The BDFL software model; pros vs. cons, is it worth continuing for Blender?

Yep so many things some would like to be in Fusion 360 with valid reasons but it would speak against what Fusion should be …
But otherwise they can listen as well.

Honestly, I don’t think that was ever in question. BDFL’s are only human and humans are prone to error. It happens. The issue, I think, is how said mistakes are handled going forward.

Does the BDFL pay attention to the community when they feel they’re making the wrong decision or do they explicitly state they’ll be ignoring it going forward until it starts seeing things their way?

Does the BDFL allow the community access to features they’ve been asking for (& implementing) for ten years or does the BDFL prevent it’s addition to the official distro until he gets everything perfect… whilst applying a far more lax standard to features he’s been promoting for a year?

When faced with a community that can clearly see when the BDFL has influence/control over many parts of the project, does the BDFL concede that their position entails (even entitles them to) such power… or do they claim they have no power at all?

It is, as is generally the case for me when it comes to social interaction, a matter of respect on both sides. It is hard to respect someone that treats you like an idiot and, when a BDFL starts doing that to the community, it is hard for the community to pay them the respect the BDFL needs for the continued success of the project.

I suppose that would work but that kind of requires the BDFL not to be that project admin himself/herself. :wink:

Its the “for life” bit that bothers me.
What if someone accidentally drops a piano out of a window on Tons head? What then eh? We need an heir apparent. Has he got kids? Is there a clear line of succession? Are they old enough to take the crown? If not would a Regent need to be appointed ? These are pressing questions. Why developers are wasting time improving the code when these questions remain unanswered is the key issue.

Since this thread touches parts of the discussion that happened some months ago in the BA thread ‘Blender over 10 years… what are your thoughts ?’ - I think I will repost some of my suggestions from back then.
About the communication and synergy between BDFL and the community - between the cathedral and the bazaar you may say.

Repost from ‘Blender over 10 years… what are your thoughts ?’ :

Better feedback from the community - concerning which features that are important/popular from the user perspective :

I think that Arnywar made a good suggestion here :

So my suggestion would be that the Blender Foundation has a public roadmap, were the users can chip in and show collectively what they want the most. It could be a extended roadmap with 3 categories :

  • A category of development targets made by BF
  • A category of features that are popular on a user request list. Features not that hard to implement with the current BF developer team.
  • A category of features (from both BF and the user request list) that will take planning and time for BF to develop. But BF would like to know what are popular in the userbase anyway. To better prioritise and plan ahead.

Blender Foundation can even choose to put some of the popular features on a crowdfunding page - something I go into detail with in this earlier post.

Anyway, the important thing is that the ‘hub’ for all this user feedback is on Blender.org. Since - in the end - it will be BF that says OK to the final code.

A note about user request list : It could look a bit like 3ds Max’s General Feature Requests feedback page.
And the feature request page - and user roadmap voting - could maybe collect 2 different vote numbers : For people with a Blender ID (that are donating) and all the other users. Having 2 vote numbers I think would make it easier to spot if a big group of people who are not really using Blender are upvoting a request (like : implement a Maya/3ds Max UI).

Better communication between BF and the community :

From a recent post by BTolputt :

And from a post from him some time ago :

My take on it :

I have for some time looked with envy on how a site like Slashdotis making user based interviews. Slashdot have a pretty elaborated moderation/user vote system as far as I can tell - to make good user feedback flow to the top (look at Moderation and Metamoderation in their FAQ). In a interview setup this would then be good/popular questions that flow to the top. And be asked to the interview subject by email in the end.

Here is a short FAQ about the Slashdot-interview format. And to give a example - here is the question process for a Bruce Perens interview. And the final interview.
(One thing to notice here, is how thorough the whole process is. So nothing like the quick format of a Reddit AMA).

So here is what I would suggest :

Maybe Blender Artist (just to mention a popular independent gathering of Blender users) could be the ‘hub’ for making a ‘Ask me 10 questions’ to Ton (or whoever is leading BF down the road) on a regular basis. Maybe every year ?

The problem is of course that BA don’t have a system to submit and vote for questions. But maybe there is a service out there on the Internet that can be used (do anyone know of something that could be used ?).

Personally I just think it could be so useful, to see the 10 most popular questions (from BA users) being answered by Ton in a thorough way.
It could (as mentioned) be done every year. Although I agree with BTolputt, that a general presence in the community should be more often.

But it would be a good start.

If he was more present here, he would need to waste his time in endless debates.

If anyone here has questions regarding running or operation of the Blender Foundation please ask directly on the mailing lists.

About the foundation:
foundation at blender.org

About development:
http://lists.blender.org/mailman/listinfo/bf-committers

Some of you are acting like Ton is some unreachable person. He’s on IRC literally every day. It’s not hard to interact with him.

Maybe its that discussing being hit-by-a-buss is a bit awkward as a first topic of conversation to someone you never met, Compared with wild speculation on forums, which is far more entertaining.

Bingo. I’d be embarrassed for my first interaction with someone to be the equivalent of a witch hunt, too.

LOL :smiley:
I’ve always wondered the same thing. but naah, i don’t think that’s gonna be a real issue. Ton’s indestructible :smiley:

Yeah his spider sense even allows him +5 to dodging falling ceiling tiles.

Again I want to see everything just the way it is, with a system to vote for secondary projects with money.

Blender is free,

Blender growing is not, we have to feed the poor thing.

Its like a cute lil’ kitten.

With Blender being one of the most successful open source projects (IMO), I’d say the development and organizational model for Blender is pretty much spot on. The BF, Ton and all the contributors are doing a great job. I trust them to continue doing so.

Aside from that, well, you can’t please everybody all of the time, we’re all different, that’s just how we humans function. There’s certainly no need for any drama on the user side.

Personally I prefer one person with veto rights who has a clear vision instead of a committee where everybody just wants to push his/her own wishes and agendas, even if that means my own wishes might get postponed or not implemented at all.

Easy for people to criticize the leadership, but hard for them to offer any type of assistance.

Blender is dead today without strong, current leadership. Maintaining software development for long periods of time is no picnic. I’ve been in many 3d forums. Many of them no longer exist and the 3d app they’re based on are dead in terms of development, TrueSpace, Silo, Hexagon, Carrara, Amapi… Modo is no longer owned by the original developers. XSI, I heard will be scrapped… Lightwave devs attempted to develop a new one from scratch, but it was discarded eventually…

And there are so many examples, not only in the commercial world, but in the open source. It’s chaotic. Man, how many Linux distros that I have to choose from? None of them can stll match the beauty, elegance and ease of use of the mainstream OSes that people came to love and stick with.

I know that the BF has listened to the community before, but the quality of which the community feedback is followed can vary.

In some areas like texture painting/dyntopo, the extent at which community requests made it into Blender is very good. Meanwhile, suggestions for other areas like the interface has had a history of not being followed near as much (considering that we saw almost no changes there from between 2.59 and 2.7 despite the massive wealth of mockups created by the community).

It’s not like these community requests require major changes to begin with either, they’re followed by the devs. or not depending on which area of Blender it concerns. I do agree though that major things like the depsgraph refactor should take priority, because they resolve fundamental issues in Blender that actually prevent the inclusion of powerful tools in certain areas.

I would also disagree with the idea of firing a developer and replacing him with a community manager, with all the things needed for Blender yet to be competitive, it needs all the developers it can get.

The BDFL model is a seductive shortcut in the short term, a liability in the long one. It is not just the case of Blender, the situation extends to other projects, like the egregious snafu that GvR (Guido van Rossum) has inflicted on Python community with version 3.

What works in the long term is something like the Apache Software Foundation or some firm-steered process like Java Community Process/Eclipse Foundation.

B.t.w., it is an amusing irony that a community of rabid individualists like the FOSS fans should worship the idea of complete and relentless dictatorship.

For Blender, my open eyes dream is Autodesk grip becoming so strangling on the 3D world that a group of important 3D shops band together, create a foundation, fork Blender and turn it into an highly professional tool.

In my opinion however, for OSS especially, a directional even if sometimes mistaken dictator is far better than a arguing, indecisive community that will end up wanting everything for itself (but in different areas).

I think a up to date users manual would help a lot.

Without consumption Blender is a private activity of the Blender Foundation, whatever that may be, which it is sharing with others.

If he was present and listening to the community more, there would be less “endless” debates. Seems like a win-win for him to join us here. He gets a better insight into the community, how it uses Blender, how it wants to use Blender, and why some of his decisions becoming much more important than he thinks they should be. The community gets less “endless” debates because the one person that can actually resolve them does so.

Also, remember, Ton can ignore the endless debate threads and simply read/interact with the community in the threads that provide value for him. Kind of like you can :wink:

I’ve already addressed this in the past. Communication is a two way street and, believe it or not, a large number of people like forums because the conversation is threaded by discussion and (more importantly) can be conducted without having to keep to Ton’s hours on the other side of the planet.

If Ton wants support from the community, wants the community to pay money towards his desired projects, etc - he needs to be developing a relationship with them that exists prior to & longer than a crowd-funding campaign. People aren’t just machines that send praise & money whenever BFI gets around to their next open film project. As the crowd-funding campaign’s success (or lack thereof) should have clued everyone in on.

Yeah, it might suck that the first interaction you have with someone comes from what you think is a “witch hunt”. On the other hand, if you had been paying attention to community discussions prior, it’s quite likely you’d not have made the decision that caused that unpleasant interaction in the first place (or, at least, have had a better response in place when people called you on it). :wink:

I’ve been in forums where the devs interact with forum members. There was also one where members had never-ending arguments with the CEO. Where is the 3d software now? It’s sold to Microsoft and it’s dead. I also remember Modo devs being active in the community forums. They eventually stopped posting. Every user has his own idea of what features should be prioritized. They’re like kids constantly asking for toys from their daddy. :smiley: Like I said, without strong leadership and creativity to produce funds for development, Blender would have been dead by now. :smiley:

Yikes, how did I miss this thread?

Couple of points/opinions come to mind.

  1. The BDFL model works fairly well with a commercial element to it…why? Businesses can fail and they can succeed. The mentality behind them is success, thus they look at market trends, conventions, competition. So what happens when you put that in an OSS environment and there is no interest to compete, no commercial necessity or rule set to attract buyers, no immersion into the larger field as a whole? In short the motivation and incentive isnt there.

  2. As far as Ton goes, him and some of those he surrounds himself with come across as ideologues to some extent. When you create an environment where that kind of mentality comes first, a culture kind of forms which reinforces it. I think thats in part why there is this big separation between the userbase and the development end (with a few exceptions). Does Ton or the development end look to other software regardless of commercial or OSS status? Do they follow the industry and field as a whole or is their world only about Blender and OSS? On the commercial end, the development is more often than naught immersed deeply in the field as a whole and the artist in that field, not just stuff tied to what they work on. Due to ideologues running the show, Blender’s internal culture seems to push for being different or outside the field rather than inside it working alongside existing 3d and 2d packages regardless of licensing.

  3. Touching on the “community interaction” mentioned earlier in the thread, its already been established that (via irc and other channels) that there is disdain from some on the BDFL side, including Ton himself, towards the Blender community and the BA forums. When this kind of sentiment exists, writing its userbase off as “noise”, theres bound to be a conflict of interest or even trust in the direction of the software. User concerns can easily be written off if the source is seen as “noise” and or the peons on the forums. The fact this exists is quite relevant in relation to this subject.

  4. My personal opinion is that the BDFL does not work with Blender when A) There is little trust or consistency from and for the BDFL, B) They are not immersed in the field as a whole, even mock commercial and professional interests, C) Put ideology and feelings over what the actual field needs and expects.

What I think would work better is some kind of set up in which the power structure isnt tied to one party or person, but rather different groups who have some kind of say or vote for moving forward, and or there exist different branches that can act independently of one another. For example I see CG Cookie as an organization and business that understands the field far better, in part because it covers the field of CG, not just Blender. If they some development and decision making capacity outside of Ton, that would create a checks and balances of sorts. The idea is to get development/decision making via teams, each perhaps with their own culture and environment which influences and or directs their results. The userbase will naturally start going with what works best and hits on their needs.

Finally, my personal belief is that Ton makes a better producer than a developer. In other words, he should still be part of Blender but more on the end that helps start projects (open movies for example), generate development revenue and so forth. He seems to get really “excited” and motivated whenever it comes to producing open movies, which makes me think he wants to be a open source film producer…and thats fine…in fact its great, but it also shouldnt conflict with or dictate the development end. It would need to be separate… and this brings me to the last point.

Back when we discussed Gooseberry in detail… and emotions ran high. The userbase was targeted for not jumping on board entirely… and thats important for a couple of reasons. The community wasnt entirely excited about it, it showed in the numbers, yet the BF didnt seek to see what the Community wanted or expected, instead blamed them for not liking their project. There is something fundamentally wrong about that when it also relies on the community to fund it. If you want funding, you go with what the masses want, not what you want nor blame them for being lukewarm about it. This is the BDFL culture we have to deal with, it reveals some serious conflicts of interest and well…management.

Going the further discussion found in the colored wireframes thread… a “told ya so” proclamation can be made by quite a few blender artist forum members. The concern and even distaste for the project was that it and I quote “hijacks the priority of development” to focus on a specific field and the interest of the BDFL’s pet project… and thats how we start to see it…as a “pet project”. If films are the pet project that hijack or take development priority for the time frame of the project (usually quite long), there will continue to be a conflict of interest between the user base (many of which are not into film or making cg shorts) and the BDFL’s personal development team. In other words, while development can still occur (that everyone benefits from), it does seem to put a wrench in the wider development pipeline as well as the direction it goes from that point out.

Given all that, I am not a fan of this current BDFL set up with Blender as I believe ideology, closed environment and niche mentality get in the way of really speeding ahead and getting both the low and high hanging fruit which is needed by actual users of the software and or what the larger CG field expects. I would rather see a Fork/Spork that cares about the larger cg field than its own little bubble world. Not everyone will agree with this of course, and theres nothing wrong with that. This is just my personal stance on the subject.