The issue with scrapping is the intensity and volume of bots.
I think that nobody would care if I use wget or curl for few pages, e.g. because I would like to read a site as offline or archive it.
Btw average age of any page is 10 years. Deletion or structural change after acquisition is common, Signal vs Noise site recent wipe out could serve as an example why we need to archive sites.
2. Or issue is not solved yet by GitHub, and meanwhile bad actors gonna try vulnerability on repos. Due to number of repos there is non-zero probability. But as with scams almost nobody’s going to admit the leakage.
CO2 is just a tip as office or home is toxic environment anyway. Plastic (e.g. carpets), formaldehyde in furniture, air fresheners… add home office and cooking at home (-> small carcinogenic particles)…
If you start reading How not to die by Michael Greger, you find out that dust, soda and sitting - not CO2 - are real killers…
It's similar to how people think sharks and airplanes are the biggest killers - when in reality it is coconuts, mosquitoes, and motorcycles.
What is the difference between your tool and lets say some skill for an agent?
Doesn’t Vercel have any ingress/egress traffic pricing? (I’ve seen a project running st Mapbox and its owner had to negotiate how to get $10,000 discount after heavy monthly traffic…it wasn’t fun at first but Mapbox forgave it fortunately.)
Efficiency is the key. I haven’t written any issue before so LLM was much quicker than manual experiment. I have personally checked the result before submission.
Why would they want to do that when it’s likely that someone will find out anyway and it could turn into a scandal?
One possibility: they wanted to keep it secret because they’re crooks.
Another possibility: there are so many calls that it costs a lot in operating expenses.
Remember when we mentioned that even just saying “hello” at the start of a chat costs extra money for no reason?
So if I create a mechanism on the client side that every transmission uses 4 bytes instead of 40… it is clever way how to spend less.
Of course, it doesn’t excuse them for not mentioning it anywhere… or for hiding it on page 385 of the terms of service… like when the Vogons told Earthlings that the notice about Transgalactic Highway was in the basement on Alpha Centauri :)
I have tried to rewrite an article with GLM-5.2 and with Sonnet 4.6. Completely different results as LLM is non-deterministic. But GLM-5.2 made a lot of subtle mistakes that needed to be corrected by hand. On the opposite, Sonnet found and corrected all mistakes in the second round.
Similar situation was with planning and coding. GLM-5.2 seems to be good “on paper” but the real usage results was different.
And I am not an attorney for Claude or GLM-5.2… :)
But as I’ve been using LLM models daily since Nov 2022 I have realized that all common tests have to be confirmed in your project - there is no “one model rules them all” - you need to dig out a specific model from that LLM haystack with thousands of models.
Benchmarks help but they start to be similar to fuel consumption specs in car ads - real consumption is different for everybody :)
> Now a beginner's tutorial opens with sixteen tools you've never heard of, half of them named after Japanese words for "fast," and the first command downloads more code than the Apollo guidance computer ran on, just to render a contact form.
After finishing the post, yes OP was funny, at the beginning only unfortunately. The rest of text is so LLMish that it hurts.
I think that nobody would care if I use wget or curl for few pages, e.g. because I would like to read a site as offline or archive it.
Btw average age of any page is 10 years. Deletion or structural change after acquisition is common, Signal vs Noise site recent wipe out could serve as an example why we need to archive sites.