[demetris]

I don’t believe resips will be with us for long, at least not to the extent they are now. There is pressure and there are strong commercial interests against the whole thing. I think the problem will solve itself in some part.

Also, I always wonder about Common Crawl:

Is there is something wrong with it? Is it badly designed? What is it that all the trainers cannot find there so they need to crawl our sites over and over again for the exact same stuff, each on its own?

[ccgreg]

Many AI projects in academia or research get all of their web data from Common Crawl -- in addition to many not-AI usages of our dataset.

The folks who crawl more appear to mostly be folks who are doing grounding or RAG, and also AI companies who think that they can build a better foundational model by going big. We recommend that all of these folks respect robots.txt and rate limits.

[demetris]

Thank you!

> The folks who crawl more appear to mostly be folks who are doing grounding or RAG, and also AI companies who think that they can build a better foundational model by going big.

But how can they aspire to do any of that if they cannot build a basic bot?

My case, which I know is the same for many people:

My content is updated infrequently. Common Crawl must have all of it. I do not block Common Crawl, and I see it (the genuine one from the published ranges; not the fakes) visiting frequently. Yet the LLM bots hit the same URLs all the time, multiple times a day.

I plan to start blocking more of them, even the User and Search variants. The situation is becoming absurd.

[ccgreg]

Well, yes, it is a bit distressing that ill behaved crawlers are causing a lot of damage -- and collateral damage, too, when well-behaved bots get blocked.