A way into the depths of dt?

(This posting could have been put into several recent discussion threads - and it includes quotes from more than one as we on many occasions touch upon the issues on how to find relevant info and get along with dt as a beginner. And this posting ends up with a few short specific suggestions for how we by technically simple means info-wise could make it somewhat less demanding to approach dt.)

Shall it really be necessary to de-learn habits and knowledge established with other image processing tools – if your motivation to get into greater depths with the dt-philosophy increases at a later stage?

Ahh, the math, the math: Of course we should aim to make the learning curve steeper - which means that learning happens fast … :wink:
But otherwise I very much agree (-- Good. I’m not alone …)

For a couple of months I’ve been “sniffing” somewhat on darktable. I’m clearly sinning in the eyes of @anon41087856, since I don’t try things out enough in practice with dt and e.g. utilize the pop-up help built into the UI. But on the other hand I’m a true believer/follower in that I, from a natural inclination, follow his recommendation to try to learn and understand much of the theory first. I do think @anon41087856 is not really quite clear on his primary message her … :wink: - and in any case I’m certain that it’s not enough to say that us dt-students “need to be willing to put in the effort to go the learning curve”.

As many others have pointed out, if people should aim to become “proficient”/“advanced”/“experts” (or whatever term you prefer to use), they still need to know where to start and with a further “learning curve” reasonably described to get going. As others has pointed to, the learning process needs to be a continuous interchange between learning and doing. And for those who get into image processing as beginners, we should anyhow be able to start with dt. We should not have to seek to other, simpler tools - and then learn to do things in a “wrong” way for later to have to de-learn to get into dt - because it’s too difficult/demanding to get along with dt from the start. Beginners like myself, will often have their focus on simple tasks, e.g. cropping, that don’t require understanding (yet) of advanced psychophysical color theory. So lets start there teaching the simple stuff.

This I believe is the right approach:

I just say Amen, (except, once again, for the inclination of the learning curve …, and I also think calibrating the monitor should come much further down the list).

But how then, should we get along with dt??
Are there really only two options:

  1. Decide from the start that I’m going to be an expert in image processing and dt, and sit down and study for hours and weeks and months to understand all the theoretical underpinnings (at least most of it) from the outset.
  2. Rather find another tool – and start “learning image processing” in an unfortunate way, (so that if I later re-decide to go for alt. 1, I have to learn things from the bottom again) ??

I do think, like others, apparently, that there is a third alternative:

  1. Start with dt as the selected tool and get on with its simpler things/functions, and then let in organically grow into more complicated themes where deeper theoretical knowledge is needed to understand what we’re doing and how to fulfill our aims.

As pointed out by others there are many things one could do before we get into the more complicated stuff like color theory/adjustment, de-noising etc – and before a “beginner” normally really feels the need to address such advanced actions/subjects themselves. I don’t need much theory to crop an image – and that task was actually the need that finally moved me from my 30 year old trannies/slides based mindset of “the image is produced when the shutter is clicked” to “maybe I should anyhow use some more time in front of that PC, rather being out and photographing … it might be useful – and fun!”. And as the curious, knowledge-seeking, pc-addict, keen to learn and master, and with strong purist-tendency that I am, I don’t know where this will end – but I don’t want to start my journey on a wrong track. From what I have gathered of information I’ve gotten to believe that the new approach in dt is the right way to go, even if I don’t know how far I will go.

Yes, dt is primairly developed for Raw-processing, but many more simpler tasks that can be done also on a jpeg image should be addressed first i.a. as a means to get to know much of the UI and some of the bells and whistles - various aspects of “exposure” adjustment springs to mind as a good line to progress along after the first “mechanical” crop and rotate stuff etc. A theme like e.g. removal of spots and other disturbing details should be of high practical value, and could be used as a platform for treating the important subject of masks and leading into many more advanced topics.

And, No, we beginners don’t know enough to sort the wheat from the BS, somebody in the know is needed to do that for us.

Yes, there are lots of interesting and valuable information on dt and image processing floating around , in the forum and elsewhere, but there is one thing missing: STRUCTURE.

(This info mess around dt is what meets you at square one. As a Linux Mint user the dt-pages points me for download to Ubuntu PPA which lacks 3.0 and tells me that it shouldn’t be used … ).

  • And there are already various initiatives out there, e.g.:

So, the question to me is WHAT CAN WE NOW DO to improve the current situation BY SIMPLE MEANS?

An advanced Wiki or something like it would of course be nice, but my suggestion now is to make it simpler, as follows:

  1. Create a new sticky in this dt-forum where we gather and put some structure to really useful information on how to get along and progress with dt – a mixture of manual-like stuff and tips/suggestions with links to videos and various texts around as appropriate. (Eventually linked to from the Resources tab on the dt home page.) This sticky should only contain the “pure” information that should be conveyed to students of dt / image processing – so therefore:

  2. Create a second new sticky where we discuss what stuff should be incorporated into the first sticky and how it should be structured – which will probably very much be suggestions to include things that have been posted in other threads. (This second sticky is needed in part because the linear format of this forum makes comments to previously posted “documents” interfere - it’s not possible to easily separate the “documents” from the “comments” (like it e.g. was in a good old hierarchical basic Lotus Notes base …)

  3. We should probably also need a third sticky: A storage for any new material (that isn’t allready found elsewhere), but which should not be a fully part of the main infostructure in the first sticky, but only be linked to in the main text similar to an attachment/Annex.

  4. The main (first) sticky will need to be edited somehow - not by an expert and not by a beginner, but by a group which contains both perspectives. Either somebody could accept such roles, OR we could try to leave it to the community to edit it on a collective basis when there in the discussion sticky appears to be some level of consensus about changes/additions. (I don’t know what the forum design technically allows of sharing editing rights.)

If we just can get a fairly good structure established for info that already exists, a structure that is reasonably easy to find, grasp and follow. I think much will be achieved - and then we can gradually fill in the holes.

(And as a beginner in many respects, I offer to contribute with my perspective. – I’ve previously indicated willingness to contribute efforts for improvements of the manual, but for various important reasons I haven’t yet learnt how to work in the Git-environment, and after more thought I now think the approach suggested here for something more “tutorial” would, at least for now, be a better way to spend time and efforts.)

Could this be a fairly simple and good way to proceed?

I have previously seen various responses in discussions like this, including “people just go wildly tryiing the program and don’t look at info material anyhow” and “just search around in the forum - you’ll find lots of gold”.

As for the first, I do think that there still are enough people who search for good tutorial material to make the effort worthwhile, (there seems to be a more or less endless market for articles, books, blogs, courses and workshops on how to do things in lightroom), and think that the availability of such material for dt will enhance the “market acceptance” of dt and make the community larger.

As for the second, I just look at my self and find that for the couple of months I have been sniffing on dt, I have found lots of gold, but I already start to feel that I lose any overview and also start to forget what I have found.

I need a structure - and I need some guidance for my further progression on the learning curve.
Am I alone on this?

Edit: PS. I now see that the names of those I quote has not followed through when I’ve copied and pasted results from the “Quote”-function. I hope you’ll bear over me for not spending time trying to correct this.

On reddit there’s https://www.reddit.com/r/DarkTable/comments/cyn5h1/darktable_collected_resources/ without vetting process but overall resources are of acceptable quality.

About your whole message in general - it seems like that’s a lot of words for “having good intro materials, articles and tutorials”. So… basically a full learning course on darktable.

have I got it right?

why discussing on work others should do just that someone can get what they want on a silver platter …
There’re plenty of resources how to use darktable - either the old way or the new way. If you want to get the depths of the linear workflow in dt, read the comprehensive articles written by aurelien, watch his video tutorials and then try and learn.
If you just need a shortcut, then read the darktable 3.0. for dummies thread.
These are the ways the designers of the new workflow explains their stuff. Fortunately they focus on improving the software instead of writing documentation for these, expecting darktable to be a lightroom replacement.

So: if you want to habe a specific kind of documentation, then you have to write it yourself, publish it and then others might help to improve it. Thats how darktable improvement is working …

4 Likes

That list does need a culling. While being nice and not wanting to exclude sources I added some that aren’t that great. It is close to the 6 month archive stage and I need to make a decision: If and if so, what to cut.

For those that haven’t guessed yet: I’m the one that created that list…

2 Likes

Why adjusting a software regardless of other ones ? Each software has its advantages and disadvantages. Adjusting a software to another would also be considering the other way (here of processing) is better ? On what base ? On what, another software is better ? Answers to that questions will depend on user. Each user will have different answer.

Of course, a different way means to learn different habits. There’s only one simple thing if you don’t want to : just stay on a software you like the process and have habits to use it.

If you have changed your software AND wanted to keep the old habits, that just means you don’t change for the good reasons. For example, one bad reason to ask for a free and open source software to adjust for a paid one (Lightroom) is the “paid” reason versus the free (in the sense of don’t pay) one !

Life also means to make choice and adapt to the consequences to choices.

Friendly advice. Ask specific questions and suggest specific things only once. Don’t write grievances because you would be setting yourself up for a world of disappointment or hurt. Personally, I would take a break from dt and try something else.

4 Likes

Mmm… in terms of simplicity for the novice, that falls at the first hurdle: dt is essentially a RAW editor. If you’re editing jpegs in the new linear RGB workflow, Aurélian recommends reordering the pipe to linearize the jpeg encoding before editing with the modules in the linear part of the pipe.

4 Likes