Follow Slashdot stories on Twitter


Forgot your password?

Comment: Re:Definitely good, but there are two sides (Score 1) 199

by dgriff (#46993897) Attached to: EU Court of Justice Paves Way For "Right To Be Forgotten" Online
Wow, don't read slashdot much these days and came here looking for some sanity but only see three comments rated 5 of which yours is one. Maybe because it's a European story. But anyway, I think you are being incredibly naive about this. What is a newspaper but a repository of stories about other people. This gives those people the ability to make such stories disappear. This is a law that the rich and powerful are going to love. Remember when the Tory party removed all David Cameron's old speeches from their website because (presumably) it might get embarrassing having their promises quoted back at them? Well you ain't seen nothing yet. This kind of thing makes China look liberal. Madness.

Comment: Java on z/OS already has segmented stacks (Score 1) 130

by dgriff (#37407212) Attached to: Neal Gafter On Java Under Oracle
Segmented stacks is an implementation issue. Java on z/OS has had segmented stacks from the very beginning because it runs on LE (the z/OS Unix layer) and that's what LE does. You still have a problem with thousands of threads though because you have to choose an initial stack segment size. Too big and you still consume loads of memory, too small and there is an overhead associated with crossing the segment boundary (e.g. page fault).

Comment: Re:The judge is an idiot (Score 1) 400

by dgriff (#37127662) Attached to: UK Men Get 4 Years For Trying to Incite Riots Via Facebook

First of all, research has shown again and again that harsh penalties simply do not work as a deterrent to other offenders.

Citation please and without it, why has this been modded up to 5? The sentence has created huge publicity here in the UK and that if nothing else should make the looter cretins think twice before posting similar incitements to riot.

And yes, it's quite possible that another riot is around the corner. NOBODY expected what happened last Monday and who's to say it won't happen again.

Comment: Why not have the matching engine on the front-end? (Score 1) 339

by dgriff (#37104408) Attached to: How Linux Mastered Wall Street
Don't quite understand this. The implication is that the time spent processing the message in the TCP/IP stack (and not the network latency itself) is a significant percentage of the overall transaction time. Which seems unlikely. Also they have (say) 1000 front-ends connecting to 100 matching engines. So why not simply add extra cores to the front-ends and do the matching there? Just cut out the network latency altogether. I know they say the traders can't connect directly to the matching engines but that's just semantics - there's no difference between having the separation physically and within the same machine.

Comment: Re:Choice is good (Score 1) 200

by dgriff (#36977374) Attached to: Is Google+ a Cathedral Or a Bazaar?

Linkedin is the only other "social media" account I have and I will never have a Facebook account and shunned MySpace when it was introduced. For me, the lack of any social decency that stems from anonymity is simply not worth it.

Unless you have some very strange friends you should give Facebook a try. Everyone I know (and as far as I can tell everyone that they know) uses their own real name.

Comment: Re:That's not Facebook's problem (Score 1) 509

by dgriff (#36707322) Attached to: Facebook Trapped In MySQL a 'Fate Worse Than Death'

Internally, the user-facing side of Facebook is in PHP. But the front end machines don't talk directly to the databases. They use an RPC system to talk to other machines that do the "business logic" parts of the system. Building a Facebook reply page may involve a hundred machines. There's heavy caching all over the system, of course, so the databases aren't hit for most read requests.

The RPC system isn't HTML, JSON, or SOAP. It's a binary system that doesn't require text parsing. Otherwise, RPC would be the bottleneck.

Um, if a single reply page involves a hundred machines, I don't think the RPC mechanism is the root cause of their problem.

The last person that quit or was fired will be held responsible for everything that goes wrong -- until the next person quits or is fired.