SCLeo

As much as I understand their decision, I hope they didn't stop. I don't use Facebook myself but now I am worried YouTube now knows that if they try a bit harder uBO would just give up.

> I'll risk sounding antagonistic and ask you this: if LLMs are not "good enough", why are they still around?

To quote the parent commenter:

> They seem to be able to solve the immediate problem,

but not long term

As many have pointed out, I think the cartel is just using the engineering perspective for justification. If avoiding racing to the bottom is the goal, the rule could be that "no one can advertise a lifetime longer than 1000 hrs" as opposed to actually limiting the lifetime.

Oh, I solved this long time ago by only opening the discussion page, never read tfa, and pretend I did. /s

I am not up-to-date in this area, and not necessarily that I don't trust you, but do you have a source for this? Just curious.

I cannot agree more. Cloudflare has some services that are really cheap (r2) to lure you into their worker "ecosystem", which is just serverless. Once you are vendor locked into their absolute garbage custom JavaScript runtime, you are pretty much forced to use their distributed database Cloudflare KV if you want good performance. Cloudflare KV is so extremely ridiculously absurdly expensive that make predatory pricing of vercel that HN likes to complain about feel like child play.

And every PR turns into a triple recursive reverse merge.

Oh that's why. I didn't connect the dots until I see your message. I had like 5 vscode open and all of them were spamming GitHub PR extension requiring login alert at me.

I actually laughed out loud in a restaurant.

Frequency of actions matter, especially for security changes. If we are talking about git, I agree. If we are talking about npm, I bet 95%+ times people install packages in order to use them, not just to admire the code.

Someone else in this thread mentioned that npm can be used to manage pure front end libraries, which is a fair point.

I don't get it. How does this help with anything? You pull in a dependency to use it, right?

People are so different. When I was in college, if I had an unsolved problem, I could not fall asleep.

They should rewrite it in rust again to get another 3x performance increase /s

There are explicit residual connections in a transformer block. Look up "residual connections" in Google images and you will see.

Thanks for the explanation. Honestly, your explanation is better than the entire video. - I watched it in full and got really confused. I completely missed the part where he said the light is pulsing at 30kHZ and was really puzzled at how he is able to move the mirror so fast to cover the entire scene.

Does using a constraint solver actually solve the question under the time ... constraints?

If not, how can you claim you have solved the problem?

Where do you even get the $3,000 standing desk? I am don't even compare prices and I got mine from Amazon for $200-$300. Sure the quality might not be the best but I just can't see there are people buying $3000 standing desks.

> And as someone who's into spy stories, I know that a big part of tradecraft is of formulating your questions in a way that divulges the least about your actual intentions and current information.

Not necessarily disagreeing with you, but if everyone started doing this, we will be in XY problem city.

I don't think they used AI nor claimed to have used AI. The interviewee just said "We monitor abnormal deformation of the buildings in real time."

Yeah, the commonjs to esm transition has been the python 2 to python 3 transition of JavaScript, except the benefits are limited (at least compared to the hassle created).

There are many libraries that have switched to esm only (meaning they don't support commonjs), but even today, the best way to find the last commonjs version of those libraries is to go to the "versions" tab on npm, and find the most downloaded version in the last month, and chances are, that will be the last commonjs version.

Yes, in a vacuum, esm is objectively a better than commonjs, but how tc39 almost intentionally made it incompatible with commonjs (via top-level awaits) is just bizarre to me.

This comment might be one of the meanest comment I have ever seen.

This explanation is extremely misleading. For many of my projects, vast majority of the operations do not use the database . And the few that does, contains long running huge joins/aggregates. Using the sync API is just straight up terrible because the task will block literally everything else that does not use the db in js, meaning generating a report in the background can literally prevent you from handling any requests. (I did end up using better sqlite 3 because they are personal projects and getting stuck for 2 seconds when the scheduled report generation happens is ok ish for the a few people using it. But I will not consider using better sqlite 3 for any future projects)

I believe they wrote their own image handling and did not contribute back to llama.cpp.

Is this website freezing for anyone else?

EDIT: yes it is.

I feel so stupid to left out h1-h6. Also completely forgot about the entire table related tags.

I think most people just start typing "HashM..." and trigger auto complete, and hit tab to import and auto complete HashMap.

Oh wow, this is probably the most creative thing I have seen for a while. Thanks for sharing.

If you have a namespace, can't people just globally-claim namespaces instead? like serde/serde or something similar. I feel if you really don't want people claim whatever they want, you have to do the Java package style where namespaces are tied to domain names.

“lossless image format" maybe?

I think all other methods require a reflow as well (internal or external). Because at the end of the day, you still need to measure the height of the text before rendering a larger box.