How much influence does @sama have around here nowadays?
For the record, I was never impressed with him - I am not aware of single consequential thing he has done or built other than take the credit for the fine work of the AI scientist + engineers at OpenAI. It feels like the company is just a vehicle for how own ambition and legacy, not much else.
Look at it this way: if the community didn’t flag it, it would be the mods’ duty to get this one off the front page. So whether it was the community or the mods is incidental.
Users flagged it and it also set off the flamewar detector. I don't think we'd turn the penalties off on this one because because this article is derivative of the threads HN has already had on the recent things - threads like these:
Statement from Scarlett Johansson on the OpenAI "Sky" voice - >>40421225 - May 2024 (970 comments)
Jan Leike Resigns from OpenAI - >>40363273 - May 2024 (391 comments)
Ilya Sutskever to leave OpenAI - >>40361128 - May 2024 (780 comments)
Edit: also OpenAI departures: Why can’t former employees talk? - >>40393121 - May 2024 (961 comments)
Those were huge threads!
Sometimes media articles are driven by the topic getting discussed on Hacker News in the first place. That is: major HN thread -> journalist takes notice -> article about topic -> HN user submits article -> another HN thread—but now it's a repetitive one. We don't need that feedback loop, especially because the mind tends to resort to indignation to make up for the lack of amusement in repetitive content (https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...), and the earlier threads have been indignant (and repetitive) enough already.
> How much influence does @sama have around here nowadays?
Zero. He never asked for any change about anything HN-related even while he was running YC, and certainly not since then. Btw Sam was the person who posted https://www.ycombinator.com/blog/two-hn-announcements/.
> For the record, I was never impressed with him
(I'll add a personal bit even though that's usually a bad idea... I remember hearing this kind of comment about Sam going back to the Loopt days. My theory is that it had to do with pg praising him so publicly—I think it evoked a "why him and not me?" feeling in readers. The weird-ironic thing is that the complaint has only grown as Sam has achieved more. Running OpenAI through the biggest tech boom since the iPhone is...rather obviously massive. I think if Sam unifies gravity into quantum theory, brokers peace in the middle east, and cures cancer, we'll still be hearing these complaints—because they're not really grounded either in objective achievement or lack of it. It's some kind of second-order phenomenon, and actually rather interesting. At least if you aren't Sam!)
You're right, posting this was a bad idea. It reads like a "neener neener you're just jealous" defense of someone you happen to like.
There is, however, a more dominant rule, which is never to contradict an angry crowd, because doing so only produces more of the same. I break that rule sometimes but not often.
(Edit: s/mob/crowd. I realized on my bike ride hours later that 'mob' was too harsh.)
I suspect in such a case people may say that Sam's just using the work of other people or his employees. But then again I know nothing much of him personally and hence wouldn't really want to pick a "side".
Yeah, creating massive hype about regurgitating others' thoughts is kind of similar to becoming the warden of the world's largest digital pris...eh, walled garden.
They're both massive somethings, all right.
But you just cannot see how it perhaps reads that way because it actually is precisely that way?
I'm not saying he did not achieve anything significant, but its not clear what those things are, other then having the backing of PG and others.
I'm older and have played the game of corporate game of thrones. I have seen far too many selfish sycophants rise to leadership, only to eventually make things worse than better by using their positions as a platform for their own self interest. It's a big reason why companies like Boeing and GE become hollow shells dependent on government assistance, while companies like Costco and Alcoa last for a long time.
So I want to know what exactly sama has done to deserve prestige and recognition the he has. Bc right now, it looks like cult of personality.
> > > I think if Sam unifies gravity into quantum theory, brokers peace in the middle east, and cures cancer, we'll still be hearing these complaints—because they're not really grounded either in objective achievement or lack of it.
...certainly reads as if you thought that just because he might do a few good things, that would make all his (presumed) prior evil acts go away / be the figments of jealous imaginations. Would you say the same about, say, Hitler[1] -- if he unified gravity into quantum theory, brokered peace in the middle east, and cured cancer, should we all agree he's a great guy? Would those of us who said "That was great, thanks, but he's still an evil asshole" just be "jealous"?
If not, why should it be any different with Altman?
[1]: And no(, as I'm sure you know), that's not how Godwin's law works.
___
Side note: And I still find it rather sus that the other article, the one that came closest to exonerating him / them, was on the front page for at least twelve hours while this one (apparently, according to other commenters who had followed it) was for max two. "A coincidence that looks aforethought", as the old Swedish saying goes; it certainly didn't look less flamewarry than this, judging from the contents. But if it really was just due to the algorithm, a manual override (either way, bumping this or stomping that) might have improved at least the optics.
The difference with >>40448045 is that the latter story contained Significant New Information (SNI) relative to other recent threads. That's the criterion we apply when deciding whether or not to override penalties (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...). It doesn't have to do with who an article is for or against; it has to do with not having the same discussions over and over.
> a manual override [...] might have improved at least the optics
Sure, and we often do that (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...), but in this case it didn't cross my mind because the current thread was so obviously derivative of previous discussions that had been on HN's front page for 18+ hours in recent days.
And in any case the next day it flipped back and this story spent 16 hours on the front page:
Leaked OpenAI documents reveal aggressive tactics toward former employees - >>40447431 - May 2024 (515 comments)
... so I think we're good on "optics". The important point is that the last link (the vox.com article) contained SNI, whereas the slate.com article was a copycat piece piggybacking on other reporting . In the case of a Major Ongoing Topic (MOT) like this one, that's the key distinction: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...