Odoo Community Association
Criteria for adding your name to contributors #185
Replies: 12 comments 34 replies
-
|
I'd say the real conflicting ones are authorship and code copyrighting |
Beta Was this translation helpful? Give feedback.
-
|
FWIW copyright lines in modules code files are AFAIK meaningless, as contributors sign a CLA that lets the OCA retain it. |
Beta Was this translation helpful? Give feedback.
-
|
hi @francesco-ooops. thanks for this discussions. could you add other similar questions, to be exhaustive :
thanks ! |
Beta Was this translation helpful? Give feedback.
-
|
Hi @legalsylvain I'm not sure how I can do this with a poll, feel free to open new threads or even better we can move the discussion in a PR against the guidelines file :) |
Beta Was this translation helpful? Give feedback.
-
|
This question has been arisen because on a cherry-picking without any modification, I have asked for removing the contributor doing the cherry-picking from the README. I think that's logical, as there hasn't been any significant modification (in fact, in this case, nothing, except renaming migration folders) on the code for adapting it to the other version. I think that can be the first guideline for this: "Only add us as contributor if there are significant code changes, or review effort, including migration to newer versions. In case of cherry-picks / forward-ports / backports, only those requiring to adapt the code in a heavy way." |
Beta Was this translation helpful? Give feedback.
-
I agree with @pedrobaeza . Furthermore I think the module authors and active maintainers should have the final word on what is a significant contribution or not. We really start to see people who got attracted and then kicked from Odoo partnership (in a way or another; so today it means not always so very open source friendly people in fact), understand that the OCA is the next "bankable" thing to invest and start doing very minor contributions in the only hope to force the name of their company into the module README, or start hijacking the projects reputation. This is not Sci-Fi, do you want to have an idea how such a thing would look at the end of the road? Just take a look a app.odoo.com where apps don't work, are copied one from another and fight to show at the top with their flickering gifs. Is that what we want for the OCA image?? I think most of the time just applying the standard migration recipe to migrate a module wouldn't qualify to be mentioned like the people who put real efforts into the modules... Also, remember that some modules are 80 lines modules abandoned by their original author. For such modules, adding 20 lines of tests might qualify to be mentioned as a contributor. So that's why I think module authors and projects PSC should have the final word what is a valuable contribution in module X or not and at some point it should be related to some kind of percentage of work. For instance, here are some unofficial rules we have been applying for OCA/l10n-brazil: there is no way you can be listed as a contributor if you did less than 2% of the work in some module (by the number of commits or percentage of code changed). There is no way you can be listed as a module author if you did less than 4x work than the module author that contributed the least. Also please never forget that in the OCA 20% of contributors do 80% of the work (or even 5% do 95% of the work). So what is important and what floats the boat is to make it nice for these 20% of people doing the hard work. The real quality ERP code won't be magically crowd sourced from millions of completely amateurish contributors. That might work to build Wordpress themes, but not for ERP code. My 2 cents. |
Beta Was this translation helpful? Give feedback.
-
Hi. could you give exemple ? I review regularly OCA PRs, and did faced such situation. |
Beta Was this translation helpful? Give feedback.
-
|
come to an ex colonized country to give it a try and see the kind of people who got the privilege at birth to have the opportunity to work with ERP software ;-) ... Here the guy who forced its way to the project PSC (unlike Renato, Marcel and Antonio), tried to fake expertise with yet another "fake review". Not all such fake reviews are easy to spot for non specialists, but here I'm sure even you can easily see that telling to do a try/catch in a loop to avoid an hasattr, is not serious OCA/l10n-brazil#3265 (comment) I could post dozens of such examples, but I didn't always make the effort to move them to English. There are also weekly examples of people doing absolutely trivial module migrations where they just applied a sequence the 5 steps and claim to be a contributor. May be not always with bad intentions, but it still happens on a weekly basis. |
Beta Was this translation helpful? Give feedback.
-
|
On one hand, I think in an OS community anyone working on anything should receive some recognition, as even porting the smallest commit from one version to another or fixing documentation is indeed contributing to the OCA and reducing technical debt for the whole community - so adding them as CONTRIBUTORS doesn't look to me like taking anything away from more substantial contributors. If we need to address the fact that not all contributions are equal, there could easily be two sections on the file "Major contributors" and "Minor contributors" with clear rules in the docs for whether an improvement should fall in one category or the other. On the other hand, I'd like to clarify that I didn't open this poll to contest whether my (dev's) contribution should have granted CONTRIBUTORS rights, but rather to address the difficulty in welcoming new contributors to OCA when so many rules on what should and should not be done are discretionary and not written anywhere. This is a hurdle for scaling, it's wasting everyone's time and it definitely does not help the experience of publishing code in OCA. I hope this can be taken in consideration and realize that even an imperfect guideline at times is better than no guideline at all. |
Beta Was this translation helpful? Give feedback.
-
|
@francesco-ooops there's already recognition ongoing through the GitHub statistics, where the number of PRs will be attributed by contributor/company. Right now, @etobella does them in an informal way, but I'm aware that from OCA, some officialism is going to be put in place. cc @vdewulf |
Beta Was this translation helpful? Give feedback.
-
|
That is great, as long as the contributor is aware of this :) |
Beta Was this translation helpful? Give feedback.
-
Well as long as there isn't such a system for "main contributors" it does unfortunately. The Odoo market is already largely what is called a market for lemon https://saylordotorg.github.io/text_introduction-to-economic-analysis/s19-01-market-for-lemons.html and this is in fact the main limiting factor for the Odoo ecosystem today. Suppose it's hard for the 5% doing the hard work to grow their companies or simply make a living contributing to the OCA, if you let people think even for a sec that in fact doing Odoo projects is so easy that each modules have dozens of such contributors, then you decrease the like-hood companies will really make the effort to hunt for the people who really make things happen. There is already such a world were only appearances matter, it's called the Odoo partner network, the Odoo award of the months thing, so there is no need to harm to OCA if some people want that IMHO. If you really want to show even minor contributions, you can always point to pages such as https://github.com/OCA/web/graphs/contributors where each commit is listed. Finally if we as a community raise the bar to be mentioned as a contributor, what we do is incentive occasional contributors to do a bit more than just trivial things and earn their contributor visibility by assuming a part of the real work load. I'm pretty sure it produce a better result than just pilling every commit in the CONTRIBUTORS.md section. Also bare in mind that what could take 8 hours for a beginner to migrate a trivial module, could eventually take only 20 minutes to an experienced OCA contributor. The PR's from the beginners will also easily suck 20 minutes to the experienced PSC doing the reviews... So at the end, the sum of the beginner work doesn't bring so much value per se sadly.
Well eventually, but will it be really simpler? We would then have authors, main contributors, minor contributors...
Then I would propose what we applied unofficially here in OCA/l10n-brazil:
In doubt ask module authors or PSC. |
Beta Was this translation helpful? Give feedback.
-
|
Leaving this link here as food for thought https://allcontributors.org/ (I haven't personally tried the tool, but I like the approach at least) The discussion so far is very interesting, but it's focused almost exclusively on technical contributions. I think writing documentation can still be extremely valuable and should be encouraged and recognized as meaningful contribution. |
Beta Was this translation helpful? Give feedback.
-
|
I agree a guideline is needed. IMHO we should stop thinking about what is a "meaningful contribution" and start thinking of "helpful contributions". In our context, we could define "helpful contribution" as anything that makes something better than it was before, or helps others doing so. That is a simple guideline that lifts everyone from the heavy burden of judging others or being judged by others. It can apply to many actions apart from code, and I think that is a good thing. |
Beta Was this translation helpful? Give feedback.
-
|
That's the kind of CONTRIBUTORS.rst pollution we want to avoid. That contribution is already in the GitHub statistics. For the module itself, we don't want to see 50 contributors in the README. @rvalyi's numbers may be a rule of thumb for being there. |
Beta Was this translation helpful? Give feedback.
-
again, I encourage everyone to think about this thing called "market for lemon": https://saylordotorg.github.io/text_introduction-to-economic-analysis/s19-01-market-for-lemons.html Basically it means in a modern economy market where the "customer" has little information about the transaction quality (hours of Odoo consulting here), the market tends to collapse and George Akerlof explained how with simple maths (and got the Nobel prize in economic sciences for it). We, OCA contributors, normally understand that it's already extremely difficult to sustain a business while actively contributing to the OCA, right? Is everybody here aware what share of the OCA work is done by @pedrobaeza for instance? I'll put it short: if @pedrobaeza were to stop his OCA contributions we are all dead I think. If Akretion stops his contribution to OCA/l10n-brazil (the largest OCA repo) the project is dead as well, so this is no joke... People attention is limited (attention economy anyone?) and by making it so that the modules CONTRIBUTORS.md becomes a mud of 50 pointless contributions what we are doing is killing the relevant information for the user/customer. We are making it difficult or impossible for him to spot who really is doing the work. By doing so, we are increasing the lack of information while the solution to the "market for lemon" is exactly the contrary: it is "labelling". By increasing the market for lemon problem, we may create very serious threat that key contributors feel their exceptional contributions will be invisibilized in the noise of opportunitisc companies fighting to put their names along once the hard work in over. If these guys move on, the OCA will be in serious trouble. IMHO it's already extremely hard to attract new talent to contribute and we better go the opposite direction of labbeling more carefully instead. My 2 cents. |
Beta Was this translation helpful? Give feedback.
-
|
I've already read the market for lemons thing about 5 times. It's quite hard to understand. George Akerlof might have been a economics genius, but he was awful at making comparisons (at which point a car becomes a lemon? 😆). Well, nevermind. In any case, we are not selling anything here. Not cars, not lemons, not code... We are a FLOSS community. Our goal is to grow and survive. Yes, we have a few big contributors, but... is that good? IMHO that's unsustainable by definition. One person can't grow more than 24h a day (in case they can sleep and eat while contributing). There are 321 pages of modules available right now. We simply need (and want) more contributors. Your points seem to care more about giving out some trophys 🏆 to the best contributors. That's a good idea! But that's another story. Odoo already does that FWIW. IMHO we shouldn't stop calling contributor a contributor, or stop recognizing contributions, just because a few members work a lot. Maybe, if we make efforts to stop making others feel as "pollution" or "a mud of pointless contributions", we'd have more contributors. FWIW Git only shows code contributions. But, remember, many helpful contributions are not code. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you @yajo, this is not just a matter of guidelines in the CONTRIBUTORS file but also the reflection on how OCA plans to welcome new... contributors. Which, as far as I remember, is supposed to be one of the main working areas for this year from the Board? It would be nice to have some of them chip-in. |
Beta Was this translation helpful? Give feedback.
-
It's a WIP, yes. Spoiler alert: hopefully we'll be able to announce something at OCA days :) |
Beta Was this translation helpful? Give feedback.
-
|
I agree that
(from https://github.com/orgs/OCA/discussions/185#discussioncomment-10657119) so the author deserves a spot in CONTRIBUTORS. In order to show how much a contributor has done, CONTRIBUTORS is not the right place in my opinion because it implies someone deciding whether the contribution is worth of going there, so it becomes subjective. TL;DR: CONTRIBUTORS is just a list of anyone who contributed, how much can be retrieved from GitHub statistics. P.S. Can someone vote in the poll? |
Beta Was this translation helpful? Give feedback.
-
|
@pedrobaeza do we already have a dashboard with stats for a specific module across versions? |
Beta Was this translation helpful? Give feedback.
-
That would be one way, but I think it's nice to see directly in the module that many people contributed to it: it gives more the idea of a common/community effort. |
Beta Was this translation helpful? Give feedback.
-
|
I think we might do both: keep the CONTRIBUTORS.md the way it is with its subjective evaluation of what is relevant and some automated git stat (per module) somewhere below (to be developed). Again, in the case of the OCA very few people/companies are doing a lot of work.Some beginners might try to help, but unfortunately they don't bring so much net help as they it takes a lot of energy to skilled maintainers to review bad PRs and in the end they would often have done a better job and quicker if they did it themselves. Recently, we got some bad news from a company that was once leading contribution as might be doing Tecnativa these days, So basically, like very often, the risk is the overworking main contributors don't get enough return for their incredible work and move on, just like so many open source projects use to die. I have been in the Odoo ecosystem since it's beginning since 2007 and I have seen the sad evolution where the initial Odoo oficial partners were very skilled and with good ethics and that is now transformed into a dystopia circus of both very poor skills and very poor ethics. The finance is destroying the project: the money is going to the greediest bastards and very hardly to the people that actually build the open source modules. There are different open source business models. But something that used to work for Odoo was that you could spend on R&D rather than marketing (say like oficial Odoo partners) and get enough visibility, at least in some niche market, so it might be working. It's pretty much the norm in open source project that project authors have the control over who they let in and who they would delegate responsibilities or give credit for their contributions. Having a technocratic entity to take over that and take away this power from the people actually building the software is not good at all IMHO. Yes, evaluating contributions is subjective. But who better than the modules authors (or at least the maintainer when the authors left) can evaluate what was relevant and what was not?? This subjective evaluation is a very important information, as I explained this is very much the solution to the market for lemon issue the new Odoo SA closed source model created. We should absolutely value this subjective information, it is pretty much like:
Finally, don't forget that:
|
Beta Was this translation helpful? Give feedback.
-
|
Thanks Raphäel for those focused words. Indeed, I have already felt discouraged for some of the small-medium contributors adding their names to contributors to "get traction" in every small PR. |
Beta Was this translation helpful? Give feedback.
-
|
I might also add that Akretion ranks between the 3rd or 4th OCA module contributor worldwide. And our entire business was done since day one by putting it on the R&D and getting visibility for it rather than smoking money on marketing. I'll now give a concrete example: Akretion main project OCA/l10n-brazil has nearly 14 000 commits, 3500 PRs, soon 150k lines of code with 90% test coverage. So just this repo is about 10% of Odoo Community codebase alone, it's the largest OCA repo. The main module l10n_br_fiscal has over 20k lines of Python+XML, that is the largest OCA module and it's a huge fight to keep its scope under control and keep extracting new modules. As I gave concrete real life example we have seen real hickack attempts from greedy companies Odoo would of course promote for some time as a Gold partner (as Odoo is unable to address the market despite their investment and also like to dive and conquer). Curiously despite very aggressive attemps to take over the localization, these companies didn't manage to push any modules in other OCA repos in 10 years... So I mean over these 15 years we invested a lot of energy into this by playing by the open source rules when we could have had a corporate proprietary software well paying job instead... Many of the people who did not spent years building these modules were actually making good money in the proprietary software industry meanwhile and are now trying to get a portion of the cake now that we made the effort to make it mature enough... It's part of the game they use the code for free while enforcing the license. It's not part of the game they highjack the project reputation by just getting a few trainees do a few superficial contributions (often bloating modules with bad code) as they often try... In the trolling examples/links I gave earlier, I had to spend like 3 weeks of work for free to prevent the guys from starting to reinvent the IAS2/anglo saxon Odoo feature inside the localization just because the guys didn't know it and wanted to fake expertise (I gave a more obvious example of fake review recently). So I mean these a real life highjacking attempts that consume A LOT of energy from us and if the OCA doesn't making clear that people who did the software are the ones who decide then it's over for companies like us... In the mean time, we actually welcomed two new PSC from different companies in the last 3 years, at least 4 different companies got their modules in the project. In a 20k lines done over 15 years module like l10n_br_fiscal, it wouldn't be fair that doing even 500 lines contributions would let you be listed as a module author. So we have been picky with the contributions and people are listed here if they contributed at least 5% of these 20k lines if code (7 people from 4 companies). The whole point is it would be very unfair for them if suddenly every company putting many trainees to do 50 lines contributions (as fiscal rules evolve) would bloat the module README and making it look their company contributed a lot more than these extremely hard working contributors who really made the difference and are still the one enforcing the modules quality today. At the same time, easily 95% of companies trying to implement Odoo in Brazil FAIL miserably, most of the time with an official partner of the year (the average lifetime of an Odoo partner in Brazil is no more than 2 years). At the same time, the market is not paying enough for the companies doing the R&D to hire and train new people. All contributors are actually investing... So I mean the "market for lemon" problem is very real. If instead of the Odoo official partner dystopian circus it was very clear the challenge to run Odoo in Brazil is extremely high and companies should do whatever it takes to work with the modules authors, then the real contributors would make enough money to grow and the whole market would grow. Or else, we might just also go from 95% failures rates to 100% if the big contributors give it up. And if you don't like it, you can actually do what open source enables: fork or create a different module or different localization project. We got 2 Odoo gold partners give it a try by starting localization forks 10 years ago... Well one abandoned its fork 3 years later and another just died 10 years later. I'll give my subjective opinion that their code was absolute shit that would never make it to any OCA repo (I can pass you the links in private if you need to see it). And as for the accusations of not letting 3rd part contributions enter, well a quick look in their own localizations easily show they did not accept any contributions, unlike us in the OCA... So big bullshit. But as these companies spent millions in trying to build such localizations to take over the SaaS market, you can be sure they would have spend quite a lot trying to fool people at the OCA (they did) or trying to highjack the modules credits if we give them a chance to do so. So again, this is not science fiction, this is real commercial higjack attempts sucking nearly all our energy way and we need the OCA to stand with the people doing serious open source, not giving credits to these opportunistic companies. Yes I understand many of the people debating here are actually well intentioned people. But remember the prisoner dilemma: in a game with lots of cooperation, it makes it easy for the bad guys to take advantage of the game. so we need to OCA to protect the people actually doing the modules code, bad guys do exist, the world is wild. |
Beta Was this translation helpful? Give feedback.
-
|
I'd like to point out that the "contribution" is not limited to the individual who writes the code. I think the PM/PO/analyst/functional who identifies a bug/improvement and often designs the solution to be implemented is contributing to the module and the community. |
Beta Was this translation helpful? Give feedback.
-
@vdewulf I can sit with you and others to see which tool is better during OCA days. |
Beta Was this translation helpful? Give feedback.
-
@rousseldenis : We will have a workshop Monday afternoon to discussed and work on ideas. You are very welcome! |
Beta Was this translation helpful? Give feedback.
-
|
@pedrobaeza please notice I haven't mentioned documentation at all as I rather meant all the work related to feature request, bugspotting, bugreporting, functional testing, solution engineering (and sure, also functional documentation as most devs don't bother much about the functional side of things) |
Beta Was this translation helpful? Give feedback.
-
|
I also do feature design, and my team includes me as contributor directly on the PR, so I don't think that's a problem. It's also true that a PR with 20 lines of code adding 2 or 3 contributors has already happened in OCA repositories, which seems also unfair. |
Beta Was this translation helpful? Give feedback.
-
BTW a good one to start with may be OCA/mis-builder#635 ? I'm pretty sure a solid mis-builder documentation would be welcomed as a mentioned contribution and you can probably ask them before starting. cc @vdewulf |
Beta Was this translation helpful? Give feedback.
-
|
Copyright issues are primarily governed by law, with ownership typically assigned to the client funding the work, or the entity bearing the costs. The OCA’s role is to manage what appears under authors and contributors, but copyright laws remain beyond the community’s control. When considering contributor recognition, it's crucial to remember that much of what’s discussed here is subjective opinion, not universal fact. Often, rare cases—representing only 5% of situations—become the primary focus of debate, diverting attention from broader challenges. Discussions around fairness often lead to dead ends, as what feels fair or unfair is subjective. Instead, we should focus on learning from larger open-source communities and applying those lessons. The OCA is relatively small, and the real objective should be organic growth, not individual accolades. Although functional reviewers haven’t been recognized, the deeper issue lies in a lack of community expansion. The OCA hasn’t seen growth in contributors for years, and contributors have left for other projects. In light of this, debating contributor recognition seems shortsighted. The more important question is: how do we build a community that fosters long-term, sustainable participation? In a space where contributions are often motivated by financial rewards or recognition, fostering a culture of humility and collaboration is vital. We need to create an environment that motivates participation and attracts new contributors—focusing less on personal gain and more on the collective good.
When doing so results in growing the number of contributors, so we can count them in the thousands, not just the hundreds. |
Beta Was this translation helpful? Give feedback.
-
|
While it is important to grow the number of contributors, it is also just as much important to make it an attractive environment for the few massive contributors. Because whether you like it or not, these massive contributors do 90% of the work in the OCA (just look at the commit distribution in key repos). So suppose we mutiply by 10 the number of very occasional contributors that today do less than 1% of the work. If we do so at the cost of creating more dumb review burden and making it unattractive in term of visibility for the massive contributors doing the 90% of the work, it is not hard to understand this is a critical absolute net loss for the project... Also you guys should really have a look at this "market for lemon" thing. The claim to make it "democratic" so everyone can participate is pretty much what did Odoo with their partnership already. 15 years ago, only people with alien ERP and coding skills would dare to try implementing OpenERP. And now, even the stupidiest greediest guys are encouraged to buy their lottery ticket, pardon, their Odoo partnership. And while it may boost the sales of the license business for Odoo in the short term, we all know the disaster it creates in the Odoo market... Skilled ERP integrators don't even want to fight anymore to be listed among the "top" Odoo partners, this is very much how the "market for lemon" collapses a market. I would even dare to say Odoo does a bit like some of these "modern" politicians: they orchestrate a chaos in the hope they will be the very ones surfing atop of it while no competitor would be able to emerge because nothing would be rational anymore. But IMHO this is not what we should do in the OCA. Again, remember when CamptoCamp guys had a central contributing role like Tecnativa has today? Why are these great contributors not as active as before? Do we really want the other skill contributors to move on or contribute less the same way? If not then we should let the massive contributors shape it the way it works the best for them and have the humility to acknowledge we aren't big contributors enough to decide things when we are not... In the recent years the number of merged PRs in the OCA grew at the same pace that the merged PR in Odoo Enterprise. The number of migrated OCA modules after 1 year the release is also increasing steadily. Certainly the crazy refactors during the VC money period between 2015 and 2020 left a lot of people behind. However since v13 the changes are less crazy to follow. Also the trend is we will have porting bots and possibly more AI so people will also become more productive. So I'm not too pessimistic about the OCA growth. One thing is Odoo doing marketing like crazy and another thing is the sustainable growth the OCA can achieve with a real working projects basis and without creating burnout or unattractive conditions for the massive contributors we all rely on so much... |
Beta Was this translation helpful? Give feedback.
-
|
IMO The real point is: what do we value?
@pedrobaeza once upon a time you told me that even for a migration where you changed nothing (just the major version number) you should be listed into contributors. Motivation: very likely you analyzed the code, tested on a new version, etc. On this particular case, cherry-picking commits, I see the same value. Why? Because when you cherry-pick a fix or an improvement your doing someone else's job: checking the code, adapt it if needed (eg: conflicts), test it on the new/old version, etc, and very likely you tested it on your customer project. As an author this saves me time. As a maintainer, this saves me time. As users of this huge bucket of code, we all save time. As @SirAionTech pointed out
Now we might argue that some contributions shouldn't be "mixed up" with maintainers work or long time contributors, bla bla. So the main question here is: how can we make everybody feel accepted and recognized? I've recently stumbled upon the requests lib repo where do this on the authors file:
and they also give recognition to new contributors on the release notes
I think we should tend to something like that. From my POV it gives more recognition to authors and maintainers while still giving recognition to smaller or random contributions. My $0.02 |
Beta Was this translation helpful? Give feedback.
-
|
I keep saying that a migration work is valuable, even if no code changes, as it involves usually all that work. And when not, it can be block by PSCs or maintainers. I have seen lots of incorrect migrations that don't work at all (because the contributor didn't do their job and only change the version number without event testing it for real). For cherry-picks, this work is more debatable, specially if it's going to disturb the visibility of real contributors. We are debating here about setting some guidelines as general rules, still allowing subjective moderation to the PSCs/maintainers, and avoiding to prostitute the environment on interested low-quality (or simply small) contributions as a way to get traction. |
Beta Was this translation helpful? Give feedback.
-
Agree 100% with that. I think behind PSC and maintainers are humans. That's the value added to OCA. You cannot automate or establish too much constraining rules if we want to keep our flexibility too. I think as @simahawk said, newcomers should be welcomed and congratulated too. Maybe a tool should be great in order to establish such a report in github pages on organization level/repo level/module level. And with social features (tweet/website) to congrats them. @etobella |
Beta Was this translation helpful? Give feedback.
-
|
I have been reading all the opinions and I would like to contribute with my point of view to this debate. Far from being presumptuous, what I’d like to say is that with perseverance, effort, dedication, and the investment of resources and time, results eventually come in the form of the “right to subjectively decide” what is correct and what is not, in the repositories where we are PSC. To summarize:
I believe regular contributors have earned this "right to decide", and it’s been through:
Everything can be debated, but realistically, 90% of the OCA is being maintained and improved by a very small number of partners, not by a large number of occasional/new contributors. It is on the major contributors (and many of us are also Sponsors) where we should focus. We don’t need to cater to the occasional contributor (who also will never be a Sponsor) instead of keeping the regular contributors motivated. If another PSC thinks adding me (or my team) as a contributor is too much (it has happened to us, as well), what’s wrong with that? They’re not telling me this to “annoy” me, but because in their judgment (in a repository they primarily maintain for free) they believe it’s not the most appropriate. I don’t think it’s wrong for someone who has dedicated (and continues to dedicate) so much effort, to have more "power" than the occasional contributors and to be able to decide according to their own criteria, as long as there are valid arguments. What I would do:
My two cents. |
Beta Was this translation helpful? Give feedback.
-
|
a few more comments: any pure "objective" criteria is going to be abused. Even cherry picks will be largely automated in the future. I gave real life examples that sucked so much of our energy that we nearly gave up on the largest OCA project. The more importance the OCA will take and the more people with zero open source mindeset are going to be kicked from their ephemeral Odoo partnership, the more pressure we will see to abuse easy "objective" criteria. So IMHO, like in the real life, we can have rules of the thumb that can be a be the "laws/guidelines" and judges. An no judge is going to be more legitimate than the modules authors or eventually their accepted maintainers. This is largely how open source use to work and self regulates. You don't like it? fork it and do better, that is part of the game and this is why open source use to have so much quality. Normally modules authors and maintainers are no idiot and are perfectly open to mention new contributors and welcome new maintainers if they evaluate it's a net benefit for the modules/projects. For instance over the last 4 years I spent countless hours to help new contributors in OCA/l10n-brazil and promote two of them as the new PSC. But it's an exchange, they had to earn it and take a real part of the workfload on their shoulders over several years to prove they won't be just new leeches (had we simply given the key to other leeches that were pressuring 10 years ago, these would have screwed the project just as they screwed their fork localizations...) BTW remember the how the xz lib in openssh was pressured from the ground up to force a new maintainer into the project that nearly lead to the biggest backdoor ever been injected into all the openssh servers? Bad guys do exist and can invest a good amount of energy to fool people. I gave real life examples of people doing fake reviews to fake expertise and gain power by fooling newcomers in the Brazilian localization. Finally, with all that said: I think it will be great to have a kind of widget/table listing all modules committers automatically so small contribs are still mentioned. As part of the current effort to map the most used Odoo modules, we could also identify "orphaned" modules where their original authors and maintainers seems to not care any more. It's probably the case of half of the OCA modules or more... And in this case, we could work out new maintainers and eventually encourage more automatically small contributions. This is quite different than very large modules with lof maintenance where their author might legitimately expect a true visibility return for their incredible efforts or else we will be a community exploiting these developpers and the good ones will move away and others will simply stop joining. |
Beta Was this translation helpful? Give feedback.
-
Hi,
to try as much as possible to stay clear of grey areas, here's a question for the community (with the end goal of clarifying guidelines):
13 votes ·
Beta Was this translation helpful? Give feedback.
All reactions