Wow that guy has been angry for, like, the past 18 months. I don’t know how he sustains it without popping. As always, a few good points (regarding the UI complexity) ruined by pointless diatribe and name-calling. Sad.
9 Likes
I’ve decided to skim over. Seems to me there is a true core to some points, but a lot of exaggerated drama around. 
I mean … who cares about how many loc for the lighttable. 10K lines … really… c’mon, I’ve seen much (and I mean much) worse at my work place.
Ah… not worth the hassle. Main thing to extract here is the really hostile view of a single person.
6 Likes
I think in reality he is desperate because the covid crisis was a major setback for the whole libre graphics/libre media movement.
4 Likes
My days! Just had a quick look see. He’s bloating up the internet with all that moaning. All he had to do was keep things simple with just one line. ‘I don’t like how Darktable is so here’s my folk.’
1 Like
I sort of agree its just a really embarrassingly divisive and painful screed… on the other hand-- maybe he is just laying out a very well-researched yet possibly slightly overly verbose well structured and thought out reason why I am neither a DT nor Ansel user.
Maybe it just hits different in English.
Classic AP rant. Makes a number of good points (as usual), in a way that (as usual) won’t win him many friends.
8 Likes
It seems if (rhetorically) he’s moved on from darktable via Ansel in the interim to (presumably) vkdt, then why all the continuing anti-darktable vitriol? I can certainly, generically speaking, understand a difference of approach but if one has divorced oneself from a project, why not just move on instead of wasting blood pressure on what caused the departure in the first place? This sort of dredging smells (to me at least) a bit like he’s trying to justify the departure in his own mind. Maybe I’m wrong.
At any rate, my time can be better used elsewhere so I’ll move on.
6 Likes
I guess he’s still can’t swallow the time and efforts he feels he has “lost” over the darktable project …
Generally I quite agree with what’s have already been said here.
On a later note I’d add that I hope he will play along nicely with the 1man team at vkdt project, because judging his work on darktable he seems to know his stuff !
2 Likes
Presumably, the developers (dunno if they do, but they should). Keeping the codebase small and well-organized is important for code quality and maintenance cost.
Personally, I like to read about discussions of code quality about an open source project of which I am a user but not a developer. It allows me to be reasonably up to date about issues, and gives a good signal about the health of a project.
These discussions are sometimes a bit heated — usually they end up as a rant after the person tried to communicate the same point in issues/PRs. What I usually try to do is abstract from the tone and look at the content.
I am not qualified to comment on DT’s code quality, because I do not actively program C++ at this point. Nevertheless, the following would bother me as a developer:
- I could not obtain test coverage information with 5 minutes of work. It should be visible on the project page as a badge.
- CI seems to run a single integration test. Is there a test suite for the processing modules in the pipeline?
- major additions and refactoring happening without unit tests.
2 Likes
LOC can be a useless metric a lot of the times and isn’t a good quantifier of code complexity. 20 good easy to read lines are better than a one liner that only the developer who wrote it understands (and vice versa).
Light table being 10k lines isn’t a problem if it has a good architecture which I can’t really comment on as I don’t know C/C++.
1 Like
While I do agree with everything said above, I got to give him points for naming the ‘while death loop’. That code is truly quite something 
4 Likes
There’s some kind of ‘while-death’ loop in most every program that doesn’t just do one thing an exit. How else would you keep the CPU’s PC from running willy-nilly to the end of memory? 
That it’s exposed in the application is another thing…
AP is just one person with an opinion. The civilized world is comprised of edifices, all designed with different intent and attention to resilience. The constant in it all is the march of time, in which new ones are created and old ones gradually succumb to the Second Law. That he didn’t start Ansel from scratch reveals his true opinion of darktable and realization of his particular capabilities; else he’d be dragging his thinking straight over to vkdt, if @hanatos would put up with him… 
10 Likes
Also, I will say that his rants about reworking the input system to support external input devices (yes, MIDI was the original driver, but the infrastructure needed to support MIDI makes it easier to support non-Midi input devices that are similar) is ignoring the chicken-and-egg nature of such things. Of course very few darktable users had MIDI devices, because those devices weren’t usable with darktable prior to those changes.
It’s just like declaring that electric vehicles will never be viable because the infrastructure is not there, and thus there’s no point in putting in charging stations that there are very few users for.
3 Likes
Keep in mind that he wants you to pay for his time spent on Ansel and in fact this seems to be his main source of income, so he now has financial incentive to try and convince you that darktable is bad and Ansel is good.
Also his liberapay has dropped by half since starting his own fork.
10 Likes
You been reading Perl again? 
(actually I liked Perl but in the wrong hands it could be incomprehensible)
5 Likes
Couldn’t have said it better 
If it was really unmaintainable it wouldn’t make any sense to fork and keep it going like that, even going through all the work of stripping many things. If that was the case it would have been easier to go scorched earth and just build something from the ground up as many have done.
1 Like
Quite the opposite. If you have a project that is usable in its present state but needs cleaning up, forking and gradually rewriting parts is the sensible strategy. Rewriting from scratch is the recipe for disaster.
3 Likes
I personally do mind forks and little competition. In fact this is core precept behind open source movement.
1 Like
Related question for you European types here as a dumb American: is there some kind of cultural thing over there with amateurs being used as a metric of incompetence?
This isn’t the only place I’ve run into that from someone with a German or French background around accreditation and disregarding someone’s input. Here in the states amateur vs professional just means doing it for the love of it vs doing it for pay and generally doesn’t have a lot to do with skill or ability. Careers that require certification aside (lawyers, accountants and so forth) it seems we get tied up less with that than y’all do over there? Self-learning here seems more accepted than over there maybe?
As an aside back when I was a professional photographer (in the American sense of that word) I knew of a local photographer who didn’t know what half the stuff on her camera did, shot in JPEG but had some popular look and was swimming in clients. Meanwhile I’d point at some of the contributors to this forum and darktable who are “amateur” in that they have a another day job but skill-wise seem much better than that.
8 Likes
I wonder which weighs more, the sprawling word count of Aurelien’s posts, or the accumulated word count of all the posts dedicated to commenting on his posts in these pages? 
7 Likes