 | NixOS Marketing | 254 Members |
| NixOS website + marketing team: https://nixos.org/community/teams/marketing.html | 54 Servers |
| NixOS Marketing | 254 Members |
| NixOS website + marketing team: https://nixos.org/community/teams/marketing.html | 54 Servers |
| Sender | Message | Time |
|---|---|---|
| 2 Mar 2023 | ||
 raitobezarius | for full transparency, I was planning to start commercial training in Paris, France on Nix/NixOS too | 13:31:48 |
| raitobezarius | I might announce something serious in 6-8 months because too much stuff going on in life | 13:32:12 |
| raitobezarius | (I also mentioned this to tfc during the last NixOS Ocean Sprint) | 13:32:34 |
 jeremiahs joined the room. | 16:35:19 | |
| jeremiahs | garbas: Thanks for adding me to the team! Looking forward to joining the next meeting. | 16:38:49 |
| jeremiahs | In reply to @evils:nixos.devI've been working on some ideas with tom to make that look more like a straight line that branches out. | 16:47:00 |
| jeremiahs | I've been looking at what tools like docker and ansible do for their projects and comparing it to nix. Not on a technical level but more of a front facing execution level. | 19:53:32 |
 evils | In reply to @jeremiahs:matrix.orgmaking it not look deliberately obfuscated? | 22:25:56 |
| evils | In reply to @jeremiahs:matrix.orghow do they stack up so far? (i'm not familiar with "comparable" things) | 22:26:34 |
| jeremiahs | In reply to @evils:nixos.devSpecifically comparing user experience, and other defining aspects of a tech project and what makes them successful. Then looking and seeing if nix is doing them, and in what capacity. | 22:32:00 |
| evils | my impression of other solutions in the problem domains Nix covers is that they're more niche or at least, Nix is more general, it covers more problem domains so the cost of learning Nix aught to be considered spread over the domains you solve problems in | 22:34:11 |
| evils | * my impression of other solutions in the problem domains Nix covers is that they're more niche or at least, Nix is more general, it covers more problem domains so the cost of learning Nix ought to be considered spread over the domains you solve problems in | 22:34:25 |
| evils | (i'm referring to the Nix ecosystem as a whole) | 22:35:52 |
| jeremiahs | In reply to @evils:nixos.devThis is one of those aspects. Obfuscation can be good as long as we peel back the veil for new users. The other aspect is SEO. Having so much of learning how to package software be in the source code only is an issue. Take every mkderivation wrapper we have for every project. When someone searches for these the internet is often barren or fails to point the user in the right direction. | 22:36:10 |
| evils | In reply to @jeremiahs:matrix.orgsorry, that was a joke about the XKCD graph actually intentionally looking chaotic and your message could be interpreted as you working to simplify the actual XKCD graph | 22:37:05 |
| evils | to rehash something i said a while ago elsewhere i feel like the documentation only provides a point cloud in the opaque fog that is the Nix ecosystem | 22:39:04 |
| evils | that analogy has the distinction between clearing things up and writing more documentation | 22:39:30 |
| jeremiahs | I'm also working on courses to teach nix | 22:39:36 |
| evils | In reply to @evils:nixos.dev or in the terms of https://diataxis.fr/ the difference between providing more understanding and providing a more complete reference | 22:40:51 |
| evils | i want to do an explanation workshop thingy next month i don't really have an idea of how to prepare for it my plan is to just try to answer people's questions and hope some other Nixers can help with that | 22:41:52 |
| evils | not sure if i should avoid it turning into a focus group | 22:42:47 |
| jeremiahs | In reply to @evils:nixos.devTwo parts of a very important picture. Take rust for example. People know it has a steep learning curve, but the learning curve isn't because they can't find the info they need. The compiler is very explicit when things go wrong and why. It even offers suggestions. But more back to focus of marketing. Userd need a clear path to getting started. A lot of users hit the biggest wall when building packages. The course idea im working on my end is taking many starting points and shifting the lessons to teach packaging, not hide it away till later. | 22:45:28 |
| jeremiahs | It is not a matter of if you will build a derivation, but when. | 22:46:20 |
| evils | i'm still of the opinion that there definitely are potential users that ought never to need to build a derivation and when those users do have to, that's because of a leaky or incomplete abstraction? | 22:49:21 |
| evils | not sure i understood you correctly, about shifting the lessons but one of the reasons i've given up on doing a linear talk about Nix is that most people have different things they need from Nix | 22:50:40 |
| jeremiahs | What's the harm in offering a linear path of how to build and learn derivation? | 22:51:42 |
| evils | because someone who just wants to config mosquitto doesn't think they need to package anything | 22:52:22 |
| evils | mostly the harm is that a large fraction of the audience will not be tempted by that path | 22:52:57 |
| evils | (any audience i'd give a talk to would be a somewhat scattered sampling) | 22:53:15 |
| jeremiahs | Give it 3 months they will find something not in nix. Now what. | 22:53:31 |