I think yes if that line was UV. But otherwise, of its just python, you have the issue that you need two tools, one for running scripts and one for managing dependencies and environments.
Love how well you articulate these notions. They are truly valid for most of us. It also means you're doing it right. The point then is to stay with the practice until you can stop fighting the notion of boredom and instead be "friends" with it. Just be fine with the boredom. Learn to relax in it. Hope you'll try it again.
While I like Proxmox and have used it in a startup, I wouldn't recommend it for enterprise simply due to how flaky the Terraform integration was the last time I checked. Probably a year or two ago. Regardless, the Terraform integration was done in the typical open source fashion of a single developer, where I'd question the amount of resources put into getting the software past 80% completion. The company behind Proxmox does not seem to prioritize the enterprise market segment.
We ended up using the REST API for automated provision management and there are certainly warts. It's a far cry from being a turn-key solution, which I'd argue is preferred by a large enterprise.
It's great for click-ops if that's your use case, but that also wouldn't qualify as an enterprise use case.
It's quite annoying to not be able to use common tools for DNS queries like `dig`, when using `systemd-resolved` for DNS. I think you might have to sometimes flush the caching feature of `systemd-resolved` as well.
It's fine, I guess. But it did take a while to learn the new ways.
I think this is the main complaint from users with decades of experience. Their scripts and old knowledge stops working.
I did have a look at numpy and on my machine the tests did not bloat it as much as you made me believe. The `core/tests/` modules are 3MB and the `__pycache__` doubled it to 6MB. What do you refer to as "test case artifacts"? The modules, pycache, or both?
Also, wrt you statement on pandas; is it the debug symbols that account for the bloat, or the libs themselves?
Which makes your content unlinkable in a way, since there is no way to link to a news item in an RSS Reader. So you can only ever send links to the version you're not supposed to read. :-)
The lack of URLs also means it's impractical for a reader of your feed to ever share your content. There's no real mechanism to link to any particular item in an RSS feed. There's also no guarantee any particular item in a feed will persist any length of time.
So if someone's feed extols the virtues of RSS feeds the only option for sharing that is to share the whole feed. If someone checks the feed a month later that great point made may no longer be in the feed. Since it doesn't exist as a post on a website it's just lost. No no one will ever be convinced RSS is the end-all be-all of blogging.
I don’t want to come across as an apologist for the author, but I can imagine him replying that this is an intentional trade off, and that for a larger-scale conversation about the work, the mailing list is a better option.
I’ll concede that at that point, a mailing list may be preferable to the RSS feed in the first place. Or Usenet.
I just tried to fetch TDARB.ORG's RSS-feed[0] with RSS Guard[1] app (via development "nowebengine" AppImage build), but for curious reasons it shown empty (even XML-file include some posts data).
UPD: It's working now, after switching RSS-feed "Type" to "RSS 2.0/2.1" (by default RSS Guard set "ATOM 1.0")
That doesn't change the validity of their point. One could respond: "why are you generating html content when it's supposed to be consumed by an RSS reader?".
In today's modern age of always-active surveillance, cancel culture and advertising, clickable URLs seem harmful. They can be logged, tracked and profiled. And most URLs are spam anyway, meaning most clickable links on the web are actively malicious. Just observe how few people on HN are even willing to click on links submitted to HN. At worst, you're taking your life into your own hands, at best you're probably just wasting your time on clickbait.
Making links clickable also forces the user to use browser or browser-adjacent tools of near infinite complexity, all of which are owned by surveillance and advertising megacorps associated with the American military industrial complex. By only having bare URLS in plaintext, I can use any tools I want to process them in any way I choose. Those tools can all be simple and free.
Maybe it's best that we be done with hypertext altogether. Clickable links were the original sin of the web. Maybe they were a good idea decades ago, in the naive cypherpunk days of the early web, but now they just provide another way to feed the panopticon, and form the basis for all of the unnecessary complexity, dark patterns and bloat in the modern web. Maybe we should just go back to plaintext and open directories.
Fancy words, but would it even be...an open, accessible web making differences in the lives of billions?
All mobile carriers in my country (or any, I'm guessing) go to great lengths to extract every last penny from my call plans. Should we go back to letters then?
What difference has it made in the lives of billions, other than inciting hatred and addiction? Indoctrinating the masses with disinformation and propaganda? Eliminating our attention spans, reducing our literacy and regressing us culturally to the level of spoiled children? I mean sure, maybe here or there one or two people have something interesting to say, or there's something worth watching or listening to, but the vast majority of it is at best objectively garbage and at worst actively malicious and harmful.
Maybe we should go back to letters. Physical, hand-written letters. A world in which everyone is forced to deliberate over their words and where the pace of information and progress is slowed down to what a human can comprehend, as opposed to the schizoid mania of constant noise we have now, seems objectively better.
We could go back to letters. And then scientific findings will be silos, our medical progress would be in the stone age, people would find out their loved one is sick after they've passed, and so on. There are wonderful things technology has enabled us to do.
But the sampling of the internet we're exposed to is heavily biased to hate speech and nonsense - the internet is pretty much social media for many. Looking through that lens, I'm leaning towards your view.
What better way to ban dark patterns than making "patterns" altogether infeasible? Why assume trust in a model already proven untrustworthy? Too much of the web is built on an honor system as it is.
Structure begets hierarchy begets conformity begets tyranny. Eliminate structure from the web and tyranny over the web can't emerge.
It's a good thought experiment, but just that. The world isn't going to go back to text documents and rss for everything or give up the internet willingly. To make this happen we would either need to make this law (worldwide) or suffer some kind of unthinkable event that damages humanity significantly.
Wrt to 1), how often do you reckon a beginner is faced with:
> By default, `args` is a set of derivation names denoting derivations in the active Nix expression. These are realised, and the resulting output paths are installed.
?? - I think the author really hits the nail on the head with this point.
