Matrix, via Element.
My GitLab profile’s readMe.MD
states:
Hello. My first name is Roke, and I always shall, and have been since I gained my first computer, a software developer. I specialize in OS architectures and GUI consistency, accessibility, and ease of use.
Matrix, via Element.
Indeed - https://xdaforums.com/t/does-twrp-function-with-non-aosp-oses.4692346/post-89709288 seems to elaborate quite well, if you agree with it. Thanks for the response.
If interesting, I’ve also posted at the XDA Developers forum about whether this is more broadly applicable.
You couldn’t even comprehend that I wasn’t criticising a person (consider this as criticism).
I evidently am able to comprehend it. If I were as incapable as you purport, you informing me would be worthless. You doing so, with this response, demonstrate consequently that your assumption is incorrect. It’s a strange one to make.
It really isn’t that great that instead of active discussion you stumble upon months-old threads.
You responding renders it active. Any alternative designation is fundamentally nonsensical.
If you’re to discuss semantics, be more pedantic, else this conversation is uninformative for us.
I made a point illustrating that switching off from reddit didn’t do any good for the community.
I’m aware. It’s quite easily comprehensible. However, it’s unsubstantiated, hence the downvotes.
Is it FOSS? I’m having a difficult time locating its source.
No, I had not. That’s certainly novel.
Really impressive work. SystemD was the sole thing missing from PMOS for me – I’m ecstatic that I’ll finally be able to say, when it’s done, that – excluding 3rd-party GUIs’ support – it’s better than AOSP.
I don’t have MarkdownLint nor Markdown All-in-One installed, and disabling Markdown PDF doesn’t resolve the problem. Unfortunately, we’re looking in the wrong place, it appears.
Yeah, I just installed it from 23.12’s repository. Works well for the basics. Many thanks.
Thank you. I’ll use that information about the USB ID for my report to KDE. However, could you elaborate somewhat? That is, would it be correct to ask for Plasma to not merely utilize the USB ID to identify whether a device is accessible, but test whether it can be accessed by MTP first, before presenting that option to the user? Regardless, per what you’ve said, I’ve consolidated my previously disparate reports under https://bugs.kde.org/show_bug.cgi?id=486931.
Additionally, I’ll definitely mention on Bugzilla that incorporating some of the patches in that package would be a feasible implementation method. Relevantly, I’d like to use that package, but because I’m not using Edge, would adding that repository do any harm (for instance, would it be added with higher priority by default than existent repositories)?
Those criticisms seem reasonable. Regarding package signing, are you referring to https://github.com/NixOS/nix/issues/613#issuecomment-134361033? Additionally, that default for pip seems veritably insane. I understand using system packages, but modifying packages outside the virtual environment is definitely weird.
You shouldn’t criticise someone because they were late to a discussion.
It worked for me. I came here instead of Reddit, which would have alternatively been my primary choice.
I’d like it opened for new users to be able to participate there without a Lemmy account, but there’s a cost-benefit ratio to everything, and the additional moderation burden and potential fragmentation of the small community appears to outweigh the benefits. Additionally, a lack of choice paralysis can be quite a significant advantage.
I’d say that that makes it more probable.
Can you elaborate?
I used it yesterday, via Pidgin. I’m rokejulianlockhart@xmpp.jp
. Why else would I have referenced it? Don’t tell me what I’ve done. That’s not a way to have productive conversations.
Regardless, I can’t provide any more technical insight than that - I know solely that the clients provide so much more functionality that irrespective of the protocol, it’s better in practice. Fedora, openSUSE, the Bundeswehr, NATO, and Beeper - all chose Matrix over XMPP, not least partially because of Element (which they also all chose).
I don’t believe that its existence causes more fragmentation than it remediates. https://news.ycombinator.com/item?id=36939482 explains why I consider Matrix fundamentally superior most (if not all) uses, although in practice it’s because the clients (Element and FluffyChat primarily) are cross-platform and support a generally uniform set of features, in comparison to the aged (but glorious) Pidgin, and its counterparts.
Its bridges are FOSS, but its client (an Element fork) doesn’t appear to be.
That’s why it’s considered good practice to act on the presumption of innocence until proven guilty.