lolinder

26 ms·

421.
by lolinder 2y ago
Most people here didn't even follow the link and are just responding to the title. They don't know who wrote the piece, much less what it actually says.
422.
by lolinder 2y ago
He's already been on Meta for years now: https://www.google.com/search?q=site%3Aafterbabel.com+instag... https://www.google.com/search?q=site%3Aafterbabel.com+Facebo...
423.
by lolinder 2y ago
This is an ignorant assumption about the author's motives that doesn't withstand even the least bit of scrutiny. Meta came first under Haidt's crosshairs and has been targeted since this article was released. Snapchat was tar
424.
by lolinder 2y ago
Have you actually read anything written by Jonathan Haidt, or is this a drive-by dismissal?
425.
by lolinder 2y ago
This is Jonathan Haidt, and he's been writing against all social media for years now. This article is about TikTok, but a quick stroll through the archives shows they released an identically-titled article about SnapChat yesterday: ht
426.
by lolinder 2y ago
Which is why this outage is so weird: the entire point of paying MarkMonitor is ensuring that absolutely nothing goes wrong with a very fraught process, and they seem to have just taken down one of the biggest brands they support.
427.
by lolinder 2y ago
It's possible, but MarkMonitor is Zoom's registrar, so there are plenty of other ways for a miscommunication between MarkMonitor and GoDaddy to cause this. Copyright complaints would be a more reasonable theory if MarkMonitor were
428.
by lolinder 2y ago
I agree that both matter, but one is more important than the other. If they released the binary as "Open Source" but had a long list of things I wasn't allowed to do with it, the fact that they didn't release the source
429.
by lolinder 2y ago
See, but my point is that this is putting the cart before the horse. The "Open" in "Open Source" is what matters most by far, the same way that the "Free" in "Free Software" is the key word that quali
430.
by lolinder 2y ago
This is why I've always considered the weights-vs-source debate to be an enormous red herring that skips the far more important question: are the weights actually "Open" in the first place? If Llama released everything that t
431.
by lolinder 2y ago
How did you get started writing your own and hooking it up? Were you just working off the spec, or are there any guides or docs you'd recommend?
432.
by lolinder 2y ago
Token based is a pretty strong downside for me that would be enough to get me to use other tools like Cursor (even though I love JetBrains IDEs). I get actively stressed watching an automated system burn through my money on its own recogniz
433.
by lolinder 2y ago
> everyone will be so used to certificates changing all the time, and no certificate pinning anymore, so the day were China, a company or whoever serve you a fake certificate, you will be less able to notice it I'm a computing profe
434.
by lolinder 2y ago
Actually, the way that these things work is usually by focusing exclusively on the usage patterns of very common (top 500) words. You get better results by ignoring content words in favor of the linking words.
435.
by lolinder 2y ago
This is the Hermione Granger approach to broomstick flying, and it doesn't work.
436.
by lolinder 2y ago
Yeah, JetBrains has so far been incredibly good to their users, and it sure seems like they know that that has been their primary competitive advantage in a landscape dominated by free editors. Hopefully that calculation stays the same in
437.
by lolinder 2y ago
Where are you reading this? I see there's a concept of "credits" but don't see any explanation of what a credit is.
438.
by lolinder 2y ago
Were you able to figure out what constitutes a "credit"? I initially assumed they were following Cursor's (early) model of 1 prompt = 1 credit, with the tokens used to fulfill the prompt not costing anything. If that's h
439.
by lolinder 2y ago
> Today's president is nothing like him, but wins because a bunch of people "vote Republican". There's a component of that, but it's not the primary cause. A lot of former Republicans stopped voting Republican wi
440.
by lolinder 2y ago
There are extremely superficial similarities here, but they're just that: extremely superficial. Along the same axis but in totally different orders of magnitude, and orders of magnitude make a difference. Obamacare and communism are a
441.
by lolinder 2y ago
Even the first wave of Republican support came from the Tea Party types more than the establishment types.
442.
by lolinder 2y ago
> In truth they voted for him because he was the Republican on offer and they're die-hard Republican. The Republican party has made no secret of its agenda for decades. This is actually simply not true. The Republican party before t
443.
by lolinder 2y ago
It's not necessary in most inference engines I've seen to set the temperature to 0—the randomness in the temperature is drawn from the seed, so a static seed will work for any temperature.
444.
by lolinder 2y ago
You might, but consumers of FOSS code are extremely unreasonable.
445.
by lolinder 2y ago
Unfortunately that's not really feasible in the current state of open source. There are enormous numbers of entitled users out there who become a parasitic drain on any project that is open sourced. Solo maintainers can theoretically j
446.
by lolinder 2y ago
Not every one of those families would find the same set of features helpful, so you have to make calls about what's worth developing and what isn't. Making those calls is very difficult because it's tricky to gather data abou
447.
by lolinder 2y ago
Kind of. It doesn't measure search traffic, it measures search results. So I could see it doing that to the extent that fewer articles may be being written in certain communities than before.
448.
by lolinder 2y ago
Clearly it isn't hard to convince investors that profit will be there, they've passed that (pretty low) bar with flying colors so far. The question on everyone's minds is if investors are nuts (which they frequently are). By
449.
by lolinder 2y ago
The TIOBE index is widely regarded as mostly useless for the kind of thing people try to use it for. Their own description of the index shows the problem: > Basically the calculation comes down to counting hits for the search query +&quo
450.
by lolinder 2y ago
Who tests the tests? How do you know that the LLM-generated tests are actually asserting anything meaningful and cover the relevant edge cases? The tests are part of the code that needs to be reviewed in the PR by a human. They don't s

More ›