mjs

I don't get why there are some many overflow.com sites. It was already hard to figure out which overflow.com to post to, and now it's even harder. The only "solution" is to post to them all. Or aggregate them as overflowoverflow.com as someone else suggests so you only need to read & comment on one.

The corresponding PHP bug report:

http://bugs.php.net/bug.php?id=47665

(Closed as "not a bug". Surprise!)

What would a RESTful API for a queue resource look like? I'm thinking of something that supports push, pop, shift, unshift, etc.

Is there a meaningful difference between:

GET /user/1

GET /search?user=1

GET /?method=search&user=1

Is REST about the URLs you use, or the operations on those URLs, or both?

They fly much shorter routes. I guess the number you really want to know is the number of staff per passenger kilometer, but the graphic doesn't say...

Other flaws:

* Misspells Aer Lingus.

* Claims low cost airlines have turnarounds of "up to 25 min" instead of "as little as 25 min."

* The 43% cost advantage is unclear (that's the cost advantage per passenger kilometer, or what?) and circles are not a good way to convey amount (is the radius important, or the diameter?).

I thought about doing this, but Growl running on one system (e.g. desktop) can accept notifications generated by a different system (e.g. server), but unfortunately the client library is Cocoa only, so you can't compile it on your Linux host, etc. (I think this is all correct.) Has anyone "fixed" this?

Brad Fitzpatrick's http://code.google.com/p/pubsubhubbub/ can be used for this. ("Subscribe" to a RSS/Atom feed, it does the polling for you, and gives you callbacks when new entries happen.)

But ... you can still Google for the URL.

The conclusion is a pretty good summary, if you don't want to read the whole lot.

A big thing that's missing from Parallel DBMSs is fault tolerance, which the paper cops to (but also diminishes):

"MR is able to recover from faults in the middle of query execution in a way most parallel database systems cannot. ... If a MR system needs 1,000 nodes to match the performance of a 100 node parallel database system, it is ten times more likely that a node will fail while a query is executing. That said, better tolerance to failures is a capability that any database user would appreciate."

Also: "Although we were not surprised by the relative performance advantages provided by the two parallel database systems, we were impressed by how easy Hadoop was to set up and use in comparison to the databases."

Interesting, it's a botnet! From the FAQ: "How can the prices be so low?" "Plura pays developers to embed lightweight widgets in their desktop applications or websites. These widgets harness the idle and excess bandwidth and computing power on the computers of people using the applications and websites."

Seconded.