DT is really great with all the sophisticated processing. But using it is not easy. My idea is to create a version/mode with a much more easy-to-use interface for all the modules. It should of course use the same pixel pipeline and produce compatible XMP files.
My proposal for ‘simple mode’ is to combine different modules in easy to use and understandable UIs like it is done in the basic adjustments:
basic adjustments (is already done)
noise and details: all the denoising and detail modules
raw denoise and denoising-*
lowpass, highpass, soften, locale contrast, sharping, etc
colors: everything that changes the colors
WB, color correction, color balance, vibrance, velvia, etc
tones:
filmic, tonemap, etc
geometric: everything that changes the geometric properties
lens correction, perspective correction, crop and rotate
special: everything that does not fit other groups
retouch, negadoctor, grad.density
finish: everything to use for the final appearance
LUT 3D, grain, framing, watermark
Possible ways for implementation: A different application DTsimple that only supports the ‘simple-UI’, or switching in DT between simple-mode and advanced-mode
It should save the mode (simple/advanced) used for a picture in the XMP/database.
If edit mode was simple create additional entries in the XMP/database for the simple parameters.
Because mapping of the simple parameters to the advanced parameters is one-way, it should show a warning if the user tries to edit a advanced-mode picture in simple-mode and show a hint if the user wants to edit a simple-mode picture in advanced-mode.
I know that this proposal is a heavy heresy for all the hard-core pixel scientist:)
This all doesn’t help if the dummy user just expects a free Lightroom.
There‘re several articles on using darktable for the i-don’t-want-to-do-sophisticated-stuff users around. You really need just a few modules and with 3.4 theres already a module group preset for this.
But if they refuse to read the basic stuff first and then complain wellinformed each effort spent in a dummy GUI is a waste of time …
I was also thinking about that and one example would be looking at what Apple is doing with Garage Band being just a striped down version of Logic Pro X.
And while they can do it because they are a big company, I don’t think that could be done with DT resources without seriously hurting ongoing development efforts.
I for one would not dumb down the UI by hiding modules etc, just make it clearer like you are proposing. DT, while being a powerful app, isn’t so hard to master but that isn’t to say it couldn’t be better and thus maybe easier. Just by by doing some UI rethinking.
Making a “simple” ui that nonetheless includes complex tools like filmic is a serious challenge.
Merely reorganizing everything may make it easier to discover functionality, but it’ll take a lot more work than that to make it actually easy to learn how to use the tools.
I think nobody (at least here) is expecting tools (modules) to get “simpler”. Clearer and more easy to understand yes. Just look at the progression of filmic rgb. But in order to be simpler modules would have to make more assumptions and that’s where we’d loose the power of that module. It’s a trade-off.
But also, since DT can have unlimited amount of modules, who’s to say we can’t have both?
Isn’t part of the problem for a new user that some of the basics in dt are different from most other programs?
E.g. in an editor that uses integer formats internally it’s a mortal sin to go beyond 100% in any channel anywhere during your editing. In dt it can be necessary to get the result you want: increasing exposure to get your “middle gray” correct can appear to blow out the highlights.
And that difference means that new users have to get some “training”, I don’t see how you can show that in the GUI.
@KristijanZic: I don’t think more modules is the way to go, there are too many already…
But perhaps it might be possible to have a “simple mode” for some modules, which hides some of the controls (e.g. in filmic, you could perhaps hide the reconstruction and display tabs in “simple mode”, and perhaps have the latitude and shadows/highlights balance adjust automatically). Combine that with a preset that only shows a minimal set of modules in the base tab, and you might already end up with something that’s less impressive for a new user.
onboardring, a nice one
You just simply make a user aware of that fact.
That was the exact problem I had when I first tried DT. I opened an image, opened shadows and highlights. brought shadows slider all the way up, put highlights slider all the way down. Got blown away by the crazy ugly result and closed DT for a few years
Yeah, maybe but it really isn’t that hard to master those modules. I don’t really advocate for simpler modules or modes, but I’m not against it either if someone wants to do that. Users would use it for a month and then they’d use the verbose mode.
But it would help new users a lot. I just happen to think a good onboarding might also o the trick.
Training could be done through the GUI in the form of “interactive tutorials” based on hand-holding the user from start to finish on a predefined set of RAW images and tooltips that describe what should be done and what each slider / setting means.
A couple of RAW images for portraits, another few for landscapes, some with tricky dynamic range situations, and so on…
The output is then verified (allowed range of settings) and the user is notified (success / failure).
Something like an interactive version of @s7habo 's editing moments:
Mixed with output verification based on unit tests, similar to this:
I had similar idea to provide a few images but make a video tutorial where a very soothing female AI/android voice (like one from the amazon voices) explains and shows you what to do and then waits for you to do it until you click next. It would be cool if the values could be checked the from the xmp too.
I wonder how hard would that be to make in lua. Probably not that hard.
The main point is to allow a layperson to associate certain settings with certain effects, without delving too deep into the subject matter. Since different photos require different settings, the user will be taken through a variety of predefined scenarios + RAW images to intuitively grasp what’s going on. Each module could contain a question mark button ( ? ) that would open the corresponding tutorial image + section.
Text is simpler to do than voice and can be re-read multiple times.
Instead of using the Luminar AI approach, a “guided manual” approach can be used. The “search module” feature can be turned into a “type to discover” feature, where one types what one wants to do in layman terms, and one gets pointed to the necessary modules (and perhaps gets told whether one needs to use local editing via masks or global editing).
For example:
“fix sky brightness” → exposure, filmic, tone equalizer
“make colors pop” → velvia, vibrance, color balance
“make details pop” → sharpen, highpass, local contrast, contrast equalizer
An autocomplete feature would give suggestions to the user in order to help one complete one’s query. E.g. user types “c” → suggestion: color
Then, the interactive tutorials could either introduce a completely new user to desired modules or help somebody already acquainted remember what needs to be done.
Or maaaybe just read docs/watch aurelien’s and bruce’s videos? Or even pay for private lesson with Aurelien? Actually IMO buying lesson with Aurelien has greater return value than time investment in developing “simple ui”
@subliminal a help page like this one that photoscape X uses with quick 3-5 min demonstration of most features would go a long way to sorting people out…again many of these resources are out there but buried in longer videos or full edits http://x.photoscape.org/ scroll down a bit to see …
Actually IMO buying lesson with Aurelien has greater return value than time investment in developing “simple ui” 
?
