![](/static/253f0d9b/assets/icons/icon-96x96.png)
![](https://lemmy.world/pictrs/image/8286e071-7449-4413-a084-1eb5242e2cf4.png)
Not true about xmpp in general. There are modern clients out there.
Not true about xmpp in general. There are modern clients out there.
What’s your problem with xmpp?
Thanks for the interesting point! I learned something today. I guess it all depends on your use-case, whether flatpaks make sense or not.
A floss project’s success is not necessarily marked by its market share but often by the absolute benefit it gives to its users. A project with one happy user and developer can be a success.
I’m not against probabilistic models and the like. I merely try to capture part of the reason they are not always well received in the floss community.
I use LLMs regularly, and there is nothing rivalling them in many use cases.
Flatpaks won’t get their libs updated all at once by just updating a library. This can be very bad in cases like bugs in openssl. Instead of just updating one library and all other software benefiting from the fix, with flatpaks, you need to deal with updating everything manually and waiting for the vendor to actually create an update package.
I’m not 100% sure about this. Flatpak has some mechanisms that would allow to manage dependencies in a common fashion.
This and on top of being inexact, it’s not understandable and un-transparent. These are two of the top reasons to push for free software. Even if the engine executing and teaching models are free, the model itself can’t really be considered free because of its lack of transparency.
I think it’s a short term vs long term debate. In the short term snaps are nice. They might help you get that software you want right now. In the long term though, it will only take away some of your rights and make you into a product.
There are also some interesting things to say about wording. Specifically consumer vs user. Software is not consumed, it’s used and depending on the specific software, the user might be abused by the people producing and controlling the software.
Thanks for considering. If only more motorcyclists cared. It’s a real nuisance for people who are not into internal combustion engine sounds.
FreeCAD is really good nowadays. You need a bit of time to get into it but then it comes with everything needed also for complex multi-part assemblies.
You’re both right!
Remeber, the more boxes you have, the more advanced you are as an admin! Once you do his job for money, the challenge is the exact opposite. The less parts you have, the better. The more vanilla they are, the better.
Not sure I really understand the issue here. Is it about installing or modifying parts of existing config files? I try to use config.d facilities as much as possible for this problem.
Not sure I understand your criticism. Debs definitely help compared to how I was doing things before. Adding some form of parameters (eg. the hostname used by some web application) to the package is necessary and I’d rather have in the form of debconf than having to edit a config after installation.
Do you have an alternative?
What about using standard shell or bash? I know they are not easy to use correctly, but at least they won’t break every few years.
Makes sense. I imagine the push model of Ansible had a lot to do with the speed issues? I can imagine how a solid .deb would be much more performant.
It’s part of the problem, but the other part is that you have to re-do the package building all the time. Alternatively you fiddle with tags and only run part of your roles (which is a hassle anyways because ansible does not really have good abstractions that help encapsulation).
KDE Connect is amazing. Also works without KDE.