bonoboTP

33 ms·

661.
by bonoboTP 8mo ago
Yes, but that better not be all the time, and around basic questions.
662.
by bonoboTP 8mo ago
It's very hard to discuss this. To some people it's obvious, to some it isn't. To me, every single paragraphs is obvious fluff AI writing. One problem with it is the repetitiveness and the schmoozing salesman feel. The other
663.
by bonoboTP 8mo ago
> If you have something to say about the text then say it. I could point out the individual phrases and describe the overall impression in detail, or I can just compactly communicate that by using the phrase "AI". If it bother
664.
by bonoboTP 8mo ago
It's 19 years since the iPhone came out, that's almost two decades. 19 years before the iPhone was 1988. Things from 1988 definitely seemed dated in 2007. In fact I think style/aesthetics change is now getting slower and slow
665.
by bonoboTP 8mo ago
This shows the downside of using AI to write up your project. I see the eloquent sentences, but don't get the message. > This works, but the actual execution happened outside the model. The model specified the computation, then wait
666.
by bonoboTP 8mo ago
> That's why replication has to be required and standard. "has to be required"... This is a passive construct. Who will do the requiring and what precisely will motivate them to such a change and what will get them the buy
667.
by bonoboTP 8mo ago
It's interesting how people never even learn about any upsides to that. Even if the balance comes out on the side of elected officials, it's good to at least have some idea of why so many societies have worked like that (other tha
668.
by bonoboTP 8mo ago
Realistically, the vast majority of people will not have a real chance to "refute" or even evaluate scientific claims. Maybe given a lot of time and foundational work to learn the field, some percentage of people can usefully thin
669.
by bonoboTP 8mo ago
Any system that grows too fast has these kinds of problems. When it's a small intimate circle where everyone knows everyone, reputation alone can keep people in check. Once it's larger you need to invent rules and bureaucracies an
670.
by bonoboTP 8mo ago
Regardless of what gets taught in school about science being objective and without ego, or having a culture of adversarial checks on each other etc., the reality is that scientists are humans and have egos and have petty feuds. Publishing a
671.
by bonoboTP 8mo ago
Someone has to pay for all this. That someone is most often not a scientist themselves. They don't have a that vague intuitive research taste that scientists have. Beyond fairly trivial levels of technical correctness, the value of res
672.
by bonoboTP 8mo ago
There is a sliding scale from that, to it being the LLM that communicates, not the person. LLMs can really reshuffle and change priorities and modify emphasis in a text. All the missing pieces will be filled in and rounded out and sandpaper
673.
by bonoboTP 8mo ago
Humans have more variability and "edge". If a person is passionately arguing for some point of view (perhaps somewhat outside the usual), it signals to me that they probably thought about this and it is a distillation of a long th
674.
by bonoboTP 8mo ago
The question is, why does anyone take the word of a company seriously which will agree with any idea if you pay them? After several iterations of this game (decades by now), someone would surely say "nah, we don't care about these
675.
by bonoboTP 8mo ago
I think it's difficult to relay to the public that a lot of this noise in "scientific publications" is not the same category as real research by reputable institutions. Yes, in certain cases the line can be blurry, fraudsters
676.
by bonoboTP 8mo ago
> Number of released papers/number of citations is a target Only in stupid university leaderships is that truly what gets you hired or promoted. It's simply not true. Junior researchers in fact are believing it stronger than th
677.
by bonoboTP 8mo ago
When, in those mythical non-"modern" times, was it easy to get tenure or a livable wage as a researcher? How open were the doors to this and what proportion of society got a realistic chance to pursue such a career? More people ge
678.
by bonoboTP 8mo ago
You either have something documented and quantified and measured and objective criteria tickboxes and deal with this style of failure mode, or you rely on subjective judgment and assessment and accept the failure mode of bias, nepotism, old
679.
by bonoboTP 8mo ago
Plenty will publish it, but those are not as highly regarded by the community. It's not a problem of journals. It's not hard to start your own journal by teaming up with other academics. In machine learning, ICLR is such a venue f
680.
by bonoboTP 8mo ago
How can it be that what you just wrote is such a widely known fact? I've been reading this and hearing this from consultancy people as well for many years now. If the guy lacks the political power, why don't his internal political
681.
by bonoboTP 8mo ago
I would think that most people can think of some. Maybe those who can't are part of one.
682.
by bonoboTP 8mo ago
People have widely different experiences regarding this. All I can say is that Claude is working great for me for doing drudgery sysadmin stuff but I'm also somewhat experienced in these things and that helps in telling it specifically
683.
by bonoboTP 8mo ago
Blogs kinda dwindled in importance as a whole. Substack brought it back to a degree, through email distribution, which is a more familiar technology to regular people compared to RSS. But even Substack is becoming more of an algo feed based
684.
by bonoboTP 8mo ago
Maybe not. But academia is going to change. Status will still have to be allocated by some mechanism but the classic journals and reviews based system will crumble under the weight of LLMs. Of course this will upset a great many of people w
685.
by bonoboTP 8mo ago
True, snap annoys me as well, but it's possible to switch packages to apt. And it's just going to get easier to customize using AI agents.
686.
by bonoboTP 8mo ago
It's likely based on just the transcript, even if it describes visual things, it likely guesses those things from the transcript text only. Maybe it's better now, but that was how it did it recently. To be convinced that it "
687.
by bonoboTP 8mo ago
It's great. Now that fancy writing is cheap and infinite, fields whose entire scholarship value was in obscurantist jargon bending have to actually start to turn on their brains and care about making more sense than an LLM can.
688.
by bonoboTP 8mo ago
It was called learning already back when the field was called cybernetics and foundational figures like Shannon worked on this kind of stuff. People tried to decipher learning in the nervous system and implement the extracted principles in
689.
by bonoboTP 8mo ago
I think thats about the first era of Apple. They faded in the background in the consumer mind from the mid 90s to the mid 00s. It was the iPod/iPhone/iPad trilogy that brought Apple back to the mainstream. In ~2002 for regular peo
690.
by bonoboTP 8mo ago
If it works, great. My comment was aimed at a person who seemed to want more freedom and troubleshooting possibility.

More ›