benterix

15 ms·

271.
by benterix 10mo ago
> you are pretty likely to be in the upper percentiles of emission culpability Since you are talking about culpability specifically, what exactly can they do about it? Or, more to the point, what have they done so that it it is their
272.
by benterix 10mo ago
The wiki explicitly mension the typical sans disadvantage: "One potential source of confusion in Calibri is a visible homoglyph, a pair of easily confused characters: the lowercase letter L and the uppercase letter i (l and I) of the L
273.
by benterix 10mo ago
This follows from the parents statement.
274.
by benterix 10mo ago
While I agree, there is no reason NOT to use a perpetual license (e.g. for Excel 2016), unless you really, really need the subscription-based version. You may notice the last edition of softwares that had perpetual licenses but moved on to
275.
by benterix 10mo ago
It would be nice if the instructions spelled out what to do, then I could do it. Otherwise I have to guess what author meant. But all in, a nice small exercise, thanks!
276.
by benterix 10mo ago
First, if this is private vs corporate, they are probably using a separate laptop, likely with a VPN. Second, doing this kind of shadow profiling is a lot of work with potential legal consequences with little gain, at least for support team
277.
by benterix 10mo ago
Unfortunately many people are unable (unwilling?) to grasp this logic and mistakenly assume that if something works for them, it obviously has to be the same way for others.
278.
by benterix 10mo ago
I have mixed feelings about this. On the other hand, developments like Deepseek are clearly marks of progress and practically a gift to the society, and it's a good thing. The Chinese are also creating and maintaining many valuable ope
279.
by benterix 10mo ago
> If AI can diminish some of the monotony of research, perhaps we can spend more time thinking, writing, playing piano, and taking walks — with other people. Whenever any progress is made, this is the logical conclusion. And yet, those w
280.
by benterix 11mo ago
> if I search on Google search for specific terms, I am not interested in crap such as "others also searched for xyz" - that is just ruining the UI with irrelevant information You assume the aim here is for you to find relevant
281.
by benterix 11mo ago
Play stupid games, win stupid prizes.
282.
by benterix 11mo ago
> Most people can't afford to have more than a few sick days in a year I believe it's true in the USA, but not necessarily in Europe. It's quite normal that you have two infections a year, plus sometimes your kid catches s
283.
by benterix 11mo ago
A devils advocate in me would say that this post was authored by one of LLM models creator realizing they really need more fresh meat to train on.
284.
by benterix 11mo ago
To have an alternative? So far the work on transformers was done in a large part in the open (except OpenAI which tried to be as closed as possible). There is zero guarantee that whoever discovers the path to AGI decides to publish the pape
285.
by benterix 11mo ago
I used to work for a short time as an IT teacher. The kids there were wonderful, even the troublemakers weren't that bad. What struck me is that clearly for some of the boys I seemed to be the only male figure they could related to in
286.
by benterix 11mo ago
Which, as practice shows, tend to be understood differently by customers and PMs.
287.
by benterix 11mo ago
> They can just integrate whatever improvements the actually existing AI models offer into their other products. If this is what users actually want.
288.
by benterix 11mo ago
For an older but more to the point comparison, see this: https://jan.rychter.com/enblog/cloud-server-cpu-performance-...
289.
by benterix 11mo ago
Did these Ocaml maintainers undergo some special course for dealing with difficult people? They show enormous amounts of maturity and patience. I'd just give the offender Torvalds' treatment and block them from the repo, case clos
290.
by benterix 11mo ago
> old and foreign content I don't really mean old, I'd even say old movies were "better" in the sense of more variety, also foreign is often not a problem provided it's good (my favorite genre here is comedies li
291.
by benterix 11mo ago
The stats from enclosed page are quite interesting. People speak today about depression/suicide epidemic, but when you look at that old numbers, "grief" and "hanged themselves" together give a top percentage.
292.
by benterix 11mo ago
> Would love feedback from the HN community. What other features would make this more useful? A web app
293.
by benterix 11mo ago
This is actual problem in general. As a rule, I never give my personal information online anywhere, always use fake info. However, there are a couple of cases when real info is necessary. If it's just my real name and phone number li
294.
by benterix 11mo ago
Having read the article, one thing struck me: the categorization of sexual content under "Harmful Manipulation" and the strongest guardrails against it in the models. It looks like it's easier to coerce them into providing in
295.
by benterix 11mo ago
> This is a reminder that benchmarks are meaningless – you should always curate your own out-of-sample benchmarks. Yeah I have my own set of tests and the results are a bit unsettling in the sense that sometimes older models outperform n
296.
by benterix 11mo ago
What makes you think it wouldn't end up in the training set anyway?
297.
by benterix 11mo ago
If you want a short answer: most people don't. But a more nuanced is: the term "AI" has become almost meaningless as everything is being marketed as AI, with startups and bigger companies doing it for different reasons. Howev
298.
by benterix 11mo ago
What a strange title. X11 is still more popular than Gnome, and formulating a wish like a fact doesn't make it so.
299.
by benterix 11mo ago
Does podman contain or plan to implement something similar? Seems very useful.
300.
by benterix 11mo ago
OTOH kudos to them for regretting AI slop (even if they don't want to point out who precisely is regretting). I know some who'd vehemently deny in spite of evidence.

More ›