This is a slight break in topic, but I’m finally taking a few steps towards taking up the dev shortage- I just submitted my first bugfix pull request to @giovanni.ghisleni . I squashed a rather annoying bug in the Rene sequencer, and it’s much more enjoyable to use now. It’s not much, but I’m still proud of it nonetheless.
I’m definitely not even close to releasing a working module, but hopefully I’ll get there someday.
In that the Bene sequencer?
I maintain a good chunk of open source software and few things bring me more joy in that activity than a bug report accompanied by a pull request. Don’t short change your work! It’s awesome and the first step is always to take the first step 
I admit that I skimmed over the previous comments, but I’d like to second this:
I lot of us miss you Aria! 
** UPDATE: I personally miss you. 
**
Typo, my mistake. Bene would use “outdated” matrix data for the trigger outputs, meaning it wouldn’t respond to tweaks until the next cycle. It has some slightly unorthodox code (doesn’t utilize Schmitt triggers for gate generation) so it took a while to figure out how to fix it, but I somehow got it to work. The gate and v/oct outputs would update fine, but only the trigger sequence wouldn’t update.
That’s great. It is probably a good idea to confirm that the developer accepts pull requests before putting a lot of time and effort into chasing down a bug and fixing it. I do not accept PR for my plugin. But that is just part of my personal security strategy. I do respond to Github issues, including suggestions as to where the problem lies and how to fix. I’m not sure how common my approach is amongst developers.
Most active projects fall into one of three categories as far as I’ve seen
It really seems to amount to whether or not you want a group of devs in your code or not; and how you want to manage the tail of support. I don’t think there’s any good or bad about any of them. We run surge in camp 3.
I would be mortified, obviously. It’s libre software so others can hack it, learn form it, create variants. Not because I sought to make it easier to take me out of the picture. When I quit, Andrew Belt made it clear that it’s all business, we’re not here to make friends around here, I remember he told me to cover my own bases. I covered them by quitting: it would be extremely hard to find a developer who is both willing to start an hostile fork, and who has the skills, taste, and dedication required to uphold my quality standards.
Mog’s a long time friend, someone similar to me in many ways, and a very private person without much of a public internet presence. I just wish to set the record straight on this: they didn’t quit in reaction to what I said. They had actually quit before me, for the same reasons. Like most people who quit, they did so silently at first.
I think you’re accidentally making the point I was trying to make far better than I did: you see a developer state that the current version has severe UI issues, yet you will tell her you have the skills and the credentials to second-guess what she said about her own software!
A simple visual comparison with v1 will surface many issues (you don’t even need to run it, you can cross-reference with the docs). The specular highlight layer on the jacks is entirely missing. The status LEDs are offset from their sockets noticeably, and when dimmed, their light blending mode makes them too desaturated. Backlit knobs lack their two-tone bezel, and their tick should be pink with a blue outline instead of black (which is a legacy codepath for Lights Off support, that no longer serves any purpose). The exact colors, along with the rest of conventions the design system and provisions made for colorblind accessibility, are documented in doc/design.md.
You tell me “Job’s done”? I tell you I would be ashamed of releasing UI in this state. No job has been done, a minimal compatibility hack has been applied. That’s not meant as a slight against Falk, it’s just a factual description of what is offered.
That’s what VCV loses by treating developers as disposable: their software, in turn, becomes treated as disposable.
I certainly would love to, too. I can’t help but feel great regret looking back at what I released: I quit before I got to get started. Starting this without experience in C++, audio development, or owning any modular hardware, made me face an uphill battle. But by the time of my last release, I no longer felt limited by my C++ skills. Were VCV a different environment, I would have achieved so much in it.
But we all know what the problem is, don’t we? It doesn’t matter one bit how nice we might be to each other. The Code of Conduct doesn’t apply to the one person who makes it miserable. And after my participations were erased, it was changed specifically to forbid such discussions.
In fact, the only reason I checked out the community a few days ago is that I received yet another private message informing me someone chose to discontinue their involvement with VCV, for the usual reasons. I wanted to take the pulse, so to speak. I have not used VCV at all in two years, just tried out Cardinal a few times, so I don’t follow closely what happens.
Anyway, I have not disappeared, those days you’ll catch me in every friendly hardware synth DIY place. Cool people there, even some fellow ex VCV module developers.
There are several contributors to DBiz and there have been a bunch of merged pull requests in the past. Even if it won’t be officially integrated, I’m keeping it as my own version because the bug is that damn annoying.
Honestly, Aria, as a trans woman myself, your blog post scared me away from VCV as a whole for a very long time. I’ve since come back and am even learning to develop now, mainly just because the community here is incredible.
I really, really hope someday we get a “perfect” modular platform: VCV has all kinds of behind-the-scenes problems, Softube Modular sounds good but it’s limited and largely abandoned, and Voltage Modular is a mostly premium-only ecosystem (with small module collections costing much more than those in VCV.)
Sounds like you have eyes wide open 
Those issues sound a lot more deep than I thought on your modules, thanks for describing them. I never used them on the official Rack (with that I mean v1.x, I lost interest on VCV things when github issues/tickets were still a thing, “fun” times…) so I was not even aware how they were suppose to really look like. To be fair, for an unexperienced user, they look fine and do their job as modules. Obviously it is sad a lot of small details were lost along the way, didnt know it was the case until now.
Did an officially apology from VCV ever happened? At least for the smug remarks on the VCV Rack issue tracker there was no such thing.
Even disregarding the lack of respect and general personal behaviour, on a pure technical level I think it really sucks that opensource modules are not allowed to be commercial. I thought it was more of a fringe case from my side, but quite a few people here mentioned having similar thoughts.
It is quite a lose-lose situation in my view. I can imagine commercial modules having a higher quality due to work/time being reimbursed in a monetary way, which in turn gives incentive to work on them even more. Existing opensource modules would benefit simply by having great sources to borrow code or be inspired by.
Could also attract external developers if they know there is a bit of money to be made, even if small.
I understand what you are saying, but open source free modules require just as much effort to create and an equal attention to quality. Many years ago there was a book “Quality is Free”. There was a lot of truth in that. At that time, American semiconductor companies liked to charge extra for quality, whereas the Japanese followed the teachings of Deming and Juran (and later Crosby, the Quality is Free author) to make companies customer centric and continuous improvement the way of life, with quality circles empowering the workers to improve things.
I’ve been working on my open source Meander module for 34 years now. I dare say there is no commercial module here with that level of effort and time put into it.
Of course I may be an outlier, but it is my experienced opinion that overall the open source modules here are as high quality as the commercial modules.
But, that is just my personal perspective.
I wonder what it is about rack that makes that true. I agree that some of the open source modules here are of very high quality compared to commercial but in other parts of the audio world that’s not as true. Logic and bitwig better than ardour, diva and zebra are better than surge, Valhalla is better than dragonfly etc and I don’t think those are particularly controversial views. But somehow in rack land there’s some amazing modules which obviousky outshine some commercial work.
My theory is that Modular is more idiosyncratic and open to wierdo experimentation and technical building blocks. And since there’s lots of modules the cost of making one is generally lower (I know meander is an exception to this rule). But it is something I’ve been thinking about lately
I suppose that as with economics in general, price is worth what the market will bear. In the long run, an expensive product can only survive if it is in fact much more capable or of higher quality than a cheaper or free product. But, sometimes the prestige of a brand name can justify a higher price for customers, regardless of the capabilities or quality.
Trying to define “quality” can drive you crazy and has many people.
But, free software pretty much has to be created during spare time. There is no equivalent in the material physical world of products. It is easier to give away our time than our money and other material assets.
Music and other art forms are totally subjective. I think this carries over to the tools and instruments used to create and perform art.
I have a Fender Stratocaster guitar and a Gibson Les Paul signature edition, The Les Paul cost 10 times what the Stratocaster cost 30 years ago. I would be hard pressed to say that the Les Paul sounds 10X better than the Stratocaster, or even 2X. It is totally subjective, but the prestige of the name carries real psychological value 
Interesting! I think the answer is… Love. And age…
People tend to do things for either love or money. As you say, it’s not controversial and no great surprise that in the audio world in general, commercial stuff is usually better quality than open source. Most of the examples you gave like Logic, Bitwig and u-He have large teams of developers working on them.
Large teams tend to produce better (bigger?) work than individuals can. Talented individuals can work for the love/passion of it but you can’t generally run large teams on love - talented teams cost money. So if your work requires a large team, it probably needs to be commercial. Valhalla is perhaps the odd one out there in that it appears to be a one man show - but he’s a genius in his field and can offer exceptional niche software at low prices that even talented teams have trouble competing with.
On the other hand, most Rack developers are individuals - or sometimes a small team of developer and designer. There are no large teams of developers working on Rack modules. Individuals or very small teams CAN work for the love of it. And an individual developing commercial modules has no particular advantage over a talented individual working for love. Particularly when the individual developing commercial modules is still doing it mostly for love anyway, given that commercial Rack development is unlikely to be commercially viable. I can’t think of one commercial Rack developer that relies solely on Rack sales for their income.
I mentioned age as another factor. As Aria pointed out, most Rack developers, whether commercial or open source, are quite old - often with a career in software development behind them. So on the whole they are generally experienced and know what they are doing resulting in high quality open source modules.
Yeah I think this is right @steve and also relates to what I was saying about the ‘scale’ of a module
Writing a weird and wonderful new oscillator in rack is a month of work and people will try it. Writing a daw is a decades long effort to get to the point where you aren’t as good as bitwig, logic, cubase, etc…
and so i think the nature of the product - both scale and size - attracts tinkerers. and layer on the socio-economic and generational reality of computer software tinkerers and voila.
And yeah I agree @k-chaffin for much of this stuff it is entirely subjective. Especially on edges of instruments and sequences and sounds. Core workflow stuff (“is your audio comping good enough”) which makes up a daw is easier to be objective about, but which is the best square wave generator in rack? Well… that sounds like a good conversation over a beer after you figure out the 30-50 records which are the 10 best of the 1980s.
This reminds me of my 3D game engine development years. I began my game engine in 1998 after I retired the first time. The estimate at the time was that to create a state of the art game engine required about 1.5 person-years of effort. When I quit doing game engine development 7 years later, the estimate was 20+ person-years. I finally had to face the facts that a single person could no longer create such a game engine.
But, I did learn a lot doing that project and that prepared me for my final two careers.