News for hackers. Stuff that matters.
Re:SPARQL Protocol for RDF by rspeer on Tuesday June 20, 2017 @10:25PM (#14600217)
Yeah, that's kind of how APIs have to work. It turns out it's not feasible or desirable to just let arbitrary people make arbitrary queries. SPARQL doesn't make it any more feasible, it just ignores feasibility.
Re:SPARQL Protocol for RDF by Vinnl on Wednesday June 21, 2017 @03:00AM (#14601486)
> It turns out it's not feasible or desirable to just let arbitrary people make arbitrary queries.

GraphQL/Relay and Falcor are trying to make it feasible. And I think it's definitely desirable from a consumer point of view, at least.

Re:SPARQL Protocol for RDF by xg15 on Wednesday June 21, 2017 @03:47AM (#14601737)
> It turns out it's not feasible or desirable to just let arbitrary people make arbitrary queries.

It turns out where and why exactly?

Re:SPARQL Protocol for RDF by rspeer on Wednesday June 21, 2017 @04:01AM (#14601810)
Denial of service.
Re:SPARQL Protocol for RDF by derefr on Wednesday June 21, 2017 @09:18PM (#14609426)
You can just 408-response queries after they execute for a given number of CPU cycles. And then rate-limit per-account using evercookies + browser fingerprinting + IP hashing. You can often disallow anonymous queries altogether, while you're at it; if your data is valuable, people are usually willing to register at least a pseudonymous account to get at it.