I’ve removed several posts that were going in the wrong direction. Let’s go in the right direction, please.
Well we have to comply with laws like GDPR. So we don’t really have too much choice.
From the software half:
I have to say, I like the way he did import/export (especially export as a popup), the lighttable change, and the color picker change is also kind of nice… I wonder if he altered the mechanics of the color picker as well.
I was not as big a fan of the masking changes, primarily because I care less about vertical space on the right panel than being able to keep the left panel collapsed as much as possible. But I also understand how frustrating that can be if you are editing on a laptop.
Oh, right! I don’t think that existed when I was a moderator on a busy forum. Which, by the way: once was enough and I would never do it again!
I’m very grateful to those that do. Thank you!
I’d argue that the majority of users, maybe developers too, liked a lot of his ideas and work, and maybe respected the coding. There’s no doubt that some aspects of Ansel will be better than Darktable and/or preferred by many users.
Personally, I’m very thankful for a lot of the work he did, and some of his modules and ideas are my most used tools. And the new workflow has led to modules like AgX, which I love. Even Tone Equalizer, which I don’t think is as good as it could be, has a couple of presets I use on many of my images and would be hard to replicate in any other way.
I’m not a developer and so can’t really speak to everything that went on behind the scenes, but I can understand that there’s a point when it’s better to go separate ways instead of constant battles and accusations being thrown around. I just wish it had ended at that, so I’m surprised to see such vitriol all these years later.
But I try not to judge too much before meeting someone in real life. You can never fully know what someone is really like or what they might be dealing with. So I’d like to think I’d still buy him and any of you a beer (or coffee) and have a good chat if we ever met before I fully formed an opinion.
This is something that’s come up over and over again, from multiple people in multiple places.
And it completely misses the point. AP invalidated himself in the community and with the other developers by the way he treated them. That’s it. There is nothing past that. It is irrelevant if his opinions were correct or incorrect, if his code and modules were good or bad. He made those things irrelevant by treating other people so poorly that they didn’t even want to engage with him anymore.
Suppose we ignore all his ranting, ignore all the bad comments and skip the past (I know it is hard)…
What can we actually learn from AP and Ansel to improve darktable?
This doesn’t mean that I agree with him at all, but if we put this through a very thick filter… what stands as something to improve darktable with?
I do understand the feelings about AP, but I don’t see how ‘dwelling on the past’ is helpful and productive for us as a community
I agree that the developers could look at what is good in Ansel and can be incorporated into DT to improve DT. The same applies to other FOSS software.
In that case, it is a fascinating example of projection from a country that rage-quit the EU recently ![]()
It reminds me of a map of how each country named syphilis after another country:
I am surprised that only the Czechs hated Hungary enough to name the disease after it. We sure pissed off more neighbors than that!
I just skimmed the recent video, but from the sample I have seen I decided not to watch it either. I find Wilfred Bion’s concept of alpha and beta elements of communication useful for digesting content like this; in short, alpha elements are reasoned thoughts, while beta elements are undigested emotion. Both use the same building blocks (words) so it not easy to avoid being drawn in to the emotional content when communication dominated by beta elements. But I have learned to simply extract myself from such experiences.
The menu idea is good, and it’s what every other software does. I never thought about it but the “stack of buttons” thing in the light table is really odd once you examine it, and the space it takes is a valuable resource since that real estate can be used by the images themselves.
The optimization of switching from light table to darktable and vice versa, and advancing to the next or previous picture. The current delay is inexcusable in my opinion.
It’s good to note the opposite though, darktable’s pipeline feels faster than Ansel nowadays due to @\hannoschwalm’s work and overall there seems to be less bugs or unexpected behavior on darktable.
but I don’t see how ‘dwelling on the past’ is helpful and productive for us as a community
of course it isn’t, but it is hard to avoid. imagine being a darktable developer with goals different from those of AP, and listening to the videos or reading any longer text of him. Any of those is a very active reminder of the past.
Its not that AP says “i had to fork because i’ve different goals”. He says “i had to fork because every other ‘developer’ of darktable is an idiot and too stupid to see things the only right way” (and he says that quite often).
Don’t count on the darktable developers being able to go past that.
@Terry wrote:
I agree that the developers could look at what is good in Ansel and can be incorporated into DT to improve DT. The same applies to other FOSS software.
me thinks this is being done.
I fondly remember asking the DT developers to create a tool for setting white balance to skin tone as was available in Photoshop Elements, but not Photoshop itself (go figure). OMG, was I attacked as a white racist person who wanted skin tone for Caucasians only…so I moved my request to if we could select a color to match WB to… Of course I meant skin tone but now my request had a politically correct wrapper. AP then went beyond what my feeble mind imagined and created the area color mapping option in the color calibration module and as an added bonus the area exposure mapping option in the exposure module. Damn brilliant in my opinion. I could pick my ‘perfect’ skin tone and color balance to that.
I miss AP’s DT input but I can appreciate that DT is not just one person and one person’s vision of what is correct. I am so happy to see new developers on board and in fairness some of their AI improvements have caused friction, which saddens me as we are free to choose to use AI or not.
Two lessons to take away from AP’s four hour video is no one wants to watch a four hour film let alone video. Even the classic film Gone With The Wind only went for 3hrs and 41 minutes and I remember we had an intermission when it was shown. The second lesson is DT is more than a single person, so move on if you are not happy.
It’s history. Interesting to those of us who came afterwards, but deeply personal to those who lived with and through it.
A four-hour rant? That itself is a phenomenon, and it tells me that I don’t want to go there. But I’m grateful for the backup history here.
By the way, aren’t we celebrating the release of darktable 5.6 here? Probably most of the dt contributors here are using using master branch, so it isn’t as if any new functionality appeared for us as of 21 June. But it is quite a milestone. And evidence that our developers have got better things to do than dwell too much on past problems.
darktable is an amazing community. Perhaps AP wasn’t the first upset; perhaps he won’t be the last. But may dt have the strength to hold together in the future as it does now.
(and I am one of those who is perfectly happy with the basic interface. And it’s customability. errmmm, customisability!)
Funny, that. I’m a white person too, and very much concerned with skin tones — as I, almost exclusively, photograph “brown” people* with complexions varying from white to black.
*I have no idea what the current “EU/USA acceptable” way of saying this is. I only know that one of the lessons I have learned is that the majority does not have to be identified by colour, so Indians in India are just Indians in India. But when necessary, they seem to use the word brown.
I get along quite well with Aurelien. The first time I offerd to support dt with a donation, I was told that there is no entity that can receive money, but a donation to Aurelien is the best thing I can do. I then met Aurelien and many of you in 2019 in Saarbrücken and Aurelien and I kept in touch. He has built me two websites and taught me how to update and maintain them and he always had an answer when I ran into a problem (I am more a photographer and not a coder at all).
We also discussed dt vs. Ansel about … 18 months ago. I old him what I liked about Ansel (UI) and what I didn’t (stability), why I would rather stay with dt instead of swinging over to his program. No hard feelings, just a good conversation.
I don’t know what happened inside the dt development team, I can read that feelings have been hurt, ways parted and different roads have been traveled ever since. I also don’t need or want to know what led to the separation, I only want to state that with all the comments here, I know Aurelien as a friendly and helping person who knows his stuff.
IMO the drop down / pull down menu construct is a long-proven and efficient means to organize and present hierarchically related choices.
I use a piece of astronomy software where for years the developer absolutely refused to use a menu, instead littering the UI with buttons for everything. The reason? He just didn’t like them.
He eventually relented and included a small three-choice menu, and I think now admits it was an improvement. But instead of using the available API he re-implemented it from scratch,l. It works similarly – but not quite the same – as the standard OS / UI menu (the user must repeatedly click instead of dragging / releasing the mouse).
Why? I guess he just didn’t like what was available… despite it being superficially identical and functionally superior.
This sounds extremely bothersome. I get implementing your own menu, but at least make it functionally the same as other menus…
–
I agree and I think the global menu in darktable would be met with similar resistance, unfortunately, based on how some PRs to change default UI functionality were received.
I really like darktable and am thankful to the developers. The pipeline and the available modules are really good and the functionality is there, but it could benefit a lot from a ground up redesign, just like Blender did, but I am not confident it will happen, so I have just accepted it for the time being.
Agreed, as long as oft-used options are also available in a context menu – having to drill down through a long menu structure every time you want to do something gets tiresome quickly. This, IMO, is the advantage of having everything available in the side panels
Fortunately the menu contains mostly configuration and other somewhat rarely used items. Otherwise it’s still button pushing, but it’s not too bad now.
FWIW I agree with you on the dt UI but as a (only very) occasional dt user i don’t push my views in that regard.
I know at least some of the devs would like to move in that direction, but as always someone will have to do the work, and they may not have the time.
