I've been donating to Gitea for years (on their Open Collective [1]) but if they start with this cryptocurrency/blockchain stuff, I am out. I am trying to fund development, not invest unstable currency technology.
Gitea itself started as a fork of Gogs, it wouldn't be too surprising to see it fork again if they start being creative with the brand.
I've also been a contributor to Gitea, but it's not time yet to call the banners.
I don't think this announcement was handled particularly well, but nonetheless, I think it's an important development for gitea to grow and become sustainable.
Specifically, Gitea is in desperate need of full-time maintainers (and probably a part-time project manager), and I see this as an essential step to creating the necessary governance for that.
I remain cautiously optimistic and would encourage everyone who is passionate about the project to raise their concerns and wishes in a constructive way. Another fork would be a disaster for the entire ecosystem and I think should only happen if there really are SIGNIFICANTLY diverging interests in the project.
I also haven't heard anything in the gitea threads about crypto, so I think they're just referring to bounties and other distributed payments.
I have nothing against creating a company. It's probably a good development, unless that company does something really weird like shilling blockchain stuff. Having professional support is great and makes sense since both GitLab and GitHub have managed on-premise options too.
I'm confused, they don't mention crypto. This blog post is simply an announcement that they started a company. The purpose seems to mainly be for providing support services and paying developers. I see absolutely nothing wrong with that.
It does mention DAO, which is crypto adjacent, however like you said the purpose listed in the post is to provide paid support services, the DAO is for governance of the community, not to shill some scam crypto coin.
I don't think a DAO is just cryptocurrency adjacent. A DAO is entirely about trading traditional, well known, corporate structures (boards, shares, Robert's Rules, ballot voting) and replacing them all with cryptocurrency structures (NFT
token holders, NFT tokens, "Smart" Contracts, more "Smart" Contracts). There is no DAO without cryptocurrency. There's no token trading or smart contract operations without trading in whatever cryptocoins underpin their DAO's platform of choice. If you think all cryptocoins are a scam to shill in one way or another, there's no DAO operations at all without shilling at least one scam cryptocoin (and usually there's more than one involved on every current platform that supports "smart" contracts).
Meanwhile there are existing open-source foundations and conservancies that operate just fine as traditional (not-for-profit) corporations and no need for a DAO or any cryptocurrencies for community governance.
Whatever the official purpose is going to be, there will be lots of people trading the relevant token and people who got it first are going to cash out during one of the many pump and dumps. And the "why aren't you collaborating with (whatever Web3 project)" / "why aren't you promoting the token more" spam will start too. I've seen it before and it's not going to be pretty.
Recently I researched a little bit about this Gogs and Gitea situation. Here's what I've got: Gogs is maintained by a single person. He was reluctant to accept external contributions, so Gitea was forked. Now Gitea is very active with lots of contributors and more features, but Gogs is still maintained and does have many necessary features.
We were migrating from GitLab, so I considered to install Gogs, we decided to go with GitHub unfortunately.
The term DAO as it's used now is orthogonal to blockchain and crypto.
It's mostly a term to describe codified governance/community systems. There may be point systems implemented via crypto tokens, but in the vast majority of cases those would be better implemented on centralized ledgers since there's someone with centralized control over either the token issuance, the communication channels, or other essential resources.
If they were developed now, Stack Overflow and Wikipedia would be considered DAOs.
That would just indicate the term 'DAO' doesn't mean anything at all.
I'm unable to find an actual example of anything calling itself a DAO which doesn't also play with cryptocurrency tokens, so I'm pretty sure the term isn't used outside of that fanbase.
Stack Overflow was developed as (and is still operated as) a for-profit/profit-motivated Corporation (Stack Exchange Inc.). No one considers it a DAO in any form, especially not the "autonomous" part. Unless you mean the gamified community elements, but you can't really call that an "organization" either because it's entirely disorganized (and sometimes highly dysfunctional) beyond the game mechanics and more importantly extremely centralized to Stack Overflow's servers.
Wikipedia is run as a not-for-profit Foundation (Wikimedia Foundation). It doesn't seem anything like DAO either. Again, unless you mean the community of contributors to Wikipedia, and that also is extremely centralized to Wikipedia servers and doesn't have anything resembling "autonomous", not even something resembling Stack Overflow's game mechanics.
The "autonomous" in DAO still only means "smart contracts" and no one is using DAO as a term for traditional corporate structures other than those intentionally confused by or wishing to confuse what "autonomous" means in the acronym.
Hi, I'm the author of the post. That is correct, I mentioned a DAO as a possible way we are investigating for empowering the community, we are very much not trying to make a new currency.
This is prejudice. DAO in its most fundamental terms is transparent auctioning, staked decision-making, et cetera; in other words, totally valid set of use cases for the purpose of managing an open source project out in the open. Perhaps one of the few cases that. One of the accidental benefits of running it in crypto is the ability to consider rewards irrespective of jurisdictions. Think contributors being rewarded for carrying out a feature, on release, in effect immediately, irrespective of where they're coming from. It could be some poor girl from Iran, or Venezuela who otherwise wouldn't be able to partake due to restrictions re: Visa/Mastercard imposed but could make some money here doing open source work. Yes, much of the blockchain space is occupied by scams but sometimes it just makes sense, especially if you're willing to take a walk outside of your average "institutional" frame of logic.
That's not something a serious, registered company would risk given known sanctions against Iran. A random scam dao maybe doesn't care, but I suspect gitea does.
Serious, registered company wouldn't know— who is that on the receiving end of the bargain and therefore can't be held responsible for it; that's exactly the point why this technology is so beautiful and why it will ultimately prevail.
That's not how law works, you can't just look the other way. If you distribute money to random people without restrictions, records, or knowing who they are, that's called money laundering. That's what crypto tumblers are.
The fun part is: "CFTC’s order first finds that virtual currencies traded on the bZx Protocol, including ETH and DAI, are “commodities” under the CEA.", which is close to "It's cute you set up your daos and automated protocols, but legally that's irrelevant."
Regarding ignoring the identity: "The order further finds that the Ooki DAO failed to adopt a customer identification program, in violation of the Bank Secrecy Act and the Commission regulations promulgated thereunder."
So they settled, interesting! Sad it didn't go to court, could have made a curious precedent. This is different, however (although I haven't done much legal work in this space) as it's an investment open source DAO can be structured in a way that does not have clear beneficiary structure and as long as the primary software developer can show that they can't exert ultimate control of the network, it's all for nil. (Now that's a whole different kettle of fish!) You have things like GNU Taler potentially being able to map the existing irl ownership structures quite closely, and all kinds of ZKP, Shamir-based stuff on the end of the spectrum where it all breaks down. I allude to the latter, which would make sense (technically using a mixer would constitute money laundering in the spirit of the law but we don't see people get charged with it, never. And doing so would set a disturbing precedent. If you're starting with some primary shareholder, you're probably doing DAO wrong, and it then becomes less decentralised and more like crypto cosplay on the prior structure.
Gitea is a great piece of software. In fact, when starting out with self hosting, I found it to be a very capable solution with a small footprint.
Over time I realized: I don’t need to be able to merge pull requests and do code reviews in a web interface. Nowadays I’m invested in a plain Git/SSH setup with cgit as a web frontend and email patches.
It never breaks, is as secure as your server is and does everything you usually need, including user permissions.
The `git am`-based workflow is less than popular outside of FreeBSD, Linux kernel development and SourceHut though. Depending on your environment, if used in a work setting, your colleagues may balk.
That being said and given the existing enterprise users of Gitea, it’s entirely natural to eventually offer solutions tailored to their use, like SSO. I wish them the best for building towards a sustainable future while staying true to their core values.
I agree. I can't even remember why I was so scared of trying out email + path workflow, but after trying it for just 1 time I was convinced. It is so obviously good way to manage patches, you get a unified interface to bugs, patches, communication and everything in-between.
I can understand some folks not wanting to do initial setup and sourcehut already provides a nice web interface for it. But for actually contributing stuff emails are pretty unbeatable IMO.
Hi, I'm the author of the blog post. Thank you for your comment. A project you may find interesting is soft-serve from charm.sh https://github.com/charmbracelet/soft-serve it's a "TUI" (terminal user interface).
Thanks, it appears to serve the same function as gitolite, which I use in my setup and somehow forgot to mention. The SSH interface seems unnecessary though.
The manual bare repo setup obviously offers no user access management beyond what the system's SSH configuration provides. Beyond that, a bare repository is actually an essential Git feature and what software like Gitea and even GitHub uses behind the scenes.
Given all users can share a user account on a server, it will behave exactly like you expect it to. Again, that's a core Git feature. Additionally, AFAIK Git has its own locking management. Also look up the `--shared` flag. `man git-init` is very helpful.
> If you are a company and rely on Gitea, especially for critical operations, please get in touch as we are now able to offer:
> [...]
> An enhanced enterprise version
It's sad to see Gitea not receiving enough donations/funding to stay afloat, and I obviously can't fault them for taking the necessary steps to generate revenue, but it's also sad to see it eventually going open-core.
This announcement took the entire Gitea community by surprise, including volunteer developers contributing code daily to the project. Gitea turned from being a community led democratic project to a company that owns the domains and the trademark, overnight.
This is an even bigger problem than going open core and is begging for a fork to happen.
Great points. I agree there will be a backlash from contributors.
As far as I can tell they don't ask contributors to hand over copyrights, but they use the permissive MIT-style licence which permits the creation of non-Free forks.
Drew DeVault, of SourceHut fame, has written on this topic. [0][1] Drew's opinion is that he probably couldn't easily turn SourceHut to the 'open core' model even if he wanted to, on account of its use of AGPL-licensed code from contributors (who retain the copyrights).
As far as I've seen it's actually the opposite. Gitea has been growing in support and popularity, and as such, it needed to put some legal structure in place to deal with increased income and distribution of funds so that it's not abused.
That is, I BELIEVE, the intention with the announcement and structural changes. As a relatively significant commercial backer of both bounties and Open Collective, I was as surprised as everyone to see the announcement. If my assumptions are correct, I completely agree with the decision. I want developers who contribute to gitea to benefit from the software they are building and think there is enormous potential for everyone to share in the upside.
It's an exciting, new junction for the project. I think we are a little bit in the dark at the moment and I'm pushing for better communication.
> Gitea has been growing in support and popularity, and as such, it needed to put some legal structure in place to deal with increased income and distribution of funds so that it's not abused.
None of that requires turning it from an open-source into an open-core project.
Hi everyone. I'm the author of the blog post above, and I am sorry that I caused so much confusion.
I have spent much of today trying to clear things up, and will be writing a follow up to the post.
I see some concern around the mention of a DAO, I tried to get across in the blog post that community is incredibly important, and so I mentioned it as a possible way to ensure community governance, not to create a cryptocurrency and sell that. I mentioned it to be entirely transparent that it was a way that we are possibly investigating to make sure community is still involved.
Thanks to our paying clients, the company has been able to hire some maintainers, and contribute back the changes we've made back to the project so everyone can benefit. We are also looking into establishing a fund to not only pay contributors, but also projects we rely on.
> To preserve the community aspect of Gitea we are experimenting with creating a decentralized autonomous organization where contributors would receive benefits based on their participation such as from code, documentation, translations, and perhaps even assisting individual community members with support questions.
This is a DAO where you earn scrip through a gamified version of FOSS development. This has nothing to do with "community" (which is a word I'm pretty sick of hearing coming from the mouths of software devs and tech companies).
I'm concerned because domains & trademark, previously under the governance of elected community leaders are now owned by a for-profit corporation based in Hong Kong. This is a takeover.
I work daily on the Gitea codebase as part of my efforts to further forge federation in forgefriends and did not get any advance warning, just as most volunteer contributors.
The development of Robot Framework(1) was originally paid by Nokia, but then things happened and Nokia decided not to continue doing that anymore.
The main developer of Robot Framework and few companies using RF heavily understood that something had to be done, so Robot Framework Foundation(2) was formed. RF Foundation has membership fees, it arranges RF conferences etc. which allows RF Foundation to pay for the development of RF.
I think that is a really good way to fund OSS development. Those companies which benefit the most from it, pay membership fees and gets to vote on the direction of the product.
This is comparable to how Codeberg [0] is organized, which is also based on Gitea and one of the larger contributors upstream. Yet the number of members is tiny in comparison to their person base and growth. According to their latest numbers [1]:
> We are hosting 42010 repositories, created and maintained by 34042 users. Compared to one month ago, this is an organic growth rate of +3372 repositories (+8.7% month-over-month) and +2271 users (+7.1%).
And further on:
> Codeberg e.V. has 246 members in total, these are 175 members with active voting rights and 69 supporting members, and 2 honorary members.
That is literally the easiest inclusive way that works beyond a single country's borders. Easier for me would be iDeal, but that's not easier for you unless you're also Dutch. Easier for people in the USA might be credit cards, but those are issued only if you have a credit rating at all and the rating is good (I don't meet the former condition so bank said no), and surely transferring money from anywhere in the world to an IBAN got to be common enough that it's doable and cheap/free in any country with decent payment infrastructure?
> Easier for people in the USA might be credit cards, but those are issued only if you have a credit rating at all and the rating is good
Debit cards are effectively the same as credit cards for this purpose, paying for things online. Many (if not most) banks and credit unions in the USA will issue you a free debit card upon opening an account, some without even needing to put any money into the account.
Credit card would also be a big issue for German users, even in tech. I know a bunch of people who don't have one, as they aren't accepted in most places within Germany anyway, no idea how they make it work when travelling though. I guess for German users Paypal would be best, that seems to be the most widespread solution.
In US, if you have no credit score or a poor credit score, you can still get a credit card by making a collateral deposit with the bank (usually they expect it to be 2x your expected credit limit).
And transferring money to random bank accounts outside of the country is rather messy in US. Some banks won't even let you do it online.
Interesting, I didn't know that was a thing. I offered this when I couldn't get a credit card, but no dice. Didn't know this was a common solution elsewhere. I figured they didn't want to deal with custom solutions for uncommon cases like foreigners with no credit score wanting a debit card that has a credit number (essentially, that number is all I want, I don't need a loan/credit at all). Maybe it's not a thing here because a credit card is not as common or necessary.
As a single system, though, it beats all of them. My mom doesn't have a paypal but she sure has a bank account. More options is always better, especially if you are happy to swallow exorbitant fees from the various systems, but I can very well see why they chose SEPA as primary payment method. I often wish more websites supported it: I don't want to pay more middle men, accept more privacy policies, deal with more crappy websites that break every few months when using an ad blocker, etc. when I'm already paying for a bank account where I can just send money from.
Don't non-consumer account holders pay much higher fees? I seem to remember they were okay if you can have a very low transaction cap, but if you need to accept/send many payments, you had to get expensive subscriptions or pay per transaction.
Almost certainly still cheaper than literally any alternative, though, since you need that bank account anyway and I will get very much doubt they'd take as big a cut as the paypal mafia.
Yes, bank accounts for commercial purpose have different fees, but unless they pick a completely stupid bank it's still multiple orders of magnitude cheaper than any other transfer.
Only downside of SEPA is that it is a lot less common outside EU/Europe and then more expensive and instant payment is not really possible (there are some vendors offering solutions ...) but that doesn't seem to be their aim. (Would be: "click here to get a license for a commercial download")
Can you not just enter an international bank account number (IBAN) into online banking outside of this area? How does transferring money work then, do you need to send couriers with cash or use third parties like paypal? IBAN is the most basic infrastructure I know of, and I've heard that in the past you also had bank routing so this paying via banks (and banks settling it later) can't be new.
I should ask my boss how we get money from customers in other countries. As far as I knew we just give them our IBAN, chamber of commerce number for tax reasons, and it just appears in our account.
> As of May 2020, 77 countries were using the IBAN numbering system
-- IBAN Wikipedia [0]
It’s not that widespread. For international payments, you can do international bank transfers which have fees associated with them which can be shared or put on one side of the transaction only. It’s often easier and cheaper to use a 3rd-party provider like XE or TransferWise.
Amazes me that there is no standardized way to do money transfers between big countries that we very commonly buy/sell stuff from/to like the USA, China, etc. Thanks for the info and link!
Hi, I'm the author of the blog post, but also a member of the codeberg prasidium. Thank you for this feedback I will bring it back, and hopefully lessen the barrier to join.
You say "one of the larger contributors upstream" do you have a source for that? Because I took a look at the contributor list, and couldn't find any evidence of that.
Codeberg maintain their own long-standing fork of Gitea and have members that are also maintainer at the Gitea project. It is the persons doing upstream contributions, not a Codeberg account.
I am a big fan of Gitea, it does exactly what I wants it too and its very easy to maintain over time.
I run it together with ZFS and SQLite. This enables atomic snapshot of both the database and git repository. And this is of course replicated to another system. SQLite is plenty fast, people should avoid more complex solutions using PostgreSQL or MySQL. I've also configured SSO with Azure Active Directory which everyone with mail from Microsoft has access for free.
For the past few years I've completed a few dozen upgrades, only 2 times it has broken. One was a new configuration property I needed to add, and one was that I upgraded GoLang a bit too early and Gitea didn't support it. Compared to many other open source projects, Gitea is a joy to use and run. And I hope it will stay like this for another 10 years.
Gitea was a community led project with democratic elections. The entire community was taken by surprise by the announcement. This is what you would call a "breaking change" in a release :-)
One thing is for certain: Gitea is no longer what it was a few days ago and there is no telling what the next 10 years have in store.
The blog post says maintainers have been hired, as well as the Blender blog post is months old, seems like this has been public knowledge for a while, but that it is just formally being announced now.
Edit: seems I am getting downvoted, but I searched
dachary who has a financial interest in Gitea, and has posted about the company in the past, so he clearly has known about this for a while.
I wonder why he happens to not mention a conflict.
> What you said is false. I suspect that is why you're getting downvoted.
It's not, you can confirm what I said. Look at the link in the blog post that points to public mention of it months ago, as well as the user I replied to posts about the company to his personal site with excepts from the public matrix channel.
I'm active in open source and lots of chat happens in matrix, maybe your coworker missed it?
All you found was essentially me guessing from weak signals with no confirmation. As for my financial interest in Gitea ... I was never involved with the newly formed company. If you mean that I raised over 100,000€ for the benefit of the Gitea project in the past year, this is a different matter. You can read about it at https://discourse.gitea.io/t/nlnet-grant-for-federation-conc...
Interesting that you forgot to mention the company YOU started to sell Gitea, even after multiple opportunities to mention it. I also checked the maintainer list, you have mentioned democracy many times in this thread, but yet you have not voted, nor maintain gitea.
By not mentioning this major conflict brings your motives into suspicion.
Oh, fuck - I had a good run I guess: I love Gitea, but this seems like a change for the worse.
I think there's an open source trap where successful projects see their adoption numbers, and billion-dollar companies using their projects and think "We're leaving money on the table" (which IMO is the whole point of libre software).
Unfortunately, this is the path to weird software licenses, open-core / community-edition product splits and a whole host that makes the software less useful to me as an individual user and occasional contributor. Besides, Gitea won't be able to make much out of individuals like me, so their roadmap will likely favor paying customers. I can guarantee that it will be too bloated to run on a Raspberry Pi Zero like I have it today, which means I'd have to choose something else.
I also suspect that switching to a for-profit company will decrease the number of volunteer contributions, after all following the same big idea that led to incorporation: who wants to work for free when someone else is making the money?
The good news about F/OSS is I have started my countdown to a community fork. After all, Gitea itself is a fork -it is the circle of life.
Edit: I forgot they are trying to pay contributors with some web3 scheme, perhaps to try and solve the inherent hypocrisy. But the cure may be worse than the disease because it further complicates the process of contributing to the project and reduces maintainer focus on actual product. Just give me pull-requests.
Hi, I'm the author of the blog post. Thank you for your comment. In reference to your edit, we are investigating a DAO to empower the community, but aren't committed to it, only investigating it to see if it is viable. In terms of funding the community, we are committed to creating a fund with non-crypto money to be able to give back.
In terms of "leaving money on the table" this is very much not that, open source is incredibly non-equitable and skews a lot to a certain way (a lot of people with who can afford to give up time to open source, not everyone but a large amount). This prevents many who have a desire to work on projects such as Gitea, but need to find a job or two to pay rent and get groceries. By establishing what we have, we are hopefully broadening the pool of people who are able to participate.
Please don’t introduce any element of cryptocurrency or blockchain to the project. Nothing it offers will solve any real problem that a simple database couldn’t do more efficiently. No matter how you turn the worm, ”crypto” is a grift. Enough incidences have shown that even small bugs or unintended omissions in the implementation can lead to hostile manipulation and theft. Mitigating the risk of catastrophic failure, once you remove the money from the process, it makes even less sense.
What you need is humans in the decision-making process, not algorithms.
The DAO was mentioned only around transparent governance and giving back to the community, but the blog post mentions hiring maintainers presumably with cash.
You seem extremely vested[1] in this, and have been very active in this thread - are you affiliated with the freshly-registered company? If so, I'd encourage you not to hide behind a throwaway because it gives the impression that you don't want to stand by your words.
More on topic: does this mean occasional/casual contributors to the project will continue to work for free while maintainers & employees get paid?
1. You looked up the name of one commenter and the donations made in their name 3 years ago.
Indeed I have been active. I am passionate about funding in open source, and saw someone link this on mastodon, and so I signed up to comment on things. I only know what has been said in the blog post. I have attempted to disclose all my leanings, and follow ddevault [0] and disclose especially around crypto things
> Additionally, we’re planning on establishing a fund to be able to provide support to contributors who not only contribute features, but also bug fixes, performance enhancements, and important refactors.
Sounds like occasional contributors can receive payment as well.
Like you, I also agree that financial disclosures and citing sources are important, and so so that is why I took the 5 seconds to validate a claim someone made, and also found someone in this thread who very clearly is not disclosing a financial stake in the matter.
Edit: I have seen a maintainer in this thread, perhaps he could share some more details as they come.
> there are a few corporations (with revenues that are greater than some countries GDP) are building on Gitea for core products without even contributing back enhancements
That's exactly what you get when you release Gitea under MIT.
Yeah I was fully expecting a licence change with that preamble. And then they mention DAO ._.
GPL scares people for stupid reasons (stupid from my fanatical perspective that if you use open source software, of course your derivatives ought to be open as well), Gitea honestly might be so popular because it had a more free license than GPL. Doesn't mean I'm sad to see they're not even trying to make these changes be contributed back going forward.
To me those are synonyms, but I'm not a native speaker. When you're allowed to do more things, such as taking open source code and using it without ever contributing anything back even if you distribute compiled versions of these people's work for exorbitant fees, aren't you more "free" to do what you want than if this were restricted in any way?
I'm not saying I morally agree with it, but that's what I understand the word to mean.
Whose freedom you're talking about matters: and your perspective focuses on developer freedom.
Which is the freer society: one where loud, all-night music in residential areas is permitted with no consequences, or one with restrictions on noise disturbances? Your definition supports the idea that late-night-parties has more freedoms, and I disagree with that. Hence the "eye of the beholder" thing applies.
My own subjective perspective is rather utilitarian: there are more users thsn developers (as developers are also users), hence, user freedom is more valuable than developer freedom, and it is worth restricting some developer freedoms to that end.
Hi, I'm the author of the blog post. Thank you for your comment. I'm sorry about being unclear with that part. The issue is less them using the project, but more creating a support burden by sneaking into community forums and hiding their affiliation. I'd much rather donate my time to homelabbers, and people not being paid an obscene amount of money who in theory should be able to figure it out themselves.
- work for a company that uses open source, look for excuses to support it (support contract, etc. I even got good value out of some.)
- are an open source project that takes donations: don't get greedy. Two projects I went out of my way to support because they were open source shortly afterwards started "experimenting with models".
> are an open source project that takes donations: don't get greedy. Two projects I went out of my way to support because they were open source shortly afterwards started "experimenting with models".
How do you know it's greed and not just insufficient donations forcing them to try to find alternative financing? Just because you donated doesn't mean enough people/businesses did.
See, that's the beauty of open source: nobody is ever obliged to do anything they don't feel like doing. If nobody wants to pay, the maintainer can just walk away. If that bothers a user, they can feel free to contribute patches and pay the maintainer's time to review and include them. (Or become a maintainer themselves using a different trademark.)
Of course, the flip side of that coin is that the maintainers are also free to try to turn their repository access into a money making machine. I don't have any issue with it, since I can just fork the code if I don't like it. It's just not the result of getting too little money and being "forced" into anything.
What’s “insufficient donations”? Insufficient for what?
What this is saying is that:
1) There are some large companies that are using Gitea as a core product, and not contributing their enhancements back to the community.
2) We are starting a company that will have Gitea as a core product, and will not contribute our enhancements (“enhanced enterprise version”) back to the community.
Not enough for what? Paying expenses? Or are you expecting that open source contributions are to be paid at developer rates, and that one should contribute only as much time as is adequately paid?
I thought open source meant that everyone was volunteering by default, just writing the features they'd enjoy to have for themselves, and that sponsorship of development time is not the norm aside from hugely important projects (like bigcorps sponsoring kernel development, but also making billions by launching Android that makes extensive use of this kernel) or projects where the sponsor has a commercial interest. Donations are typically for raising money for things like hosting the website and download bandwidth fees.
Adopting GPL used to be a philosophical mindset consciously chosen (as opposed to freeware), and required some level of zealotry. Now, it's just a means to an end / pragmatics / a tool in the "growth hacking" toolbox because "everyone knows" having an open source project gives it an edge over closed-source equivalents on adoption.
IMO, it's not that open source developers have to earn a living (which they always have since the beginning), but that developers who want to earn a living in any way have started adopting open source by necessity. I'm no purist, regardless of motivation, the more open source projects there are, the better it is for everyone. For once the source is out: it is out.
Yep. I recall the blog post by InfluxData on monetisation/sustainability and that there's very few options available for an open source software to become self-sustainable, and even fewer working ones. Influx went with the open core route, removing features from the open source version, which kind of killed it.
I took a look at Gitea's Open Collective, and they have something like $300 coming in each month, and that there is a large amount in the account because of a few large companies, but not re-occurring ones.
I'm going to assume you are not an enterprise - forgive me if I'm wrong.
Enterprises have very different needs to individual developers and small shops. So there are always features which apply to enterprises but not to small companies. Consider for example Active Directory integration.
Equally purchasing in an enterprise is different to small companies. If you work for a small company you can basically go to a decision maker, make a case, and if accepted then money glows within the day.
Enterprise businesses don't work like that. They basically can't just "donate money". There are a million layers to work through, and there have to be tangible reasons for spending the money - it can't be things like "good karma" etc.
So for OSS projects that are used by (especially non-tech) enterprise customers, it makes sense to be an entity, and sell a product, that enterprises can buy.
The inhibiter for an enterprise is not a lack of money on their side, its a lack of process on the OSS side which matches up to the process they have to fulfill.
The issue, in my mind, if I was making a similar comment, is that projects seem draw the line in the wrong place for where "Enterprise" begins.
I suspect this person's fears may be similar. The issue isn't that they'll offer enterprise, the issue is that they'll carve up the non-enterprise version as a differentiator.
Inevitably, and I've seen this before, the non-enterprise version loses all steam because time and effort have to be focused on the enterprise customers, and feature parity can't be maintained because those differentiators have to show value in the enterprise offering.
Those OSS users who built the community and contributed to your product for years end up having been beta testers, unceremoniously ushered out to find an alternative.
Elastic for example made the mistake of making "secure" (by default) an "Enterprise" feature of Elastic Stack, then Amazon came along and ate their lunch.
For me, I've seen this enough not to worry; if they go "south" in this respect, there are alternatives; and failing that, there's the minimalist approach of plain git.
This is pretty much my sentiment as well. The problem is that the obscure integration problems are better solved by a system integrator if they are so bespoke they don't belong in the code base.
I've worked with enterprise customers before and yes it tend to be LDAP and active directory stuff they ask for in terms of features. I run my own LDAP, am I enterprise?
Further this undermines their argument. It comes off as "hey we want some of that pie". The alternative is the foundation route. Or just to start your own system integration business, not to muddy the lines.
So, yes, (to answer both followups here) - obviously there's a danger that the OSS version gets less time and attention, and in many cases it'll start prioritising actual customers (ie those paying money) over folks using the OSS version.
I certainly understand why those who are using software for free see this as the thin-edge-of-the-wedge, the start of the product morphing to something that is less free.
On the other hand, despite the health risks, eating is still enormously popular. Developers want to get paid, and so they tend to gravitate towards models that, well, earn money. OSS is great (I'm a fan) but it doesn't pay the bills. OSS should be what you do in your spare time, not your day job. If OSS is your day job then either you're lucky, and someone is funding it, or it's only a matter of time before you need to switch to a different day job.
Bottom line - if OSS doesn't sort out an actual funding model that works - if users (of the Libre version) aren't willing to pay real money, on a regular basis, en masse - then don't be surprised if the developers need to switch it up to earn a living.
There's this perception that "big companies" should just fund all OSS for everyone, but I'm saying that this can never work - that's pretty much exactly what "big companies" are set up not to do.
I understand all of this. I'm a software engineer myself, and FLOSS advocate.
The "complaint" is not that I don't understand _why_ this is happening, or even so much that it _is_ happening, but in my experience, this "commercialisation" of F/L/OSS is inevitably (I'd argue) net negative for the community built around it.
Even if the core product doesn't become "paid" and we don't have this issue of whether or not people _should_ have to pay to use it, the quality of the "core" or "community" product will fall, anything new or interesting will get pay-walled and updates/features will be few and far between.
I'd call this more lamentation of a loss than anything, Gitea isn't dead (yet) because of this, but I move forward cautiously to see what comes of it.
What would be really interesting, would be to revisit this discussion in a year, two, maybe more and see what became of it.
The problem is when any of the features the enterprise wants to / can buy is closed source and not available on the same terms as the rest of the project.
I'm not wanting to insult you here, so please read in that light. I'm not trying to be the dick here.
>> The problem
Problem to whom? Not the customer (the one paying the bills), not the developer (the one getting paid) but by "the rest" - the ones consuming but not contributing.
>> closed source
Unfortunately the term "open source" encompasses a group of freedoms - the lack of any of them leads us to the term "closed source". For some projects the freedom "removed" is the right to _distribute_ the proprietary source. In other words the source might be "open" in the sense that the customer has it, but it might be "closed" in the sense that they can't distribute it.
I'd suggest that the freedom-lost in the "can't distribute" model is a good compromise (developers wanting to get paid.) I'm less favourable to a "binary blob" model. Clearly in the abstract projects could go either way.
>> not available on the same terms
ie - for free. I hear you, but I feel like developers have a right to take their project in a paid direction if they want to. The demand that they continue the project forever for no return seems, well, unfair.
> ...but by "the rest" - the ones consuming but not contributing.
"contributing", when it relates to F/LOSS software, includes non-monetary activities, which you seem to be ignoring. Nevermind the fact that the so-called consumers are the whole reason why Gitea is in the position it is today. Without adoption, it would be a bunch of people making commits on an unknown codebase. Characterizing users as moochers is bad form, amd will not end well for Gitea, IMO.
> I feel like developers have a right to take their project in a paid direction if they want to.
The other side of the libre software coin is that the users have the right to fork the project and take it in a community-driven, non-profit direction if they want to.
>> includes non-monetary activities, which you seem to be ignoring.
sure. Except those non-monetary activities don't pay the bills. I suppose I kinda fell like they have value, but on the other hand the business has to pay bills and salaries first. If that isn't done then the non-monetary value is, well, meaningless.
And of course, if there are specific users who are contributing materially with code etc then nothing stops them from getting a gratis enterprise license.
I think it's also worth noting that the "community" consists of two groups - those that are contributing to the project, and, well (your word) the moochers. Yes, even quiet users have some "value" - but it's really tiny. The ones that are contributing code and so on, obviously they have real value.
I certainly don't want to minimise the value of active contributers, but at the same time I don't want to over-value the contribution of "n anonymous users". Yes, there's a business model built on acquiring some huge number of users, then selling them on to some big corporate (like GitHub did), but I'm no sure that's in play here.
>> Nevermind the fact that the so-called consumers are the whole reason why Gitea is in the position it is today.
I'm not sure they are so-called - they clearly are consumers. And that's great for a small fledgling company starting out. But there comes a point where the bottom line has to be fed, and they aren't the ones doing the feeding.
This is obviously not a new thing that Gitea invented, we've all seen this play out a million times with startups. It's all free all the time until the money runs out. Then something has to change to make it sustainable.
>> The other side of the libre software coin is that the users have the right to fork the project and take it in a community-driven, non-profit direction if they want to.
Of course yes. If there are some group of developers who want to work on features, they are of course most welcome to do so. That is quite literally the whole point of an Open Source license. But you say this like it's a bad thing? Like it would somehow hurt Gitea if this happened. I'm not sure that's the case. (you'd just have a new set of developers with the same financial problems, and Gitea would have gobbled up some Enterprise mind-share in the meantime.)
> Except those non-monetary activities don't pay the bills.
That is why the vast majority of free or open source projects are not businesses.
> And of course, if there are specific users who are contributing materially with code etc then nothing stops them from getting a gratis enterprise license.
This is one way to kill an open source project that is new to me!
> I'm not sure they are so-called - they clearly are consumers
They do more than that (in a power-law way): the evangelize, translate docs, file bug reports, donate, provide free support online, among other activities. Reducing a community to mere consumers is self-defeating.
> you'd just have a new set of developers with the same financial problems, and Gitea would have gobbled up some Enterprise mind-share in the meantime.
Ah, I see we have a fundamental philosophical difference. I don't believe that all open source projects have to be profitable enterprises - or even a full-time job for single maintainer. A lot - I dare say most - successful projects are volunteer-driven.
The world, in my opinion (we all are being opinionated rather than right or wrong, here).
Like [almost?] everyone, I have ideas about what I think would constitute a better world. They include, within the realm of IT, a pretty much "IP" and DRM free society, open general purpose computing for all, and more truly free [as in freedom] and open source software.
I think that open core and enterprise service based models and their ilk make that world less likely. I'm not interested in finding ways to make more "silicon valley" companies profitable if they do so by making a world I don't like.
It means there will be a proprietary version of Gitea, named Gitea Enterprise or something similar. The company is based on an Open Core business model https://en.wikipedia.org/wiki/Open-core_model.
Is it? The blog post calls out their participation with Blender, and reading the linked blender posts and videos where they talk about their work with the company they talk about the relationship being entirely about providing advice and support, I don't think blender bought some closed source software.
Lots of focus on the DAO mention in the post, but it looking at the concrete things in the blog post instead of some "experiment", there are direct funding models of selling hosting, support services, training and more. The DAO is likely just a way to make the governance of the community more transparent, but I think people are confusing it with creating a crypto coin.
I’d rather see funding come from a crypto token that keeps the software completely open source than enterprise sales, aid hosted versions, and going “open core”.
I don’t understand how anyone believes that the current trend of so called “open source” startups is good for free software or anyone other than investors.
Given that, I’m very up for experimentation with new types of organisational and finding models for open projects, including ones that use cryptocurrencies and smart contracts.
Hi Barnabee. I'm the author of the blog post. Thanks for your comments. I'd be interested to know why you would prefer funding come from an external crypto token instead of a paying for a hosted version. I would think that dogfooding is a way to stay on focus on the main project instead of being distracted by something else.
For me the main deterrent from Gitea has been their approach to UI design. The shameless inspiration (if you can call it that) they took from Github is pretty startling, the two interfaces look pretty much identical (well, Github has recently redesigned its repo page, but many can remember the old design).
Sure, get inspired, make something similar here or there, but this is an outright copy of the whole product design.
I just compared Gitea and Gitlab side-by-side and they're very, very similar. So either there's also copying between GitHub and Gitlab, or the design of git lends itself to a very specific interface.
> So either there's also copying between GitHub and Gitlab
Isn't it likely that there's just a certain kind of interface and functionality that both works and people have also gotten used to it over time?
It might make a lot of sense to copy it to at least some degree, instead of reinventing the wheel: if you look at MS Office and something like LibreOffice, you'll notice that both of the spreadsheet apps are rather similar (and you can even enable a ribbon interface in LibreOffice, if you want).
I think the same more or less applies to every piece of software, from how phone OSes look, to how desktop environments look and work, as well as why the majority of websites out there look a bit samey.
For me this is a benefit, GitHub's UI in my opinion is really good and I'm glad I don't have to learn a new one for Gitea.
Whenever I need to report a bug on a GitLab instance, its UI is a big source of friction (especially since GitLab's UI is still annoyingly slow when loading things like issue comments).
The implication of a DAO and Crypto involvement has stirred attention in the comments. I have 2¢ on the matter. First, I don't think we've seen many examples like this in open-source communities. For that reason alone, regardless of the idea's merit, I welcome it; if it is to fail, let us see it, that we may have a case study in the future. Worst case, there is wide support for a fork, and I don't see that becoming any more unreasonable over the next decade. Secondly, just as a brain-storming idea, what if we had like the blockchain and stuff but the rewards weren't pure value (ie. contextual value, one man's treasure; titles, hats, cosmetics, merch vs purchasing power and/or influence)?
I hold zero crypto, but by reading of the DAO is that it is one of the ways mentioned of trying to benefit the community or perhaps it will be used for community governance (elsewhere in the post it mentions hiring maintainers, and also creating bounties to fund development), not a way of creating some crypto coin to sell.
Even if it was a way to sell crypto coins and fund development (presumably while also giving backers some visibility and say on the project’s governance), how is that worse than making part of the project closed source and paid for?
Looks like Blender is onboard with moving to Gitea. However:
> To preserve the community aspect of Gitea we are experimenting with creating a decentralized autonomous organization where contributors would receive benefits based on their participation such as from code, documentation, translations, and perhaps even assisting individual community members with support questions.
So Gitea wants to create a DAO and "receive benefits based on their participation such as from code" assuming it going to be similar to Gitcoin, which may involve using crypto tokens, NFTs and the like.
Creating a DAO for this sounds _incredibly_ risky to me. One tiny flaw in the design of that organization and the project could be destroyed by people gaming the DAO rules for profit.
There's nothing literally stopping someone click "fork" on GitHub. There are tons of things that make it difficult for people to create a successful forked version that people use and contribute to.
Though in this case I guess they could just use Gogs anyway.
Hi Simon! First, let me thank you for all that you do with Open Source. I package a few of your Python libraries for NixOS, and they are all so wonderful. I'm the author of the blog post above, and agree with your concerns. That's why we are not committing to creating a DAO, but wanted to to let everyone know it was something that was being looked into to empower the community, and make sure the voice of the community is still very much present. I mention community 4 times in the blog post it is that important to me.
If they didn't mean blockchain they could have not used a phrase that is very strongly associated with it, and instead used any number of other terms in colloquial use that match the properties of the organization they're setting up.
If you look at the sponsors at the bottom of the page linked, the first name is a blockchain company/organisation, which in my mind lends credibility to the fears that they really do mean blockchain and aren't just jumping on the buzz-phrase bandwagon.
We have the words "corporation" or "company" (or "foundation" or "conservancy") for easy to find synonyms of "organization" and despite appearances none of those words directly implies either "centralized" or "decentralized". There are companies and foundations that are built quite decentralized meeting only remotely over the internet today.
Companies and corporations are far from "autonomous" however, and that word currently always implies "smart contracts" and blockchain.
Why not just use shaming to get big companies to contribute? Name them, keep a list, put them on display. I mean, they should be ashamed to use a free open source and not contributing in any way.
This makes no sense. I have never heard anyone say github was ugly and gitea pretty much replicates most of it.
It could be more singular/original but people using that kind of tools are looking to get the job done, not to use some fancy website. I'd rather have speed than anything else.
It is funny that on the same community people can blame an open source tool such as gimp for not being a straight copy of photoshop, then are not happy when another open source project is staight copu of a proprietary one.
There is however a difference and that matters significantly. GitLab was a company when they made the move to Open Core. It was their decision to make, unilateraly.
Gitea was a community driver project, with elected leaders assumed to care for the need of the community before their own. But they secretly created this company and transferred the domain and trademark to the company. So the community is gone an the volunteer contributors have been taken by surprise. This is not good.
The thing is, I'm very happy with that not being part of the model, but the whole model.
I don't want to deal with experimentation or complex organizations or "an enhanced enterprise version" or free tiers. I want someone to take money from myself and other customers, and use it to make a good product without shenanigans.
E: As far as I know, it is still possible to get free accounts on sr.ht - however, it is very transparent that this is not a permanent offer.
Merely for your consideration, if you were to rewrite the 502 template to mention your status page, it could save these kinds of "communication gaps" in the future
Well, [text](href) is just a lousy syntax. Quite apart from how easy it is to forget which way round it is, the way round that it is is syntactically problematic: the parentheses are URL code points <https://url.spec.whatwg.org/#url-code-points>, so URL encoding won’t percent-encode parentheses, so Markdown doesn’t actually support all valid URLs, leading to injection attacks if all you do is regular URL encoding, deliberate or accidental, and deciding where an href ends is troublesome and inconsistent, with some Markdown implementations terminating at any right parenthesis, and others trying to match parentheses as a heuristic that helps most cases. The other way round, with the href in square brackets, would have been better in this regard, as square brackets aren’t URL code points, and thus will be percent-encoded. But better still would have been to lean on angle brackets more, matching long-held custom and the other style of links Markdown already uses (just plain <href>). In my own lightweight markup language that I’ve been working on for a while and am now polishing up and implementing properly, I’m currently using [text <href>]. [text]<href> is also quite tempting, with slightly different trade-offs.
(When I speak of the details of URL encoding and which characters get percent-encoded, these things weren’t quite so clearly-defined back in 2004 as they are now, but I believe it was all still true.)
HN doesn’t use Markdown in any way, shape or form, so of course it doesn’t do anything special with it. But anything that does actual Markdown will handle it.
Hi, thanks! I'm the author of the blog post, and I'm not sure if you were the person who pointed it out in chat, but I have since fixed it thanks to that heads up.
Gitea itself started as a fork of Gogs, it wouldn't be too surprising to see it fork again if they start being creative with the brand.
[1]: https://opencollective.com/gitea