> I am glad that you acknowledge that it is only how it makes you feel as the original comment that are the cause of such obviously weren't intended for you interpret them that way.
Yes, but the unfortunate thing about Nix is that so much of the documentation and interactions with the community come off that way, to me at least.
> It's a bit like if I told you that you need to use vim to actually be able to use it efficiently and you told that you had read the manual and understood really well the concept of modal editing.
I think you need to read my comment more carefully. I fully acknowledged that using Nix made it easier to understand:
> Third, I agree, after all my study, that Nix is easier to understand once you use it.
What you replied to was me saying that I felt like I knew what its value is. To take the example of Vim, it's like I have watched many users of IDE's and gotten some idea of how fast they can be with it, then saw an expert Vim user and realized he was 10 times faster than any IDE user. In that case, I would obviously see the value of Vim, even if I don't understand Vim itself.
Reading Nix's manual, the pills, and many blog posts probably won't help me understand Nix fully, I agree, but I disagree that I won't see its value without them.
In fact, if the documentation cannot help potential users see the value of Nix, then they won't take the time to switch. It's a high cost to switch and not something that should be done on a whim. (I don't consider trying it out in a VM as switching, but doing so adds to the time cost of switching.)
> I used to be like you, very confident in that reading, even regularly reading blog posts/articles on a subject was a great way of absorbing knowledge but at some point I have had to admit that that book/article I had read I did not remember much about and when faced with actually using/implementing something and I needed to actually use/implement multiple times to be able to do it by myself.
I do not deny this. I have never denied it. What I claim is that I understand Nix's value.
However, that brings up three points.
First, to be told repeatedly that I don't understand the value of Nix also feels elitist despite that not being your intention.
Second, if I actually do not understand Nix's value, that is a massive weakness in the documentation. Helping people understand its value is crucial to marketing Nix to a wider audience. It's been almost 16 years since Dolstra's thesis was made public. Nix needs that marketing. The industry needs Nix to be in wider use.
Third, the best technical documentation I've ever seen has made me actually understand the thing it was documenting before I ever touched it, and when I touched it for the first time, I felt like nothing, or very little, of my understanding was wrong, and that my understanding was sufficient.
(Funnily enough, one of the things that did this was the RFC for AES. I had to implement AES for an assignment in college. My code is not published, of course.)
I know that the Nix community acknowledges the need to improve the documentation. I believe that the aspect I mentioned above is one thing that could help a lot of people make the jump from sitting on the fence about it. If not, Nix's lunch will probably be eaten by something that comes along, does the same things or provides the same value, and yet has accessible documentation.
To be frank, I've been tempted to do that myself at times when I've only been stopped by knowing that I do not understand what it's doing.