[MichaelGG]

Why is it morally wrong? It may be a risky technical decision, but how can it be morally wrong? Would it have been better if the data was processed via a client-side browser plugin?

And Uber's "fuck the rules" are on a totally different level than sending HTTP requests to a server to aggregate publicly accessible information.

[sangnoir]

1. The owner of copyright determines how the product is to be used

2. Provider of a service can set out terms of service

3. When using someone else resource, it is polite to do as asked ("Please don't finish all the staples" or "Please don't use my stapler John. Jane can still use it")

Increasing CLs server load & increasing their bandwidth costs was morally questionable, it became clearly immoral once they knew CL didn't like what they were doing.

[rev_bird]

> The owner of copyright determines how the product is to be used

I'm not entirely sure I agree with this part. If I say, "John's website is at this address," John doesn't get to say "HEY! I get to decide who links to me," because copyright isn't for facts, it's for content. That's why I disagree so vehemently with the PadMapper decision: All they were saying was "There's a CL post for this address." That's not content, and it's certainly not enough content to replace the functionality CL sued over.

Also, FWIW 3taps wasn't using the CraigsList servers: http://arstechnica.com/tech-policy/2012/09/3taps-countersues...

[geofft]

> 1. The owner of copyright determines how the product is to be used

Padmapper mined data (apartment locations, prices, number of bedrooms, URL to Craigslist post), not creative presentation. Even if we believe that it's somehow morally reasonable for Craigslist to have copyright over the text of user postings, the data Padmapper required was clearly not copyrightable. This was a CFAA case, not a copyright case.

(Yes, in practice Padmapper also displayed descriptions and photos, but they could have avoided doing so, and the service would still have been useful.)

[ericd]

Thanks for clearing that up. We actually didn't include descriptions, though, just the title as part of the link back to the original.

[MichaelGG]

If we're talking about morals, then the fact that CL only got copyright via a click wrap assignment, and that most postings aren't copyrightable because there's no creativity, and that the actual creators (posters) benefited from this access, I don't think things held up.

I'd also be surprised to find a real increase in load, if the scraping service served end users. Otherwise those users would be requesting directly to CL. And, if the 3rd party service worked by riding on the back of users (say, via a browser add-on), would that suddenly change things?