how do you know which tools you (really) use? - Community & Forums Related Discussions
Users browsing this thread: 2 Guest(s)
|
|||
Hello.
I was debootstrapping a debian 11 on some partitions yesterday (because my upgrade path won't be easy considering my tweaks, between the init system and the rebuilt packages...) and started to create a mindmap of stuff to do next time (turns out it almost looks like a roadmap of stuff to do to create a distro, but that's another topic). In the end, I noticed... or rather, am noticing, that I do not use that many programs, and that this tool (the mindmap) allows me to get a better understanding of what I use, in which context, and why, if I would like a replacement or not, etc. I might end in sharing the file, when it will be a bit more refined, but it spawned the (one in title) questions of: do you really know which tools you are using? Do you know why you're using them? Do you know any alternatives to them? Why not use them? How would you collect those informations for future use (when a program stops working (depends on deprecated tech), when your needs evolve, etc)? Note that I only have in mind the tools you use, not the ones the tools you use depends on, that's a very different thing. I think it would be interesting to have thoughts from fellow nixers on those questions, and maybe on other ones you might ask in the discussion? |
|||
|
|||
I like to play with tools which I don’t know much yet, just for the knowledge itself. So I guess that I don’t use most of my tools. But I do notice missing POSIX tools on servers, as I usually try them first when fixing stuff.
-- <mort> choosing a terrible license just to be spiteful towards others is possibly the most tux0r thing I've ever seen |
|||
|
|||
It's funny that you mention that your "mindmap" looks like a roadmap to create a new distro. In the process of making a distro myself (which I never accomplished btw), I wrote a package manager for myself (to track stuff installed with "make install"), which I now use on whatever distro I'm using, on top of the default package manager.
And in the end, I found that for the tools that I really use, I install them with that package manager (even when they're available in the repos, so that I can choose the version I install). And because of that, I know more precisely what tools I need, and why I need them. |
|||
|
|||
I often install software to test and forget to uninstall them, which is a problem, since my space on disk is not that big.
Let me see how many packages I have manually installed (not counting their dependencies): Code: $ pkg_info -m | wc -l That's a lot! I do not keep track of which ones I need to reinstall in a future new system installation. But now that's something I need to do: keep track of which programs I need to have installed. In my server, however, I have only vim installed (OpenBSD has a lot of services out of the box). I'll probably install dovecot and git when I set up my mail server and git server. I think I'm not alone. Most people install packages carelessly in their desktop while keeps just the necessary on their server. |
|||
|
|||
Code: ~> pkg_info -m | wc -l I actively monitor what i have installed and remove all unnecessary stuff. Back then as i did package maintenance for a small operating system i used to install everything* , to be able to port stuff even without 4g/wifi. *except colliding packages, obviously. |
|||