wswope

10 ms·

91.
by wswope 2y ago
Speaking as someone with similar views to the OP: my “better solution” is to write an idempotent shell script targeting a specific Debian release/ISO that handles system setup end-to-end. It is for nearly all intents and purposes funct
92.
by wswope 2y ago
> the authors point is that a little bit of extra thought and work can make a lot of difference to quality Care to bring home the thesis on how that’s actually really insightful?
93.
by wswope 2y ago
> Definitely not near the extra protein serious gym people attempt to consume. Plenty of people exist who aim for >200g of protein daily - but that’s objectively overkill. There’s very little benefit, if any, to going over a half gram
94.
by wswope 2y ago
Yeah, time-series on that scale has unique requirements that can really cripple most DBs. > I'm currently leaning towards python driven dashboards since we can connect them to a REST api which will talk to our data structures and ke
95.
by wswope 2y ago
IME trying to use Tableau & kin for analysis, especially time-series work, is a case of holding the tool upside down: you can do it, but it’s not the intended use-case and you’re likely to wind up frustrated. The viz tools are just viz
96.
by wswope 2y ago
Gonna have to give it a proper read-through over the weekend, but this looks like a stellar guide at a glance. Sincere thanks for sharing your work and looking forward to further entries in the series!
97.
by wswope 2y ago
> That said, maybe there are populations which are much more sensitive. I suspect the lawsuit would be more informative to learn more. Yeah, the LD50 is misleading for potassium in this context. Most people can tolerate higher doses comp
98.
by wswope 2y ago
The underlying implication of the linked comment is that Anthropic is using quantization or similar quality-reducing strategies to help keep their service online due to the same shortage that has been causing availability issues.
99.
by wswope 2y ago
Great summary - tyvm! I can absolutely see how quick feedback from those metrics could help people dial in their maintenance.
100.
by wswope 2y ago
Thanks for posting! Out of curiosity, could you talk more about the practical utility of the sensor readings you get while monitoring compost? Temp and moisture seem straightforward, but e.g., does gas composition imply anything about C&#x2
101.
by wswope 2y ago
It’s a plug for the author’s discord server (first entry under the “must-join” communities list). Pro tip for any aspiring DEs: you can ignore 98% of the linked junk on this repo. Learn python, learn SQL, read DDIA, and you’ll do fine.
102.
by wswope 2y ago
I’ve only toyed with them a bit, and had a similar experience - but did find I got better output by forcing them to adhere to a fixed grammar: https://github.com/ggerganov/llama.cpp/tree/master/grammars
103.
by wswope 2y ago
Firefox is going to be massively disproportionately more popular among LLM users, the issue has been happening to people for over half a year, the error message is terribly misleading, and from a CS POV, it should be relatively easy to flag
104.
by wswope 2y ago
You’re not wrong, but I’m more leaning on the anecdote to make a broader case that the big walled-garden players (Google/OAI/Anthropic) all kinda suck in similar ways. I.e. - I think Anthropic is seeing a boon right now not becaus
105.
by wswope 2y ago
Yeah - I run the ESR version of Firefox, which is probably the root of it, but wide prevalence of people hitting the same issue for so long + the total radio silence from support is what really threw me. At least OpenAI pretends to care by
106.
by wswope 2y ago
As much as I dislike OpenAI’s ongoing shenanigans and disdain for their own customers, I tried to sign up for Claude last week. Turns out that Anthropic’s signup flow has been silently broken for months for Firefox users: https:/&#x2F
107.
by wswope 2y ago
You inspired me to check, and I’m paying $50 per month for my home club. We have monthly member appreciation nights to burn off excess budget where they order a bunch of expensive pizza for everyone.
108.
by wswope 2y ago
While I think these venues are cool, they’re inherently VERY limited in maximum possible scale. They’re more of a competitor to bars like SPIN, rather than traditional table tennis venues - kinda like how Top Golf isn’t really trying to ste
109.
by wswope 2y ago
I read the JACC article too, and thought the NYT claims were decently supported: > The second unexpected observation was the lower rate of non-CV death with semaglutide vs placebo, particularly infectious deaths, including in patients wi
110.
by wswope 2y ago
Rather arrogant of you to spew forth unfounded conjecture without even bothering to skim the high-level details of a six paragraph article. > the protective effect occurred immediately — before participants had lost significant amounts o
111.
by wswope 2y ago
Think of it this way: the release of VS Code didn’t mean people suddenly stopped using or updating Emacs/Vim. VS Code simply offered a more polished, beginner-friendly way of setting up and building software projects than the old TUI e
112.
by wswope 2y ago
Most projects do just fine with pip + venv, and that’s what they stick to. The exception is if they have specific dependencies outside the CPython ecosystem - in which case they’ll probably be using conda. Examples of such dependencies incl
113.
by wswope 2y ago
Note to self: impersonate djyaz1200 when launching cyberattacks under the new world order.
114.
by wswope 2y ago
I pull music from a wide array of sources (Bandcamp, yt-dlp, etc.), drop each album or release in its own folder in ~/Music, run beets on those files to clean the metadata and consolidate into V0 MP3, then rsync the beets library over
115.
by wswope 2y ago
Congrats to the Windows Terminal team member who went through the process of getting a PhD to implement this.
116.
by wswope 2y ago
If you’re open to a paradigm shift, I can’t recommend atuin highly enough. It’s truly the pinnacle of shell history tooling, but requires a ~1-2 day muscle-memory learning curve. https://atuin.sh/
117.
by wswope 2y ago
Hot take as a datamonkey who’s done a lot of work in this vein professionally: General-purpose self-serve is hostile to non-technical end users. Most do not have the mental model of SQL to guide their usage of the tool, so giving them an “o
118.
by wswope 2y ago
Most asthma medications aren’t steroids. The types of steroids used to treat asthma aren’t anabolic steroids - they’re corticosteroids and are generally performance-reducing when taken long-term. The asthma medication discussed in the artic
119.
by wswope 2y ago
Your question is a cop-brained false dichotomy. The real world isn’t divided into good guy/bad guy “sides”. It’s not an either/or - we can and should address crime while also respecting civil liberties.
120.
by wswope 2y ago
> a couple of years ago You’re bearing a grudge based on an unintentional bug introduced 18 years ago. Weak take and irrelevant to topic at hand.

More ›