dwroberts

4 ms·

31.
by dwroberts 2mo ago
One thing that grates with that style of comment though is the implication (whether intended or not) that the link/post author wasted time and could have just had an LLM do it instead. And I think it’s important that we make clear that
32.
by dwroberts 2mo ago
Dean has pretty big symbolic importance inside of Google regardless of what his role title is/was Edit: sibling story in front page links to this which explains better in its lede text than what I just wrote: https://www.nyt
33.
by dwroberts 2mo ago
Yeah I actually read this more as them scrambling because Jeff Dean is leaving? Because that is the role he is taking up
34.
by dwroberts 2mo ago
I agree, it has now become a cliche buzzword that product managers slap on boring non-technical things to make them sound important. Isn’t there even a bug tracker that some company calls an ‘issue OS’ or something absurd?
35.
by dwroberts 2mo ago
They reference gravity in this paragraph and I have no idea what it’s supposed to mean > Not only is it very hard to create interfaces that have limitations, but a bad decision is not just precedent – it’s code that can be copied and reu
36.
by dwroberts 2mo ago
The link is 404 but not just the repo - the author’s entire account appears to be gone
37.
by dwroberts 2mo ago
Can someone explain why we’re not using it to solve the obvious big conjectures/problems though? If this stuff is solved, why isn’t Riemann the first thing to go? Even if it means a fund of several hundred $k to let it churn on it.
38.
by dwroberts 2mo ago
But it says that the agentic authoring step patches the migrator when things get stuck. So while the execution of this stuff is deterministic, its actual content is not.
39.
by dwroberts 2mo ago
I really like sourcehut but I do find it frustrating that it does not offer LFS. I know I could point at another providers’ LFS, but then I may as well use their entire git offering too
40.
by dwroberts 2mo ago
Because the premier league introduced this voluntarily rather than in response to specific legislation, I doubt this kind of loophole finding will happen. A team skirting the rules with a technicality will just be called out and punished (a
41.
by dwroberts 2mo ago
They also don’t really provide an explanation for the big uptick in bugs found M146+ in the first place. Is that better testing? Or more bugs were being introduced in the first place?
42.
by dwroberts 2mo ago
I now think they’re doing it intentionally. Let the models do something outrageous and you have prime marketing material.
43.
by dwroberts 2mo ago
This is an advert
44.
by dwroberts 3mo ago
I didn't want to be rude but to be more specific: It comes across as "god this guy is insufferable, I don't blame them" especially the tone with which he asked to be dismissed/leave?
45.
by dwroberts 3mo ago
OP, feel bad for obviously unpleasant situation, but I don't think the way this is written makes you come across in a good light at all. Maybe just me?
46.
by dwroberts 3mo ago
> Riddled with irrelevant, non-essential details You’re describing natural language too
47.
by dwroberts 3mo ago
Not sure I quite get how this works from the wall plug PoV: if you have smart metering on your electricity, will it actually show you drawing less from the grid due to the input provided by the panels to your local wiring etc? Or is that it
48.
by dwroberts 3mo ago
I think the article is omitting animated features as they’re treated as family/children’s films, and not really held to the same standards as dramatic movies for adult audiences
49.
by dwroberts 3mo ago
> I can still take pride in something I “made” with an LLM You didn't make it
50.
by dwroberts 3mo ago
> European countries suck completely on their own. There are people in jail without due process, we have histories full of systematic abuse of children/people with disabilities in care units, blatant disregard for constitutional rig
51.
by dwroberts 3mo ago
Yeah heresy has ‘returned in a modern form’ - if you overlook the glaring issue of the fact that people were typically imprisoned or executed for heresy in the past. What a ridiculous false equivalence.
52.
by dwroberts 3mo ago
Not saying the paper isn’t AI, but “for example” is an extremely commonly used phrase that you should not be using as an AI indicator
53.
by dwroberts 3mo ago
This is an advert for a product, no idea why this on the front page, did anyone actually read it beyond the title
54.
by dwroberts 3mo ago
This is a really cool page, never really thought about how much information you can squeeze out of just IR (recommend selecting the dust RGB and reading the PDF it links to explaining just how dust is tracked etc)
55.
by dwroberts 3mo ago
Yeah I got basically nothing from reading it. I kind of wondered if he was contractually obligated to offer up some kind of statement, for PR or something, didn’t really care much, and a quick garbled post on X was all he could be bothered
56.
by dwroberts 3mo ago
> I know several people that have difficulty writing but are still f*cking smart. So what, they should refrain from using AI to help them because the AI police says so? Do what other people do when they cannot do something: don’t do it?
57.
by dwroberts 3mo ago
Although I’ve not been laid off (sorry to hear that), had a similar experience tending to my garden recently. Working to keep some fussy plants alive in my garden has been surprisingly therapeutic and put a lot of other stuff in perspective
58.
by dwroberts 3mo ago
When I read stuff like this, I think the people involved need to go outside and touch grass. I don’t mean that in a mean or reductive way. But something about this kind of assumption that things will get more elaborate and more abstract for
59.
by dwroberts 3mo ago
Important part before parent comment gets dismissed: > Jane Doe 4’s case shows how that pattern played out: xAI’s mandatory report to NCMEC included only the original, non-CSAM photograph, omitted every one of the AI-generated CSAM image
60.
by dwroberts 3mo ago
Linux is a much larger project receiving changes to tons of systems from lots of different sources. The combined behaviour of those things working together is massively harder to understand and test. Copyfail being introduced by an optimiza

More ›