jbellis

56 ms·

691.
by jbellis 16y ago
Firefox 4 removing the menu bar is a one-off attack on the same problem. Better to solve it once than require each application to come up with its own fix.
692.
by jbellis 16y ago
That game is _hard_. Thexder HD was released on PSN recently, btw.
693.
by jbellis 16y ago
By the time someone feels that pain he's already decided to leave and it just pisses him off.
694.
by jbellis 16y ago
(Sorry about the accidental downvote. Arrows + phone got me again.)
695.
by jbellis 16y ago
Note that deb and rpm packages are available, too. http://wiki.apache.org/cassandra/DebianPackaging http://rpm.datastax.com/
696.
by jbellis 16y ago
Link to the last time this was posted: http://news.ycombinator.com/item?id=2237158
697.
by jbellis 16y ago
(Last changelog entry: 2005.)
698.
by jbellis 16y ago
One very interesting aspect that I haven't seen mentioned yet is that Netflix does a LOT on AWS. I wonder if seeing their infrastructure provider start competing more directly with them, makes them rethink that.
699.
by jbellis 16y ago
Hey Alex, I was hoping you'd jump in. :) So why would I use ElephantDB instead of Voldemort read-only storage?
700.
by jbellis 16y ago
Cassandra supports reading/writing MapReduce jobs natively ( http://wiki.apache.org/cassandra/HadoopSupport ), and Voldemort supports read-only k/v views of Hadoop data ( http://project-voldemort.com/blog/2009/06/building-a-1-tb-da... ). B
701.
by jbellis 16y ago
Upvoted, but what TFA is claiming is that codine w/o acetaminophen is roughly as affective as the combination, which isn't addressed in that analysis.
702.
by jbellis 16y ago
... and large-paste support. ... and backlog replay when you go offline. ... and searchable history. ... and (like most IRC clients, but unlike Campfire) built-in notification support. You can get some of these with IRC, if you have a serve
703.
by jbellis 16y ago
"If you're a web designer for local business, they would not be fine with it." That's kind of the definition of "clients that are not worth having."
704.
by jbellis 16y ago
bq. Why not climb on their shoulders and adopt the WAL approach The WAL approach solves a different problem: "how do we provide durable writes with minimal performance impact?" WAL is not queryable, so the question of "how do we efficiently
705.
by jbellis 16y ago
+ Cassandra :)
706.
by jbellis 16y ago
bq. My understanding is that whatever database AWS supports it will let you scale up as much as required If by "as much as required" you mean "to larger single EC2 instances," yes. If you mean backing it with a SAN or transparently partitio
707.
by jbellis 16y ago
In fairness, you don't have to tune MongDB poorly (deliberately or otherwise) to get poor performance with a workload involving substantial numbers of writes; it's well-documented that there is a global lock ( http://www.mongodb.org/display
708.
by jbellis 16y ago
Well, that's partly right. Yes, clustrix is more expensive than the open-source databases. But, companies like Twitter and Netflix certainly have the budget for Exadata (which is what clustrix wants to be when it grows up); they are using
709.
by jbellis 16y ago
Agreed. Even where you can optimize the hot code in C, Python is no speed demon. Cassandra's java stress test can push out about 10x as many ops/s as the python one, even though Thrift C extension for Python is quite good. /still a Python
710.
by jbellis 16y ago
> error messages from MUMPS Is fixing that tractable? Cryptic error messages is one of the most frustrating turn-offs to new would-be contributors. /probably one of the few people here who has written MUMPS code
711.
by jbellis 16y ago
Sure... until your database is your bottleneck.
712.
by jbellis 16y ago
The release announcement has links to posts by Riptano and Rackspace engineers with more details on the new features: http://article.gmane.org/gmane.comp.db.cassandra.user/11569
713.
by jbellis 16y ago
There's a strong team behind a new competitor to Blackbord, http://www.instructure.com/ . Decent article about them: http://mfeldstein.com/instructure-canvas-a-new-lms-entrant/
714.
by jbellis 16y ago
A couple years ago I bought two copies of that book to give away as jokes because the Engrish is just that awesome. Both recipients were appropriately baffled. :)
715.
by jbellis 16y ago
If you're running a "real" cluster (i.e., where you actually need multiple machines instead of just one big one running postgresql), then yes, the recommendations there are realistic -- not because cassandra is inefficient but because you w
716.
by jbellis 16y ago
FWIW, if you're looking at HBase alternatives because of stability or reliability concerns, Cassandra is probably going to be an easy move because of the common BigTable datamodel ancestor. (With some notable improvements, e.g., http://ww
717.
by jbellis 16y ago
http://en.wikipedia.org/wiki/Loss_leader
718.
Apache Cassandra in 2010: Code, Community, and Controversy(spyced.blogspot.com)
35 points by jbellis 16y ago|2 comments
719.
by jbellis 16y ago
http://www.emacswiki.org/emacs/EmacsReleaseDates What changed to go from releases every 4-7 years, to releases every 1-2 years?
720.
by jbellis 16y ago
> facebook is an independent entity - it's not quite the same comparison qualitatively, I don't see much difference between investing VC cash vs investing parent company cash.

More ›