as long as terminology discussions are done to avoid new users having to understand the functionality it’s fruitless.
A tool that behaves differently won’t meet ‚intuitive‘ expectations even if the sliders are named in a known terminology…
You can call a trumpet a violin, but using a bow won’t help 
1 Like
I re-read and completely miss-understood you, my bad 
That seems to be a fairly specific case in which the suggested names might cause confusion… whereas the current names always cause confusion, unless you know the math and how the process works or you have your own mental model that is already probably flawed in one way or another.
I understand that people have reasons why they do things. We almost always have reasons why we do things. But having reasons why something is done is not the same as doing something well. Terrible things can be done for the “right” reasons 
I totally get this. I am not trying to say that all control names should have a committee behind them, just that module developers should try and make control names reflect their effect on the image when possible rather than a direct reference to an underlying parameter.
1 Like
That is a very good question. On the one hand, digital photography is building upon the 100+ years of analog film and camera technology. So yes it relevant. A very simple example is shutter speed. That is literally the same. And that we measure in seconds. Sure!
But there are terms that don’t translate that well. ISO is one of those terms… People that used film for them it is very easy. But being born in digital photography… For me ISO was a strange concept. Of course it is not hard to understand the effects of changing ISO. But how ISO correlates to what my camera ACTUALLY is doing and what 400 ISO means in the digital photography… well that is hidden in my digital camera. That the software of camera takes 400 ISO as input and then perform an amplification on the signal, well… I would argue that it would have been better if the camera companies would have come up with a new standaard, something like ‘sensor sensitivity’ or so… So that we ACTUALLY can compare ISOs between camera by the numbers, rather then doing all kinds of test like DXO is doing…
It’s not just film…these are also common scientific terms used when curve fitting s shaped curves. These have been used in enzyme kinetics, elisa’s , dose response curves and other applications…
I guess it comes down to it do you describe sliders by what they do to the math or a parameter of the math or do you try to use an effect on the image if there is a clear and singular effect… Which ever way you go someone will always have an opinion of which is better
1 Like
“functionality” is not the same as “effect”. Color calibration is meant to be used to accomplish the same “effect” as White Balance. Do we need users to always understand how an effect functions under the hood?
Color calibration does not behave differently in its fundamental effect. It accomplishes white balance. It might accomplish white balance using a different method, but the same effect is accomplished. Yes, it has other features besides white balance, but the mainline effect is white balance.
Also, I don’t understand how the control names I selected do not accurately convey their effect on the module behavior. Besides, novel acronyms like CAT and CCT convey exactly zero information to users unless those users are already intimately familiar with the math behind it, and are therefore never intuitive. Good tools are good at what they do AND nice to use.
I don’t have a wide range of experience in a bunch of different software, so maybe the way color calibration is designed and presented is common, but I don’t think so. A layer of abstraction could make the module more user-friendly, without hindering its effect or modifying its functionionality.
2 Likes
I really don’t have a problem with toe and shoulder. AgX is a recreation of a processing method from a well known software (Blender). Yes, you could reskin AgX, and rename things to try and make it more understandable, but I don’t mind it so much.
I think the correct answer is it depends.
If the target user is someone who understands the math and therefore will appreciate the technically correct parameter name, then we should definitely use the parameter names.
However, if the target user is someone who is trying to accomplish an effect specifically, without needing to know the implementation, then the control name should try to inform them on how it will modify the end effect.
In my opinion.
Yes, and I think that is where the discussion about your target audience comes in. If you write software for scientists: it is very clear: use the mathematical properties. I have been involved in those projects. And it always have been fun discussions.
But darktable is not for scientists…
So… I would argue that it cannot be assumed that photographers have the knowledge of the mathematics behind ‘curve fitting s shaped curves.’.
So we should’ve come up with a technically accurate name for this instead of calling it the thing that people would intuitively understand but isn’t very technically accurate? Hm.
The toe and shoulder in AgX are, which is what we were talking about.
1 Like
If you coming from the film world, ISO is very intuitive. But in my experience ‘born into digital photography’, ISO was not intuitive. The effect of changing ISO is. But that is not what I am arguing.
So let me ask you: Explain to me what does the 800 in 800 ISO means? Don’t come up… with a higher number the sensor amplifies the light more…
As far as I know… but may I wrong. It is a not a direct physical unit like volts or ‘number of photons’. It’s a standardized rating number derived from how the camera maps sensor signal → JPEG brightness (and/or RAW metadata), assuming a reference scene and tone curve.
So the meaning of ISO 800 is far from intuitive, there is a whole standard behind it.
PIxls.us did not used to be this constant bickering. I think we can do better.
1 Like
I don’t think this is bickering. This is the hard work of UX development.
Yes, some arguments/discussions will never come to a full agreement, but the discussion itself is important for understand where various people stand and the perspective they come from.
6 Likes
Back in college in 1970 we called tone curves H&D curves, after Hurter and Driffield. Spent the entire 1st semester exposing film on a sensitometer, reading the steps on a densitometer, and plotting the values on graph paper. Phew, I’m ready for a nap just thinking about it.
Your argument is for naming things is to align them with what people already know, and that is what was done for digital ISO, even though the term doesn’t make technical sense, and you’re saying it doesn’t work and we should’ve come up with a new more accurate name.
I know what digital “ISO” does, and if I didn’t I’d look it up and educate myself, which is the whole crux of my argument for “toe” and “shoulder.”
1 Like
Just had to do a very quick test based on this claim. ISO isn’t just metadata (at least on my Canon), it’s a gain added to the raw data. Same shutter speed / aperture, I can induce clipping in a scene just by changing ISO.
But your comments made me think about it, and now I find the ‘digital ISO’ concept very humorous. It’s a term for a standard, that has been hacked to mean ‘gain amount,’ and is now non-standard between models/manufacturers.
No wonder my image is so noisy, there are a whole 3,200 International Standards Organisations inside it.
Ha Ha…no PTSD I hope…I remember preparing figures for publication with Staedtler ink pens and a French curve for the S-curves… We had the original on graph paper underneath the tracing paper… one slip and start over … 
1 Like
Just a perspective from someone reading with interest but not wading in: I agree that this thread, while full of heated debate, has not really devolved into bickering. Hopefully it stays that way.
There have been some new contributors recently to Darktable with fresh ideas, and it’s great to see. But it’s never plain sailing to get new ideas implemented, so hopefully everyone can remain mature and respectful in these discussions. Some people have left angry in the past, and it’s always a shame.
4 Likes
I don’t think your get my point.
Actually there is a little confusion in this thread about what I am saying. Let me make if a cleaner
-
Digital photography is building upon the history of film obviously. And where the concepts are technically the same let’s keep them.
-
where they are not the same (like with my ISO example - film sensitivity is achieved in a different way than sensor sensitivity), it is worth a discussion if the terms should be same. Because what is intuitive in the film world may not be in the digital world. Again ISO was an example.I have not argued that darktable should drop the term ISO because that is stupid, nobody else is using other term for this. However, if tomorrow all major camera companies announce that they will call it ‘sensor sensitivity’ I would argue that we should adopt that.
-
Although raw processing is very mathematical in nature darktable is not a mathematical program / build for scienctists. (Actually I have done raw processing in Matlab, way back). I am arguing that if there is the choice between terminology that is used in other raw editors or the mathematical property, that is better to pick the common one. Because if all other editor are using it, it is a more established term then the mathematical one or the one referring to some film related concept.
-
I prefer terms that are descriptive for most people (hence the example of shutter speed) and are technically (or mathematical) correct. If we stumble upon a term where one of these conditions is not met (and for me the toe / shoulder labels is one of them) we should either adopt what others are doing (if that makes sense of course) because that creates consistency in the field of digital photography or come up with a term that is descriptive.
Yes, and as I’ve argued, we do that in places where it makes sense (exposure, tone curve, rgb curve, even color balance rgb is labeled fairly straight forward), but in places where darktable exposes things that aren’t really found in other photo editors, we should choose the name that aligns with the prior art (e.g. toe and shoulder are taken from blender and blender took them from film photography) or choose something that is more technically accurate so if one does choose to look something up, you’ll find something that makes sense.
We can’t label every single slider in every module “highlights” or “shadows” or “compression” because if there are too many, then the term becomes meaningless because those sliders do not do the same thing.
The whole point is that “toe” and “shoulder” are already descriptive and in photography jargon have a fairly precise meaning. In fact to me, because I spent time in a wet darkroom, lots of the AgX silders made sense to me because it is modeled after the old process.
I can’t repeat myself any further.
2 Likes
I have not commented on those, because they are indeed straightforward.
The film processing jargon and the photography jargon are not the same… I think we keep on disagreeing here. So pointless to continue.