4 ms·
The title makes this sound much more severe than it is. According to the article they only get access to _public_ posts, i.e. posts they would have been able to
by DangerousPie 13y ago
The title makes this sound much more severe than it is. According to the article they only get access to _public_ posts, i.e. posts they would have been able to access by crawling the site anyway.
- gojomo 13y agoNote, though, that Facebook's robots.txt and Terms-of-Service typically prohibit the crawling of these 'public' posts. Users shouldn't be surprised that others can see them; they may still be surprised at the new level of discoverability by strangers and via other sites. Developers may still be a bit miffed at the mixed meaning of 'public' here – world-readable to the extent it benefits Facebook (and negotiated partners), but not allowing downstream automated indexing/analysis/excerpting by the general public.
- nly 13y agoHonoring robots.txt isn't a legal requirement, and I don't think a crawler can read or agree to TOS. Personally I'm in favour of anything publicly accessible being fair-game for indexing purposes, but this kind of news about database access makes me uncomfortable.
- gojomo 13y agoRe: honoring robots.txt isn't a legal requirement In common-law jurisdictions (like the US), I wouldn't be so sure of that. It has a lot of precedent behind it, as a longstanding convention for indicating site-owner/rightsholder wishes. Ignoring it – deploying software designed or configured to be oblivious to it – could create legal risk. A literal reading of copyright law and laws about 'authorized' use of computer systems would assess all bulk copying/reuse of web content without explicit advance permission as illegal. It's the force of traditional/reasonable industry practices (like robots.txt), and offsetting considerations like fair-use, that make it legally defensible.
- jonknee 13y agoWell, they're getting the firehose which is a lot more valuable (real time).