、开发工具或维护的需要使用它们。此外，一些用户批评 Jellyfin 团队拒绝外部资金用于项目开发的决定，认为更大的预算可以促进雇用专门的员工并改善整体功能和支持。

Not happy about this post. Instead of asking to stop, they should do just what they suggest, without further ado - distribute the income to the ecosystem devs the way they see fit. Make it transparent, make application to these funds posssible, make feature bounties, write it besides the donation button.

Once someone stops donating, it is unlikely they will put up the effort of continually researching on which client/dev to support. This would be much better handled by Jellyfin maintainers.

Spending it on other projects is a bad idea:

- it'd be distracting, they're open source devs/maintainers not fund managers

- it could be divisive and they'd risk taking flak if they get it "wrong"

- it creates a class of "approved" clients and related projects

I think the Jellyfin team's approach is pretty reasonable.

That said, if I was them I'd probably do what the Helix devs do [0] and rather than telling people to stop donating, remind them that donations are a "tip" to the project and shouldn't be assumed to be buying you anything or paying to ramp up development (or marketing or anything else). I really appreciate that stance and their attitude that it's great to have the money and they'll spend it if and when they need it, while also encouraging people to consider donating elsewhere if they want to have a bigger impact.

[0] https://github.com/helix-editor/helix/issues/2220

Exactly. This is something we've discussed internally a LOT and this is basically my take as to why not to distribute the money elsewhere, with the added #4 of "people who donated to us, donated to us - is it really right to take their money and give it to other projects, even if we think it is?".

Yes. I'm not sure what a donation is for, if not for the recipient to spend as they see fit. And this isn't a matter of spending donor money on unrelated but deserving causes - as you say, supporting clients supports the ecosystem.

I understand the argument that deciding who gets the money and how much is nontrivial and perhaps better left up to individual donors, but I disagree that it would be dishonest or misappropriating funds if you were to do so.

It's not that black and white.

IMHO, most people donating to a software project would assume the funds are used for something related to that software project. Of course you can invent a narrative saying that the funds are now given to of the project, and now that he owns the money, he wants to donate to -- it's probably a valid legal argument in case disputes arise, but it probably isn't what the original donors had in mind...

> Spending it on other projects is a bad idea:

Disagree

> - it'd be distracting, they're open source devs/maintainers not fund managers

True. Dedicating time & effort to this would take away time & effort from development. But the harsh reality here is:

1. Client devs simply will not receive anywhere near as much funding as the drop Jellyfin are proposing/asking for.

2. It's clear from the article that Jellyfin's main devs would prefer a world where those client devs do receive such funds.

It's ultimately a trade-off: Jellyfin central dev would take a hit from taking on such a fiscal role, but the Client dev community would benefit, which sounds like it would benefit Jellyfin ultimately. Entirely up to Jellyfin's devs on whether they feel the hit would be worthwhile but I don't think it should be dismissed out of hand & it's worth suggesting.

If I donated to a specific project or organization, I would feel seriously deceived if they turned around and donated that money elsewhere. This is the correct move

Mozilla is a prime example of this.

I think many have donated to Mozilla thinking they were donating to Firefox only to later realize donations never reach Firefox.

I know this adds complexity, but one way to handle this is simply transparency, and another way is to add choice: on the donation web page, give donors a choice if they want their money to possibly be used to fund other, related projects, that might be short on funding, and then list those projects. Funding ffmpeg, for instance, seems fair since it's a dependency. They could also just make a selection matrix on the donation page, allowing people to select allowed recipients, so for instance they could allow the Android client, and disallow the Windows client.

The Mozilla case shows why it's bad to not have transparency.

And even then, redistributing money will inherently make the organization have political issues. Choosing who "deserves" it is just hard work, takes time, needs an internal political alignment and strategy where the people working on this project are just doing it for fun on their free time.

In fact their decision is pretty mature and nice : they have enough to support the project costs but they don’t want the project to become their job and that’s exactly what money flowing over a hobby does : changing it to a job.

They could donate to their dependencies that need it?

Still directly supporting the code running in the project. Balances out that people seem happier to donate to the UI/product layer than something like ffmpeg. Might be a bad example, but I’m sure some of their dependencies are underfunded.

It's still opening up a whole political mess. Which dependencies need/deserve it the most? With the backlash on this post, I cannot imagine an actual list with actual dollar amounts. No matter the list, there will be an army of upset folks, this time with actual dollar amounts to complain.

Like others said, they're hobbyists providing a service in spare time. Asking them to be money managers is not a trivial ask.

May I ask which open source projects you maintain?

It's one thing if you donate to wikipedia and it ends up at some random wikimedia project you've never heard of and a completely different thing if you donate to Jellyfin and it ends up at Jellyfin Android.

But as others have said, the hax office might not like it

It may be a bit different, but there's still plenty of potential for people to donate and not be happy that their donation is forwarded to a third party Jellyfin Android if they personally have no interest in the existence of Jellyfin Android.

(I also agree with various other reasons in other people's comments, most of all that this lets donors vote with their money on which third party projects that work on top of Jellyfin get the money, rather than Jellyfin having to decide which ones deserve what proportion of spare money.)

I'd prefer to donate to the clients I use - there are many I wouldn't use (for whatever reason) and so donations to them do not help my goals. For that matter there are causes completely unrelated to jellyfin that I also support: knowing jellyfin has enough money frees to some money to donate to those other causes.

There are more deserving projects in the world than there is money. If you don't want to manage who gets the money directly (this could be called micromanage!) then something like united way which redonates to projects that you haven't heard of is probably a good use of your money. (United way is controversial and they do not support open source but they are still a good use of your money if you want to do the most good with the least thought)

If you're trying to make a donation to a project but want to be able to specify what they can(not) do with the money, then you're not really trying to make a donation.

ALL donations to non-profit organizations are purpose-bound to some extend. That's why the relevant organizational structures that allow for tax-exempt donations usually legally bind the origanization to using that money to further it's stated mission and not blow it all on hookers for the board or whatever other pet projects the organization's leadership wants to further. This is how donations are SUPPOSED to work: contribute money towards a COMMON goal, not towards someone else's goal. Donations are not free money. It is a real problem when organizations become so broad you don't really know what you are donating towards. Often such organizations end up using deceptive tactics where they use one of their popular projects (e.g. Wikipedia) to drum up funds that they intend to use for unrelated projects that are not advertised to the donor. The best solution is to keep the scope of the organization tight and tell donors to donated to other organizations as well based on their own preferences. This seems to be what Jellyfin has chosen to do and it's the right choice.

Steve has a car. The car is yellow. All yellow cars are driven by Steve. (Except not.)

You have posted an observation about non-profit organizations and described a defining characteristic of donations that go to them. You have not given a defining characteristic of donations.

> This is how donations are SUPPOSED to work: contribute money towards a COMMON goal

You mean like shoring up the Jellyfin ecosystem?

> Donations are not free money.

Donations are free money. (When the donation is money; donated clothes are obviously not free money—they're free clothes.)

There's a whiff of a suggestion with this comment that makes it seem like the intent is to frame it as a contradiction of the comment it's a reply to, but it isn't.

Jellyfin of course spends the money on ingredients that go into the project: “if you do want to help us cover some operating expenses like our VPS hosting, domains, developer licenses, metadata API keys, and other incidental expenses, check out our OpenCollective page to donate.” https://jellyfin.org/contribute/

Why would anyone feel deceived?

I instead like what they did: they were transparent about their finances and told their supporters that there are others that would benefit more from their support at this time. They were under no obligation of doing this, but it likely felt them like the right thing to do.

Yeah, they could have done what e.g. the Wikimedia foundation does (find ways to spend the money and keep asking for more), but they didn't. Saying "we're good, we have all the money we need for the time being" is very refreshing...

> the Wikimedia foundation does (find ways to spend the money and keep asking for more),

I remember reading an article about that, how Jimmy Wales was driving super expensive sportscars around and such - it made me pretty much resolve to never donate to them. Ever.

Not withstanding the super aggressive please for cash every September or whenever they do it.

> I remember reading an article about that, how Jimmy Wales was driving super expensive sportscars around and such - it made me pretty much resolve to never donate to them. Ever.

What is the connection between Jimmy Wales’ personal spending choices and your decision to never donate to the Wikimedia Foundation? As far as I can tell, Wales is not paid by the Wikimedia Foundation.

You don't see a problem with someone leading a 1%er lifestyle asking for donations from people substantially less well off while being misleading about what those donations are used for?

I don't understand your viewpoint. Is it specifically about Wales? Are you really that much against driving? Or against expensive sports cars?

Like, David Heinemeier Hansson drives super expensive sportscars - does that mean you don't use anything he works on?

I checked the Form 990s back to 2004 at https://projects.propublica.org/nonprofits/organizations/200... . Wales has never received compensation from the Wikimedia Foundation.

Is your issue that he makes his personal wealth from some other source, and doesn't transfer enough of it to Wikipedia? If so, do you have the same views for board members of other non-profits?

> Instead of asking to stop, they should do just what they suggest, without further ado - distribute the income to the ecosystem devs the way they see fit

Probably not. The amount of money they are taking in is very small, and if they start to ramp up revenue, then they actually need to start paying people for non-core work, like lawyers and accountants. Assuming that the money is received in Canada, by an Ontario registered non-profit, then it is really easy to do the accounting work for small dollar amounts (I ran several Manitoba and BC based non-profits previously for conferences and community work). If you start paying people who can't invoice you, then you need to sort out cross-jurisdictional payroll, and if you are making alot of purchases through other services.

Comments like this are unhelpful because it assumes that the developer of the project has a desire to run a business (and non-profits are businesses) rather than building and shipping an open source tool that scratches an itch. It's cool to speculate on what could be, but if you think that is the right approach, fork it, run the business transparently, and allocate a portion of revenue to be held in reserve the moment the core team for the project you forked asks for it.

>Not happy about this post. Instead of asking to stop, they should do just what they suggest, without further ado - distribute the income to the ecosystem devs the way they see fit. Make it transparent, make application to these funds possible, make feature bounties, write it besides the donation button.

Additional admin that will lead to more costs in a self-fulfilling cycle. This is a bad idea. People who are donating are already doing so with intent. Trying to gues where their money would be best placed is playing unnecessary financial games. They're a software developer, not a pension fund manager.

>Once someone stops donating, it is unlikely they will put up the effort of continually researching on which client/dev to support. This would be much better handled by Jellyfin maintainers.

They're already using the client just as they're using Jellyfin, how is this any more effort than what they've put into donating to Jellyfin directly?

I agree, it would be nice to make it easy to donate, and then the donations are spread to others within the ecosystem. But perhaps they're worried about a ruckus when they choose to make a donation someone doesn't agree with, so they'd need to have a layer of bureaucracy that they don't want to manage.

I don't think that it is the job of Jellyfish but it would be nice to have a foundation dedicated to do that for all projects.

Especially in Europe where you can get a nice tax deduction but only if you do the donation to a well (and complexly) registered non profit

No. Tax deduction effectively excludes middlemen if they don't exclusively distribute to tax-deductible purposes. And at least in Germany, the hurdle is high, as is the risk of being retroactively declared non-deductible, after which you need to pay all the retroactive taxes, usually bankrupting your org and even the middle-men.

Footing the legal costs of making sure that doing such complies with the open collective policies, the policies of the group OC is delegating management to, their prior statements, etc, would easily take up such a small amount, leaving nothing for no one.

They can't do it retroactively for already received donations (not ethically at least, I don't think it would be illegal in this situation), but it wouldn't at all be a problem for them to announce "donations from today on will be used in this new way" instead of making the announcement they just made.

(I personally think they made the right choice, am just responding to your comment disagreeing that it would be a transparency issue if they changed how things worked moving forwards.)

What you suggest is closer to fraud to me than the right thing to do. If I donate to a project, it's not for the maintainer to choose to to use the money for something else.

They should probably do both.

Make an announcement like this, but add that beyond x years of runway the remaining donations will be funneled to the community projects.

How do you know he's not already donating the extra?

If I was running a popular open-source project, I wouldn't make it public that I've got extra money that I'm giving away. That comes with all sort of complications...

If Jellyfin operated as a 501c3 charity, the IRS would have standing to ensure that funds were used in accordance with those requirements.

I couldn’t find evidence that they do. If you donate money to me (not a 501c3) because you like something I do and want me to keep doing it, I can do anything (legal) with that money without IRS comment. (However, your donation to me is not deductible to you.)

But then you have to tax the donation as your income; after that, you can redistribute and the receiving party has to tax it again... that's what charities are shielded against.

By donating directly, there's one income tax layer removed.

If it's able to be treated as a gift, it is not taxable to the recipient. I an not a tax professional, but I would expect most of what people call "donations" to Jellyfin would be considered gifts by the tax court.

There are many different options to register jellyfin in the US, and many other options to register in other countries. Each as their own set of rules. If Jellyfin isn't registered in any way then the law gets even more complex (if you are registered anywhere other countries will generally recognize that and accept the laws of where you registered apply, but if not registered nobody knows but the lawyers can spend millions fighting it out)

I'm not clear what exactly you mean. However every country has their own equivalent of the IRS with their own rules. I'm not going to look up the legal structure of Jellyfin (boring...), but whatever it is there are legal rules as to what they can do with any money they get.

> distribute the income to the ecosystem devs the way they see fit

It sounds like they have a general policy of "no paid dev". Further, it's possible that part of the motivation is the belief that if they did have it, it might lead to negative consequences (fewer contributions from non paid devs, squabbles over compensation, generally having to manage payouts etc)

IMO they should probably put their thinking hat on and try to find a way of spending more money. If the community says they want $20,000 spent on some project; then spend it. Buy some ads or something. In theory a project funded by donations is best run at around break even. This is a signal from their supporters that they should be spending more.

$20000 isn't a useful amount of money. If they are getting $20k/month they can think about hiring a developer to work full time on the project, but at their current donation levels they could hire a developer for maybe 3 weeks before they run dangerously low on money.

Remember, Jellyfin is a fork of Emby which was a fork of Plex - both those projects were open source and then went paid. After being burnt twice like that I think it is understandable why many Jellyfin developers have had enough of that and will reject anything that even slightly could be seen as going paid. So while hiring someone might be possible and in the best interest of Jellyfin in a different world, they cannot do that in this world without losing support.

That is assuming people are donating money to change the project rather than support the project as it already is.

The money doesn't have to be spent just because it's there. Personally I'd be a little annoyed if I donted to a project and the only change they made was to burn the cash on advertising that they otherwise didn't do.

A dedicated forum, open to read without login, and controlled by the project. They have easy access to all their data, decision-making power on features to support, and it's their call whether or not to allow advertising (they don't) or AI training of their data.

Seems like a perfectly good place for it to me.

We definitely need more forums. And with Discourse being a huge improvement over BB and other forum software from the 90s, it's not hard to set up multiple forum accounts and even tie them to some other login if you want.

I recently started using Jellyfin for friday movie nights with me and a few friends who no longer live near each other. The sync feature on the web UI works surprisingly well, we hang out on VC and it's as close to watching something physically together it can be. Overall I've found it to be really solid, although I've not really dabbled much with other media center software to compare it to.

My only real complaint is that for whatever reason it really does not like my folder structure - most of my files work but it'll randomly decide that a bunch of episodes in a folder are a single "file" with multiple "versions". Reading their docs it seems like they really want you to conform to a specific folder structure, but not only would this take me forever (I've been growing this collection for 15+ years now!), I just don't want to change it; I'm happy with my folder layout and it makes sense to me, it's really surprising that Jellyfin can't just show me the raw files.

I suppose you could change the structure rather fast with a few well-placed batch operations in a shell, though I also understand why you wouldn't want it.

Regarding the randomly merged episodes, perhaps the culprit isn't the folder structure or file name patterns, but metadata on the files themselves? I never had this particular situation, but I wasted my fair share of life dealing with assumptions music players make about ID3 tags, and how they're routinely broken by files sourced from random places on the Internet.

> I wasted my fair share of life dealing with assumptions music players make about ID3 tags, and how they're routinely broken by files sourced from random places on the Internet.

Musicbrainz Picard is a life-saver. I don't add any audio to my collection without putting it through Picard first.

I haven't tried implementing it, but my idea is to write a script that automatically creates a "correctly" organized directory tree, populated with symlinks to the arbitrarily-located real files.

Manage your stuff with tinymediamanager. It will auto rename files folders, download nfos thumbnails, covers.

TV shows just need s01e01

Then import the folders and dont allow jellyfin to fetch online data.

The *arr stack can be used to rename files in a sane way

A good tip is to include the imdb or tmdb ID in the file name for movies and in the directory for TV/Anime, it'll make things unambiguous

> The sync feature on the web UI works surprisingly well.

Hmm ? I've been having issues with SyncPlay since forever. Media stuck for some people, stuck at loading. Playing until some point, then freezing.

Basically whenever I'm doing something with SyncPlay it's "press play and if it does actually start never press pause"

But that's literally my only issue with Jellyfin, I've been using it for the past year and it's awesome.

Yes, I've tried separating them out, I've tried putting them all together, I even tried putting it as a photo library because apparently that was supposed to be closer to a raw folder layout, but the issue still persisted.

I think there's a kind of disconnect between the kind of users that use these media libraries - a lot of people seem to really value the metadata aspect of it, how it collects all the info the IMDB or whatever and tries to sort and match stuff.

And then there's people like me who just want a server that can handle any file I throw at it and play it over the internet, I'm not too fussed about the other aspects. Part of me thinks that maybe I'm using the wrong tool for the job, but Jellyfin does otherwise work really well for what I want, folder gripes aside.

> And then there's people like me who just want a server that can handle any file I throw at it and play it over the internet

I just use nginx (with directory listing enabled), let's encrypt and HTTP basic auth. That basically gives you what you're asking for. But it won't do the fancy web sync so that you can watch it with your friends.

That's part of why I now have multiple Shield TV boxes in my home, mostly using Kodi over SMB/CIFS shares to my NAS. It's not as friendly for outside the home, but it's nice internally.

I've been thinking I'd like to expand to start capturing the YouTube channels I watch most and dump into series directories for them. I've gotten tired of trying to work around what YouTube seems fit to shove in my direction. I really wish there was a "don't show me content from this channel" option, as if you click on any bait, you keep getting that channel for weeks after under suggestions.

Jellyfin doesn't even work quite right if you try to follow their documentation to the letter in how to structure and name your directories. I hope they improve that aspect, because it can be time consuming and irritating to fix when it doesn't work right. It largely works though.

Does Jellyfin allow using external metadata agents and scanners like Plex? I basically couldn't use Plex at all if I didn't have HAMA and ASS installed for scanning anime files.

That's not really the same thing though. With Plex, you can customise the code it runs to scan the filenames and you don't need to create any additional files.

There are multiple approaches to curating metadata. Maintaining .nfo masters | backups is handy when migrating between media managers or finetuning descriptions not in line with online data DB's.

So if you give it "[GroupName] Name of the Series - Electric Boogaloo - 03 (1080p).mkv", it understands that it's season one, episode one of a show called "Name of the Series - Electric Boogaloo"? Because Plex in its default state would rather you using the Name.of.the.Series.Electric.Boogaloo.S01E03 formatting, which would entail renaming basically everything. That's why having ASS as the scanner to avoid renaming is essential for me.

I'm not certain about that specific example. I think it should work if it's, say, a single season show. Otherwise if you put it under Show\Season 1\, I think it should work too. I have a range of different naming schemes and Jellyfin is able to recognize all of them. I've only had to put in manual effort for extras like interviews and trailers, which can be put in an "extras" folder in a show/movie directory next to the actual video file(s).

    Some very-well-requested features/clients have gotten no traction at all, with no one coming in willing to start/help developing them. We've had to abandon some (like Chromecast for a while, though it's getting new life in the last few weeks) because of this.

    We're aware that probably the biggest complaints about Jellyfin are about the lack of client support, and the rough edges/lack of polish. We do hear you. We do want this to improve this just as much as you do.

    But we need people to help us do so. We need more volunteers who can help make the code better, write new code, document, and generally improve things. We need your help to push past what I call the Development Bystander Problem, get some new blood into the project, and especially, help to make it better!

    https://jellyfin.org/posts/a-call-for-developers/

It's noble to want to be a 100% volunteer force but it's frustrating that they know they have issues and a big pot of money but won't solve them.

Even without paying for development, money could be spent to improve the developer experience and attract new devs.

They have acknowledged client development is an issue in the OP and the link above. Could they not support client devs with hardware, licences, costs, etc...

Why should distributing surplus funds fall on them, though? That's work, and it's not fun.

Setting up a wider Jellyfin-ecosystem-donation-fund (or your favorite FLOSS project) is something anyone can do. Maybe someone reading this comment will get the motivation to step up? (And no, we don't need another platform, just plain honest human volunteering)

Not immoral, but it's a huge HR headache. Once people get paid, you have to deal with what people are worth. And everyone thinks everyone (including themselves) is worth different amounts. Coming to a consensus about that is a huge hassle, and will still end up with hurt feelings.

Bad moral is bad for code.

If you have enough money, you can swallow that pill and push through. If you don't, it will make things worse.

They have enough money for what they do, but they haven't got nearly enough income (donations) to pay everyone market wages.

I find it hard to believe receiving nothing is better for moral than receiving something.

They acknowledge they have issues they can't fix due to a manpower shortage, they also acknowledge they have more money than they can spend.

Paying market wage may not be realistic, but supporting or donating something to developers who support them would go a long way to attracting additional volunteers.

Some people might be receptive to getting $100 for something they spent dozens of hours on, but then others will wonder why they didn't also get $100 for their dozens of hours recently.

Others will look at that and calculate how far under minimum wage it was, and start to question if it's worth their time after all.

You end up with bad feelings on every side unless you pay market wages, and even then you end up with people questioning if they just took on another full-time job, instead of a hobby.

It's really, really hard to add money to the equation and not make things worse.

> I find it hard to believe receiving nothing is better for moral than receiving something.

Check the ultimatum game for proof that people are definitely able to prefer no money at all.

I’m not saying this is what is happening here, but money doesn’t come alone, and overall it might be worse than no money. My point is, money is not everything - sounds really old, but it’s surprisingly true still.

I'm not sure I agree.

I mean, a silicon valley developer getting paid $200k/year might fix a bug in some open source software for free because it was bothering him and he wanted to give something back to the project.

But if I offer that same developer $100 to bail on date night with his wife to fix a bug for me? That's not an offer I'd expect him to take me up on.

I broadly agree with your point - that in many situations people feel happier and more willing to donate their time than to be paid but paid less than they feel their time is worth commercially.

But you made the point badly, confusing things by adding in the date night - so that instead of comparing "fix bug for free" vs "fix bug and be paid, but much less than their usual hourly rate", you instead compared "fix bug for free at a time that suits them" vs "fix bug at an inconvenient time and get paid less than their usual hourly rate".

I can't actually work out why you bothered to bring bailing on a date night into it at all...

I mention that merely to illustrate that our hypothetical $200k developer has good things he could be doing with his limited discretionary time.

The time spent fixing a bug for $100 doesn't get magicked out of nowhere - that's time that could be spent meeting friends, doing sports, spending time with family, reading books, creating art, enjoying good food and wine, learning new things, or even sleeping!

For a similar concept expressed in a wordier way, read about Maslow's hierarchy of needs [1]. Our hypothetical developer's physiological and safety needs are fully met - and their unmet needs won't be much helped by $100. I avoided this and chose the wording I did because some of Maslow's wording like "self-actualization" and "transcendence" kinda invites confusion IMHO.

Far less confusing and questionable, I thought, to merely argue that sex is more fun than software development. But apparently not...

[1] https://en.wikipedia.org/wiki/Maslow%27s_hierarchy_of_needs

If they really have something that can't be done by their current developers, rather than just something they don't have time for yet, that might work. But it'd be a lot more than $1000 bounty, and they'd still need people to maintain it.

They're the FOSS entrant in the category dominated by Plex, which has a history of monetization strategies that upset the community. Jellyfin probably wants to stay completely clear of even the appearance of being a paid project because the main reason someone would choose them over Plex is to get away from the influence of monetary incentives.

This is exactly why.

When I (and the other original folks) decided to fork Emby to make Jellyfin, we had seen exactly what trajectory other projects in this space had been taking:

1. Start off small and FLOSS. 2. Attract a userbase. 3. Start requesting more and more money. 4. Start "paying full-time developers" or similar. 5. Add nagscreens, premium features, and the like to "increase revenue". 6. Go proprietary.

My extremely hardline stance on this has been to nip this trend in the bud right at steps 3 and 4 by taking ALL money out of the development process.

What we use donations for is a very small list, as mentioned on the OpenCollective page:

1. Paying for Infrastructure. Domains, VPSes, etc. 2. Each team member (contributors who are distinguished and invited into the org) gets a single one-time $300 USD credit for buying a client device to help them work on the project.

And that's it.

This is why I made this post, to basically say - in many more words - "hey, we have enough runway for a few years. Donate to individual people or other projects instead".

They make a 'free software media solution that puts you in control and respects your privacy' and if they're not naive, they probably suspect a decent portion of their users are playing pirated movies.

So from a certain perspective, saying "no-one gets paid" is more consistent than saying "us developers get paid, those anime studios don't".

For another thing, if you pay for the upfront development of software, it still needs ongoing support. And that support commitment comes with a funding commitment; you don't just pay $x0,000 for an eastern european developer to spend a year adding smart TV support - to keep the developer around and the feature working, you've now got to raise that money every year, forever.

Yeah. I didn’t know this before, and I’m now beginning to see why Jellyfin never comes close to satisfying me whenever I use it, when compared to Plex.

What they’re trying to do is really hard. As much great work as they’re putting into it, they’re really treading water with what they’re currently able to muster with people that are willing to volunteer their time. There’s obviously a lot of people that want to put money toward the effort.

I hope the project succeeds and think this is a good move. But I have a lifetime Plex pass and Jellyfin still seems not as good. I have both on my box, but whenever I try to use Jellyfin theres just something missing. I will keep checking back.

So glad I stopped using Plex, it worked fine but was so bloated with useless garbage and I did not feel like I could trust it with my data.

I need a plex.com account to use my own self hosted instance? Uninstalled.

This goes against my experience. I too have a Plex lifetime pass and it has been rock solid, my library lives in network mounted NAS drive, formatted with filebot. I've unpinned the default views that it comes with when you install the client and pinned just my libraries. It just works on all devices, absolutely no issues, no memory leaks, the UI is beautiful and Netflix like, which is important for my family. The central account is what makes claiming the server and sharing the library with others and the overall remote login experience very easy, this is what enables plex.tv/link functionality.

Yes it is not perfect and they made mistakes along the way but comments as such just tend to ride on the sentiment that vilifies all non OSS products as if it's something inherently bad, the "us vs them" attitude is something I see time and time again here.

> The central account is what makes claiming the server and sharing the library with others and the overall remote login experience very easy, this is what enables plex.tv/link functionality.

The fact that this is all it enables just reinforces the idea that it should be optional. All of the core functionality is possible to implement without requiring an account.

> this is what enables plex.tv/link functionality.

> The fact that this is all it enables

Not sure where you got the idea that linking is the /only/ thing an account gives you. One feature I quite enjoy is syncing watch status between servers. So if one has multiple servers they use (multiple they control or friend’s servers) their watch status can be synced. In my case I have a travel/portable Plex server and my main Plex server so it’s nice to have my watch state kept in sync.

And there are more features from having a central account, this is just one of them to disprove that linking is the only reason you need an account or that it’s the only feature of having an account.

It does match your experience, you just explained their justification. There is no reason to require a central account. Linking discovery, etc is completely possible without it.

Even if they did have the account, the unforgivable part is that you cannot use the mobile app for your family and have people have different profiles because the plex pass is tied to a profile. That’s a sleazy cash grab that drove me off of plex.

Unless they changed it, having Plex Pass on the account that manages the server “blesses” that server, unlocking all Plex Pass features for everyone using that particular server.

They have changed it.

An example of this is sunsetting Plex Sync and replacing it with Plex Downloads.

At the time they used the whole 'Sync never worked right' argument to justify why they got rid of it. In reality, Plex sync used to allow any user to download videos from any server where the server owner had plex pass.

Now with downlods, the user must have Plex Pass to download anything.

https://support.plex.tv/articles/downloads-sync-faq/

There are other examples where they have done this as well.

I used to feel the same way about Plex til they started flooding my less savvy family with ads for their own content and useless features unrelated to what they want to do. Really not impressed by that one. I realized how bad it was when I got a call about a broken movie I didn't even have.

Plex still seems slicker than Jellyfin in some ways but after that experience I'd certainly consider a switch. Offline is the only reason I still use plex, but their offline setup is pretty buggy too.

Plex is great. There's rock solid and then there's rock solid to the extent that my spouse and family members can all use it with no support calls to me. My dad was able to download movies to his iPad for a flight to NZ all on his own.

Yes, they keep adding cruft, I just ignore or hide it. That said, I didn't know I wanted Plexamp until they gave it to me, and now I love it.

If you like PlexAmp then you might be interested in Prologue [0] as well. It’s not a first-party app but it’s audiobook app that uses your Plex server as a source. So you can have a library of audiobooks that it pulls from.

I buy all my audiobooks on Audible, remove the DRM, then put them on my server so I can use Prologue instead of the default Audible app. Yes, you give up WhisperSync but I rarely use that anyway.

[0] https://prologue.audio

The Plex Hater community is easily in my top 3 most hated online communities, and has been so for at least 5 years now.

There are few environments that seem to attract the sort of discussion where every single person feels justified in writing a 12 paragraph vitriolic entitled screed drawing from the same bucket of recycled quips. To say that you actually like Plex is some sort of great offence that attracts at least a few replies telling you that you’re wrong.

I’m so incredibly confident that there’s a large contingent of people that wouldn’t care an iota that Plex phones home, if not for the fact that they’re so deep into the Plex Hate scene that they’ve learned to be up in arms about it.

I’m not a gamer but have interacted with gaming communities here and there as I’m sure we all have to varying extents. I see the Plex community as pretty similar. I think it attracts the same sort of ‘power user’.

You might be onto something. I think it's to do with the fact that self hosting attracts a certain crowd, and when you have a product that is popular to self host, but is not open source and with some paywalled features, it irritates them deeply. The narrative is popular enough to hate Plex that I suspect some people who don't even care try to farm karma from the bandwagon. Podcasts like selfhosted.show also ride these tropes...

I have never used Plex even though I've heard it's superior but I just didn't bother when I saw exactly what you said. Creating an account somewhere to use my own instance? Even though I could just block all outbound traffic and hope it'll still work - no thanks. Jellyfin turned out very fine.

I stopped using Plex almost immediately after installing it. I couldn't figure out how to proceed without creating a centralized account with them, which is the last thing I want or expect from a "self hosted" software package.

When I used it a few years ago it was still possible to use without an account if you knew the exact extremely-hidden-behind-dark-patterns steps, much like making a local account in Windows these days. But from the other comments in this thread I get the impression that even that tricky process is no longer an option.

I think this is the worst part of Plex, it's self hosted but not really. Still depends on the company's continued existence. If the company goes bust my solution will probably just stop working. It took some time to set it up properly so the bloat is thoroughly swept under the rug, never in sight. This gives me a more than decent experience and the interface/UX is spot on for me.

The best of Jellyfin is that it's truly self hosted. But no matter how much set up I do, the experience is still never too good. I don't really like the UI/UX but I could get used to it. My biggest issue is how it handles the folder structure and metadata for series. They will always have to be neatly organized in folders to be properly picked up. You can't throw an episode file here, an episode in a folder there, the rest of the season in another folder and so on. They're just seen as independent material, no metadata. This makes the watching experience very stunted.

> I need a plex.com account to use my own self hosted instance?

And anybody could log into their plex account on my instance! The way they handle accounts for self-hosted instances is either deliberately convoluted or ineptly designed.

I can give a couple examples:

1) Jellyfin supports reading NFO files to help determining matching the file with the correct metadata (https://jellyfin.org/docs/general/server/metadata/nfo/). This is arguably superior than Plex's way (you have to name your file with IMDb id or tmdb id https://support.plex.tv/articles/201018497-fix-match-match/#...), but I still constantly find media w/o metadata in Jellyfin.

2) The way you fix the matching in Jellyfin has a pretty terrible UX. You select the movie w/o metadata (there's no filter to "find all media that Jellyfin fail to match w/ metadata) and click "Identify" in the 3-dot menu, then fill in the information and search (why can't Jellyfin prefill useful metadata, at the very minimum the year that should be easily identified using regex?), then pick either IMDb or tmdb (why do I have to choose one of them? Can I bulk-switch my library to use IMDb instead of tmdb?), with a checkbox to confirm you want to "replace existing images" (why do I have any images to replace? Jellyfin did not match the media to any metadata right?) you now have some useful metadata.

3) When I need to force a rescan of a library, there's no way (that I can find) to do that when you are viewing the library. The only way to do this is to go to Settings -> Dashboard -> Scan All Libraries.

4) There's no offline media playback support (AFAIKT).

1 - I'm not sure what you mean. Jellyfin can use the NFO, and if not will do a normal search and match.

2 - I agree that Find all without set metadata in a library would be nice. However when identifying you don't need to chose one or the other, you just type whatever info you have (the name, or the year, or both, or ...) and search and it search in all providers it has a plugin for (so in your case, both imdb and tmdb) and show all results from both.

The do you want to replace image is because it's very common to switch to another provider for obscure series for exemple, but want to keep you hand replaced image. It doesn't really apply to "blockbuster" series or movies, so I guess it could be nice to have a default you can set.

3 - It separates the admin part and the viewing part, I think it's good. If you're solo user it can feel bothersome, but for me it's a better thing.

4 - You can download the files for later viewing offline.

Overall the only one I agree with is the figure out which movie or serie you didn't identify and let me fix that in bulk, that process could be improved, but it's also a one time thing at first setup.

1) I do generate NFO for all my media and Jellyfin should have enough information to match the metadata, yet it fail to match constantly.

2) When I try to identify the media, I would expect Jellyfin to pre-fill the form with the name, year etc, basically anything it can find, either from parsing the filename or the NFO. But Jellyfin did nothing here to make it easier.

4) Oh I just found this functionality in the Android client! But this is over simplified, for TV shows it doesn't support "download the next N unwatched episodes". I also tried to download a movie just now and I got a notification "download unsuccessful" w/o more insight to debug the issue (although to be fair Plex is not better in giving a reason why download failed). And I don't see in the UI to manage all downloaded media (maybe because I haven't had a successful download?).

For 1) I use folder and file naming myself rather than NFO so I can't help you more, but if it fails on such a large scale you should submit a bug with an example as it's obviously not normal behavior.

2) I understand what you mean,but I would disagree, usually it would get in the way for me (whatever it identified or thought it did was wrong).

4) The log will be on the server, admin, logs. There won't be a manage my upload though, it's a client thing and the current client does consider that outside of its scope.

Personnaly when needed it's download then use Vlc when I want to watch.

I have 20tb+ of movies, tv shows, anime, documentations

I do not allow jellyfin to scrape online. Everything goes through tinymediamanager (kodi format). There isn't a single missing piece in my library .

I've been able to identify everything by just using the title. Jellyfin could be better here, but there's almost no friction in entering the title, hitting enter, selecting the right movie and then completing it. You don't have to change anything after selecting the movie, since it by default replaces images (and simply provides the option in case you already have images you'd like to keep).

Are you advocating taking away an option other users might use because it confuses you?

Never used Plex, but I have 2 gripes with the Jellyfin iOS client.

1. I can't download shows from the iOS client. I can on my Android device

2. The music player doesn't interface with the device's media player. So you have to keep your phone open when listening to music

Jellyfin helps fit a niche between a more complicated Kodi setup and Plex’s easy setup with everything just working.

Non-techie family though prefers Plex, hands down. Looks good and really easy to setup on any machine.

Ditto. I would love to use jellyfin in principle, but in reality I find myself often butting heads with it when I just want to watch a damned film - and Plex, despite its considerable bloat, just works.

Yep. Every time I've tried Jellyfin, I thought it was great... Until a video wouldn't play, or the sound wasn't synchronised, or something else. Whereas Plex just works, and always did.

Hoping Jellyfin keeps going, maybe I'll check it out again at some point.

In my own experience (and the takeaway is “software is hard” not “I am right”), every movie I watched via plex had audio desync issues, which has never been an issue with Jellyfin.

The client I use is Apple TV 4k and Plex does not have reliable playback for that device. It’s pathetic this has been broken for so long. I have to use Infuse for playback, which is $1/mo.

Does Jellyfin work well on this platform?

Jellyfin works well except that if you’re not transcoding, you’ll need Infuse for (IIRC?) a couple not-ubiquitous-but-not-rare audio codecs that Apple TV doesn’t natively support.

I use Infuse with Jellyfin. My server’s way too weak for reliable transcoding, plus not using it avoids a large proportion of bugs in the Jellyfin issues tracker at any given time.

Using the Jellyfin webui on AppleTV without even installing an app works decent.

There’s a dedicated AppleTV app, which works great for a month, then reliably forgets its credentials. I use it, but the kids don’t want to learn how to type urls, usernames, and passwords so they don’t. That’s fair.

I have more issues when using Plex than I do with Jellyfin. Jellyfin can still be annoying, but Plex can be infuriating trying to get it play nice with my library. I've stopped using Plex.

This is my first time hearing about Jellyfin, and wow what a breath of fresh air compared to the typical hyper growth model employed by a lot of OSS projects.

Any Jellyfin users here that can vouch for it? I currently have a SMB share on a Raspberry Pi 4 and I connect to that on my Amazon Fire Stick using the VLC SMB features. It works ok but the VLC UI leaves much to be desired. Would Jellyfin be better for this? Is there a client that works on the Fire TV stick? (This one I think? https://github.com/jellyfin/jellyfin-androidtv)

I've been using Jellyfin since the beginning and it's been a total joy to use. Even though I trust the project now, in the beginning when I was migrating off of Plex I had both running simultaneously in separate VMs referencing the same read-only library. This dual config worked great, though I didn't use Plex much at all after Jellyfin quickly proved itself even in the early days of its development.

The diverse client support is awesome. In addition to streaming video content to various house devices, my favorite is my "jukebox" music listening setup consisting of a RPi 3B+ with ALLO Piano 2.1 DAC hat feeding separate speaker and subwoofer amps with desired crossover frequency. Running on the Pi is Mopidy with the Mopidy-Jellyfin extension to access Jellyfin's libraries along with the Mopidy-mowecl extension which provides a slick web front end for the DAC (you can also queue music from the Jellyfin GUI and "play to" the DAC). Highly configurable and fun to tinker with. For example, I have a USB numberpad keyboard plugged into the Pi with assigned hotkeys by way of the triggerhappy service. I love being able to keep the music playing if my desktop workstation is off or rebooting.

https://github.com/jellyfin/mopidy-jellyfin https://github.com/sapristi/mopidy-mowecl

The best thing is it's all FLOSS, so I don't have to worry about the rug being pulled out from under me!

> Any Jellyfin users here that can vouch for it?

I've been using it since the fork from Emby and it works well for what I use it for.

That said, they're right IMO to redirect funds to the clients. The Android TV app is in a really rough state. The regular Android and Web interfaces work great, Roku as far as I recall works well...but the Android TV one is really not good.

I abandoned a Shield for Apple TV over the many issues on it. Should-be-supported codecs had problems that made many videos unusable. Plus bad UI jank in general, not just in Jellyfin. Shouldn’t have tried to cheap out, it’s not like I’ve never used Android before (I’ve developed for it… including for set-top devices) so I ought to have known better.

I switched from Plex to JF after being a plex user from the very beginning. It's not as polished as plex but it has none of the bloat and works flawlessly for me. It's super easy to manage my libraries and the meta scanners work 95% of the time, very rarely needing me to manually tweak data or images for media.

Jellyfin android works fine on firetv. The only thing you might struggle with if you run the server on a pi is transcoding, especially if you have 4k media files.

I use it with a weak x86 server (used workstation off EBay) and stronger clients (Apple TV, iOS devices, browser on strong laptops) so I don’t need transcoding, since my server can’t manage real-time transcoding. Works great with that set-up, crazily better than XMBM/Kodi ever was in a dozen or so attempts, and people who aren’t me can pick it up and confidently use it, without immediately getting stuck in some weird UI mode and giving up forever, which is what Kodi always led to.

[edit] I use the download links in the browser UI and then VLC playback on an iPad for kids’ movies on long road trips. It even (with an assist from VLC) fills that use case.

I've been using Jellyfin for several months now, for me it works nicely. I did a bit of tuning over time (e.g. "OK, now I'll spend a bit of time improving how subtitles work"), and after learning how to set that aspect up, it just trucks along working. I have a handful of friends using my setup too and haven't had complaints.

Beware though that if any transcoding is needed, the RPi probably won't be up to it. Instead I shelled out a little bit for a small Intel NUC thing that can do hardware conversion.

I've been using XBMC (now Kodi) clients + a single NFS file server for 16 years, now installed on three media PCs throughout my house. Looking at Jellyfin's feature set, it's not clear what it offers over my current setup. It just seems like an apples-to-apples alternative, which is fine.

Yes it handles my library (~60tb) just fine but the Apple TV client isn't great. I run it and Plex side-by-side and switch between the two depending on which device I am using.

jellyfin handles my 12 TB media library with transparent ease. i use infuse as a client on my apple tvs, including devices at my family and partners' places via tailscale on aTV

Absolutely, it's been running on my NAS (via Docker) for over a year now with no issues. It supports hardware transcoding which is nice for downscaling streams while watching outside of the house.

Jellyfin integrates really well with other services like Radarr, Sonarr, and Jellyseer so you can request media, have it downloaded, indexed, and available automatically.

Client-wise it's a mixed bag, Infuse is probably the best one out there, but it's only available for Apple TV / iOS.

Their Jellyfin client for iOS and Apple TV was a bit flaky for a while, but last year some update fixed those issues and it's been rock solid for me. I also bought Infuse and it's a decent alternative interface.

I use Jellyfin, and it works good enough most of the times. Sometimes it just won't find subtitles, or metadata and it's highly annoying to fix it. The TV app is a bit rough sometimes. The android app feels very barebones. The HDR tone mapping is either horrible or non-existent (can't remember which, but I could never play proper HDR).

So yeah, there are a lot of rough edges, but I think this is the best on offer if you want a true self-hosted media library that mostly works out of the box across various platforms.

Very respectable. It's behaviors like this that keep me on Jellyfin instead of Plex, even in the face of social pressures to switch.

Switched from Plex to Jellyfin a few years ago and have been happy with it. Plex had issues with movies with subtitle files. It tried to constantly reencode stuff. Couldn't work it out. Jellyfin works fine.

Also, I'm not getting nagged for features that I don't want anymore. I happily paid for my account, Plex, but just let me watch my local movies in peace. Leave me alone with television streaming, free movies or whatever else was the latest thing that was being pushed.

I’m a happy user of Jellyfin, and I’d like to amplify what others have said about $24,000 or 40 months’ expenses being not nearly enough. At a conservative safe withdrawal rate, $24,000 of capital is only $60/month — in other words, a tenth of what the project needs to cover its monthly expenses.

If the project wants to be self-sufficient, they thus need ten times as much money. Then they can invest it, and cover expenses with the income from their investments.

If the project doesn’t want to be self-sufficient, that’s something else. Maybe they would prefer to think it will better align incentives for their users to keep them hungry? Personally, though, I think that self-sufficiency should be everyone’s goal.

Yes. ‘Safe withdrawal rate’ is the annual fraction of an investment which one can sell while maintaining the principal’s value, adjusted for inflation. That might sound very complex, so here’s an attempt to explain it simply.

If one has $24,000 and spend $600/month, then after 40 months there will be nothing at all left. On the other hand, if one invested that $24,000 in Treasury bonds one would have something like $26,229. So (ignoring inflation) one could spend ($24,000 - $26,229)/40 = $55.72 each month, and still have $24,000. Why would one want to do that? Because one could keep going, investing the money, collecting the interest and paying one’s expenses: that $24,000 can last forever, paying $55.72 a month.

Now, in real life one can’t ignore inflation, and that $24,000 will be worth less than $22,000 in 40 months. In real life inflation tends to outpace the risk-free rate of return one can get lending money to the U.S. Treasury. So one needs to get more return by taking one more risk, for example by investing in the stock market as a whole. But that exposes one to economic downturns.

To make a long story short (too late!) folks have run the numbers and figured that one can conservatively invest one’s money in some broad indices, withdraw about 3–4% a year and maintain the post-inflation value of one’s capital. More than that, and one runs out of capital; less than that, and the capital continues to increase, but you have less money to spend today.

If Jellyfin wants to be able to pay their $600 (in 2024 dollars) hosting bill forever, they need to invest $240,000 today. And then they’ll never need to ask for money again (assuming all sorts of things, like no decade-long recession, no world war, an asteroid doesn’t crash into the Earth and so forth). For the same reason, an American making the average wage of $63,795 needs a $2.2 million nest egg to never work again.

Yeah, but they specifically don’t want that:

> No, this doesn't violate our policy of "no paid development", because donations are just that - donations. We will still not honour bug bounties or similar, and still not use our collective finance here for paid development.

IMO, there's nothing worse than paying people too little money. Well, other than unevenly paying everyone money.

"But it's $100 a month they wouldn't have had!" Sure, but now there's something to value your time and effort against, and it will not stack up favorably.

> IMO, there's nothing worse than paying people too little money.

In a for-profit project, I'd agree. Jellyfin isn't though, and they're pretty up-front about that, so "paying too little" doesn't really apply to the people who end up contributing.

You do not want your volunteers looking at the effort they put in and the money they get back for it. It's going to create negative feelings, and at the very least performance will suffer for a while. It's more likely that it'll cause a permanent problem instead, possibly even driving away volunteers.

> You do not want your volunteers looking at the effort they put in and the money they get back for it.

Why not? I think it's good for everyone to evaluate what you're spending your time on, and if that time is time well spent.

What do you mean with "the money they get back for it"? It's a FOSS project that doesn't pay for engineering hours from volunteers or anyone. It's literally $0, and everyone contributes to Jellyfin with the expectation of getting back exactly $0.

Is there a Jelly Cloud solution? i.e a company that deploys Jellyfin to a EC2 instance and hooks up an S3 and gives you custom domain e.g: username.companyname.com and charges you a monthly fee for the S3/EC2 costs + other costs?

I was thinking of creating it but wanted to see what demand there is? Most Devs can easily do this themselves but I want to extend and sell it to other people as a service who wouldn't be capable and just happy to pay.

Plex and Emby would be competitors but not open source afaik.

There are many providers who sell seedboxes, which is exactly what you're looking for. They generally include support for Jellyfin as well as other *arr apps. I personally use ultra.cc and have been mostly satisfied with the service.

I can't speak for a generic cloud solution. but various seedboxes (managed bittorrent providers) offer Plex and also Jellyfin as part of their offering. That's how I run Plex as well as Jellyfin, personally.

Why wouldn't they redistribute to other maintainers of the plugins/clients. They have much more knowlegde of where money is needed or where resources are lacking.

Or pay devs, not in their circle to build the features the community wants but they can't (lack of expertise, or otherwise) implement?

Why streaming? I can understand when one needs some media to play on different devices at the same time, also remotely, but for individual use I found a kodi box near my TV and configured to navigate NFS/SMB directories and filenames on my NAS, to be so immediate that I can't imagine doing things differently. Also, by streaming from box A (NAS) to box B (PC) it would probably need a lot more computing power from the usually smaller and less powerful hardware, and it would also move a lot more data on the home network because the video is transcoded by the weaker hardware and before being transmitted. ...Unless streaming in this context means a completely different thing and therefore I'm wrong.

> probably need a lot more computing power from the usually smaller and less powerful hardware

Streaming does not mean it forces a transcode. I mean it can, but if you watch from a decent client it often just sends the raw media file.

And you can change from PC to phone, to tv and it remembers what you watched.

I actually use kodi as a client for jellyfin.

I mainly use Plex, but the key benefits are that it remembers what you've watched, allows for multiple accounts, allows for easily continuation of a film or series without having to remember what time/episode you were on. Also allows for easy remote viewing.

Kodi does all of this, except remote viewing that possibly could be done through external addons, but I'm not sure about that. Some features like remembering what you were watching requires it not to be turned off. Sleep is fine however, and on my mini PC it goes to sleep when it detects I turned off the TV via HDMI/CEC connection. When you wake it up you'll be in the same place and it remembers also the time watched, while if you turned it off you have to navigate back to the place you were watching, but it still keeps the time of that file and any other one, prompting you to continue from when you left or start over. Watched files and time are saved as part of the backup that can be scheduled to run automatically in the background, so one can upgrade it, then restore the backup to get back every change to the interface, addons, configuration, etc. It also runs on phones, and I installed it also on my Pixel with GrapheneOS along with some addons, but the interface is really hard to operate there.

I am sure about Kodi not being able to perform remote viewing with the same compatibility and convenience as any of the popular dedicated media streaming platforms.

I get that you like Kodi, but listen to yourself. Kodi does all of this! Except you might need a plugin that doesn't exist for the one big feature that spurred the existence and popularity of Plex, except that some features work but not reliably, except that you need to actively manage the backend, except that the phone UI sucks.

So, there's your answer?

I'm not a kodi fanboy, and I would certainly change a few things about it if I could, but I needed a client because having a NAS where everything is stored in their appropriately named own directory makes things a lot faster; all it needs is a player on something that can mount remote directories. Also, my NAS runs XigmaNAS which I doubt could stream videos, but more importantly I don't need the stream functionality because usually I'm watching something while my SO watches something different. They're just different solutions for different needs.

You were overselling it a bit, going so far as to assume a plugin existed for it when it didn't. It gave a bit of a fanboy aura to the comment.

I like Kodi too, I've been using it since the XBMP days on modded Xboxes and still do for certain things, but eventually I needed something more flexible and something that worked over the internet. Being able to share my movies with my grandma or my next door neighbors wasn't really achievable with Kodi.

Software like Jellyfin and Plex allow for streaming to more than just Kodi boxes (your phone, tablet, set top box smart TV apps, web browser) and transcoding allows the flexibility to adapt the bitrate to reliably push video over the internet and to work around compatibility issues with clients that might not support certain codecs (keeping in mind transcoding only happens when it needs to).

To be honest, it's right around a line that I feel is fine to make this request.

The thing that triggered this, that isn't really mentioned by me in the post, is that we've had a few sudden swells of donations around both the 10.9.0 release and the latest 2 Plex kerfuffles, and I get the very distinct impression that people donating are not reading all the "fine print" about what we use it for.

We also have a lot of recurring donations which should pay for a huge chunk. The goal here is to basically just make another public "hey, we never really asked for donations, and we appreciate it, but hey for now we'd rather you donate to individuals than the org since the org can coast on this money for a few more years". Which we absolutely can. I can't envision any real situation that would require us to use a huge sum of this money in one go, so it's really just the monthly recurring costs to cover which are relatively stable and consistent.

I was expecting a much bigger number than 40 months, but then again for an entirely donation supported project it's quite good. I don't think I can remember a project telling people to stop donating directly to them before, though. It's a little odd really. I get it, but seems like investing in the S&P500 and then withdrawing your money the moment you get the slightest gainz.

Don't port forward it's a pain in the ass to expose your home network to the internet. Just use something like Tailscale VPN (p2p wireguard) and buy a domain to point to the internal Tailscale IP.

“Don’t click a few buttons in your router UI to forward a port and setup a dynamic DNS client in the same way that has worked for 25 years. Instead use a bunch of other 3rd party shit that builds a VPN and tough shit if people are using a TV.”

What you’ve described is a pain in the ass from a setup perspective. I think what you were trying to say is “be careful about jellyfin vulnerabilities”, but that’s definitely not what came out.

Exposing a port to the internet is a huge pain nowadays. You never know if it is going to work, and even if it works it is incredibly flaky. What's changed is that a lot of ISPs are using double NAT or CGNAT now, which wasn't as common before. Which means router based DDNS will simply not work. On top of that every single router I've used is extremely unpredictable about respecting uPNP or whatever. So for most people a tailscale vpn or cloudflare tunnel is the best option.

It depends on the situation.

In my situation at home, port forwarding is stupid-simple and just works.

My ISP does not saddle me with CGNAT (or any other form of NAT). I don't use UPnP.

I have a real (dynamic, but just-for-me and almost never changing) IPv4 address to use, and I simply use it.

It works predictably. It works reliably. It is not even a little bit flaky. There is no voodoo involved.

And it doesn't require me to teach my elderly mother how to use Tailscale with her Roku STB.

(I recognize that others may have different situations. But the existence of different situations doesn't mean that one must declare a particular solution to be the "best", does it? KISS.)

Have you managed to get TLS working with a setup like this? I have a custom domain that isn't used but I'd like to point it to a machine that's on Tailscale. Do you just put your Tailscale DNS on public DNS servers or do you use an internal one? Do you use a reverse proxy to route port 80/443 to the port your app is running on?

You could just get a wildcard certificate with lets encrypt, via a dns challenge.

E.g. lego supports many different dns providers

https://go-acme.github.io/lego/

And then internally inside of tailscale you could have your own dns server, which serves subdomains of your domain, and for all subdomains you can use the same wildcard certificate.

This also does not 'expose' your subdomains on Certificate Transparency logs

Depends, if you only want dns and nothing more, then probably dnsmasq. That's basically one of the most used dns/dhcp servers.

Otherwise you could use solutions like AdGuard Home or PiHole, which both have a Web Interface for configuration, and the ability to block ads and tracking domains.

Note that I don't use Tailscale myself, so I don't know if Tailscale 'needs' something else. But I use pure wireguard, and all of the services mentioned above work with 'pure wireguard'.

I use Google as my SSO provider, all of my personal devices are under my own email. For friends, I just made a throwaway Gmail account which I give out the username and password for so they can connect their computers to the tailnet.

I must be a networking genius because I run about 30 websites in dockers behind a reverse proxy(one static IP) with gazillions of port forwards and static routes over a consumer router and I have no issues.

Look, if you're serious about this, set up a proper DMZ. Segment your network, throw in some DNS magic with internal and external zones, and slap a reverse proxy in there for good measure. That way, you can have your cake and eat it too - internal and external access without compromising security.

Is Jellyfin an all in one solution for media management and streaming?

I normally just use transmission and serviio (vastly superior to Plex IMO), and only recently found out about sonarr, radarr, etc.

Really though, is there not any 'all in one' solution, instead of all these various programs chained together? I get keeping the serviio bit separate, and maybe the torrent client, but the rest could easily be integrated into one app.

I always had Jellyfin as second option with my Plex Lifetime membership. Glad they improved a lot since the first days on their reddit community.

Now I prefer using Stremio+Read Debrid, Kodi as my main player for local movies.

I have an android TV running Kodi. Kodi connects via SMB to my NAS (Odroid HC-4, OpenMediaVault) and can easily show all my content.

Would Jellyfin improve my experience? I could run it on odroid without transcoding, and then use the official android-tv client. Would it be worth the effort of switching?

Depends on how much you value your Kodi setup and how many "difficult to auto match" shows | movies you have.

I wanted to run Kodi | JellyFin in parallel (two Kodi client Android TV's in house) so that visitors with iPhones | Androids | Tables could browse content.

The local Kodi TV clients used the NAS box MariDB to hold meta data held in the local to content .NFO files (show ID's, descriptions, etc)

I've held off on that as the JellyFin docs are not clear and the small experiments I tried in isolation proved tricky to resolve issues.

The issues being getting JellyFin to load details from local.nfo files w/out making changes and|or getting JellyFin to not make additional .INFO files (more or less duplicates) and|or getting JellyFin to not trash existing artwork with new downloads.

I'm sure there are solutions here (eg: JellyFin can use seperate directories for it's meta-data (but then doesn't read the local .nfo files to get ambiguous shows 'right')), I've yet to put the hard hours in and these are the kinds of hiccups you may face.

It's moot if you have a small collection | don't care about existing artwork | etc.

IIRC you can disable transcoding on fly by server and trust the network | viewing client to handle full original content, etc.

I like JellyFin and used Emby in the past and use small JellyFin setups today - I just haven't resolved the issues in pointing it at a large curated collection and letting it rip making it's own parallel meta data w/out affecting pre-existing meta data.

As someone else mentioned, the benefit of Jellyfin is it updates with newly added media quite consistently. I previously was a heavy Kodi user, on a small Android TV set top box. I now use Jellyfin exclusively, without transcoding as you mentioned.

Another benefit I've found is it's much easier to add new subtitle to a show/movie, without needing to setup specific subtitle databases as I did with Kodi.

The only issue of note is the occasional crash, but given the level of polish with Jellyfin, and zero cost, it feels absurd to mention it. Oh, and if you have multiple household members, checkout Jellyseerr [0]. It lets others easily request shows/movies (which you can get notified about and approve) without bugging you.

[0] https://github.com/Fallenbagel/jellyseerr

I'm not a Jellyfin user, but I use Emby, which has the same ancestry and similar features.

I can connect to my media server from literally anywhere in the world, using basically any device. I have a Chromecast with Google TV that I pack with me when we go away, and I pop it in the TV, configure wifi and I have my entire media collection right there. My phone, laptop, tablet and the wife's devices all have the client installed & configured.

I have friends and family that I share my media collection with and they are on the other side of the planet. All I give them is a URL, username+password and they are off to the races.

If this is not your use case, then Kodi with SMB is probably more than good enough.

Single-client is IMO where Kodi shines. This way you also have less complexity as Kodi doesn’t need a ton of things that are integrated in Jellyfin, I’d not switch if you don’t have any pain points with this set up.

There is a Jellyfin add-on for Kodi which you could use if you want to keep using Kodi for other stuff like streaming. That's what I do (Kodi runs on a Raspberry Pi connected to the TV) and it works fine, though honestly I'm not a particularly heavy or advanced user so my needs are low. Another advantage of that setup is that I can use the Jellyfin clients on my laptop while using Kodi for the TV.

I had the same setup until recently. I always had troubles with Kodi not indecing very well my movies; and also when wartching my moview far from home I had to copy them first, than think about marking them as "watched" in my library - and if I stopped watching mi-movie that was even more complicated.

Really, Jellyfin has been quite the upgrade. The indexing is better, I can watch things far from home, everything is there. As a bonus, I can create accounts for close friends/family so they can watch what I download without hassle.

Just try it.

I had this setup with a Netgear NAS (old and slow). I switched to a better NAS and Jellyfin and could not be happier. Well, I could be happier because Jellyfin has some quirks, but I'm pretty happy.

The main difference is that it works everywhere, from any device on the LAN (browser, iPad, Android phone, you name it), and even remotely (using Tailscale, to avoid exposing Jellyfin to the world). It's like you have your own little Netflix that you can access from absolutely anywhere.

（评论） (comments)

（评论）
(comments)