And thanks to Hibernate, Java kicks the collective ass of every other web enabled language out there.
And JBoss (the company) just happens to be deeply involved with Hibernate. Plus the hibernate model is pretty close to the new entity-bean model in EJB 3. Smart buy if you ask me...
T-Sql is a poor replacement for PL/SQL, but money talks.
Pl/pgSql (That's PostgreSQL's pl/sql) is VERY much like Oracles.
Naturally it lack some of it's features, but a rewrite from pl/sql to pl/pgsql is dead easy. That means less manhours... money talks:)
Sudo is a tool not the entire solution
on
Sudo vs. Root
·
· Score: 1
Sudo provides a number of features:
Issue commands as root whith the sudo prefix (and some password checking)
Logging of commands issued using sudo.
Handout of semi-root permisions to assistant operators (PFY's ?)
The first one is all about convinience. It makes it easy to be logged in as regular user and issue root commands as needed. This lessens the incentive to be logged in as root al the time and thereby can reduce the
risk of accidentially issue unfortunate commands as root.
The second is a help to figure out what went wrong in case you need to un-fsck the system after an accident.
The part about handing out semi root to PFY's is really the least interesting part about sudo. Either you trust people or you don't.
I agree that sudo becomes a hassle when you need to perform surgery but for daily tasks it's really great. If you need to be root all the time then there is something really wrong with your setup.
My use is real world and in reality, PostgreSQL is slow...
Fact: Out of the box (source install) it's slow. You need to configure/tune it to get performance.
Fact: Good performance tuning info is hard to come by and the tuning takes some time
Fact: Once you tune it right, it's blinding fast
... and a bit buggy.Nested parentheses in SQL can cause an engine crash.
Newer seen it happen in 1.5 years produktion use.
Hand-holding the query optimizer is not. Quite often, the optimizer gets the query plan wrong. Sending special commands to disable internal features is often the only resort.
The optimizer has had a major overhaul recently. I'd think you like the results.
Vacuum is a total non-issue. Nothing to see, move along.
I've personally put together a pg server that holds a 160 gig
base that is used as backend for customer self-service in a Telco. Every 15 minutes it get a load of updates from the company
main (oracle) db so there is no shortage of writes.
I'm currently (after the weekends upgrade party) on pg 8.1.3
and performance is blinding. Admin'ing that box is sooo boring
it just chuncks away. Zero issues.
The young idealists who let themselves be bought are the only ones affected. Everybody else can still fork if they have any kind of major problem. This is a non-issue.
Bull.
Two-three years ago the was a really great open source VoIP platform named VOCAL from a silicon valley startup named Vovida. Then they got bought by Cisco. Guess what. There is absolutely zero activity on the project now. Sure I could fork, but then I'd have to restart the entire development effort.
The best biplane pilots would have been eliminated by guided missles before they ever got their guns close enough to take a shot at a jet fighter. (Assuming they could catch a jet aircraft, which they couldn't.)
Even if the biplane could get it's shots in, then it's highly unlikely that they would have any effect. WW I planes fired riffle caliber ammo and with fairly low velocity. They were effective at the time because the planes were exeptional fragile.
During WW II planes got tougher and the guns shifted to around 12.7 mm. In Korea/Nam they shifted to around 20 mm again due to tougher targets. The targets (enemy planes) got tougher due to the higher loads associated with higher performance from higher speeds and higher g-loads.
Shooting down an old bi-plane would not be cost effective by the way. The missile would be far more expensive than the target.
Except that Uranium comes if far, far smaller quantities than oil, last far longer, and also comes from more stable regions of the world.
Stable regions like... Nigeria?
As far as I recal the major deposits are in: US, Australia, Former soviet union (can't remember where) and Nigeria. I guess both Australia and the US are stable enough;-)
But, damn, Lisp isn't dead-end knowledge! I "think" in Lisp all the time. It's a superset of all modern procedural programming languages! When I'm hacking Ruby or Perl I imagine it in Lisp and then translate.
Sad but true. State-ot-the-art is having a hard time catching up with the 70's. So many great ideas from Lisp and SmallTalk is still trying to get a foothold within mainstream languages. Ruby is a promising example.
And no I couldn't even spell computer when those languages where invented, but I do try to learn from history. Languages like C and Basic are the true dead-ends, were just not all the way down the road yet...
And yes, you can teach old dogs new tricks - last year I made the switch from Cold Fusion/Sybase to OracleForms/Oracle/PLI.
Please, please stay away from Oracle Forms and in particular Oracle Reports. It's cruel and unusual punishment. Unless you're into B&D and is a sub: STAY THE FUCK AWAY !!!
Don't get me wrong. Oracle's DB is cool but the rest of their stuff really is a drag...
These are EMC devices, not cheap USB harddrives. RAID-5, extra volumes, metadata, SCSI and Fiberchannel, NAS, SAN, etc. pp. Sorry, different game.
Sorry. The TB's at work ARE EMC, fiberchannel, SAN and RAID5 (would have prefered RAID10 but that's a different story).
A two tera (partition, thats after raid'ing, hot standby disk etc.) on such a beast is less than ten grands these days. We are just putting in a storage system that's 20 times larger and we are an absolutely tiny telco. So my point stands: big ISP's have way more than that on hand.
I know a major ISP in Europe who has an EMC storage with several TB of capacity.
Stop the press. An ISP with several TB of storage !!!
Get out of the 90's man. I've got half a tera at home and about five tera at the few box'es I run at work. And those are small piss-ant servers compared to what major ISP's run.
Give it a couple of orders-of-magnitude more. Then were talking major business.
The existing approach was proposed by Microsoft Research and is called Vigilante. They find that it is possible to quickly detect worms automatically, construct automatic filters for just the worm and not benign traffic, and distribute it quickly to vulnerable hosts in a secure, non-forgeable way.
Can anybody explain to me why they haven't put this into action ? They would be hailed as saviours of the (electonic) planet. Are there still a 'few kinks' to be ironed out or are we into tinfoil-hat-area ?
Jesus that's a big server for Asterisk. I've pinned up 600 calls / 60 cps with RTP (mind you, ulaw) against the echo app and sat at an average 70-80% idle on a modest old dual Xeon.
Codec and transcoding is everything when it comes to Asterisk and CPU. Try running the same setup with g.729. Hint: My box with dual 3.6 Xeons max out at around 120 calls when it needs to transcode g.729 for pstn termination. If Asterisk only needs to pass the packets along without transcoding then it can handle thousands of calls.
To a very large extend we, as IT proffesionals, don't tell the managers that we use OSS.
They don't understand it and are afraid of things they don't understand.So there is little incentive to inform them. We only really need the managers when we need their approval to buy something. So they only ever hear about things that costs money and gets a distorted view of things.
Perhaps it would have been easier for IBM to ship DB2
with a copy of McGovern and Date.
I had the distinct pleasure to discuss exactly this topic with Date
yesterday. Yes, that Date. To say that he's not pleased with the idea of XML in databases would be a very british understatement. In his words "it's a throwback to the hieracical database model, that has already been proven defect once and for all".
On top of that I would like to add that it's extremely rare that one encounters an international celebrity within the academic environment that is such a nice person.
Slashdot Mirror
OS X Leopard server looks like it's going to be big as mail/workgroup server and Apple's got a nice server for it.
For games, use windows or a PlayStation.
A life without the Internet would be like solitary confinement ...
Oh, so you've got yourself an AK47. Good for braging rigths but it's actually a crappy weapon, the epitomy of spray-and-pray.
Get yourself a decent gun and learn how to aim.
You know, there are no polar bears in Iceland. They require pack-ice which doesn't reach down to Iceland. ( http://en.wikipedia.org/wiki/Polar_bear/ ).
Been there, check'ed it, got the t-shirt to prove it.
And since they require pack-ice, they are not to fond of global warming either...
And JBoss (the company) just happens to be deeply involved with Hibernate. Plus the hibernate model is pretty close to the new entity-bean model in EJB 3. Smart buy if you ask me ...
Pl/pgSql (That's PostgreSQL's pl/sql) is VERY much like Oracles. Naturally it lack some of it's features, but a rewrite from pl/sql to pl/pgsql is dead easy. That means less manhours ... money talks :)
The first one is all about convinience. It makes it easy to be logged in as regular user and issue root commands as needed. This lessens the incentive to be logged in as root al the time and thereby can reduce the risk of accidentially issue unfortunate commands as root.
The second is a help to figure out what went wrong in case you need to un-fsck the system after an accident.
The part about handing out semi root to PFY's is really the least interesting part about sudo. Either you trust people or you don't.
I agree that sudo becomes a hassle when you need to perform surgery but for daily tasks it's really great. If you need to be root all the time then there is something really wrong with your setup.
Fact: Out of the box (source install) it's slow. You need to configure/tune it to get performance.
Fact: Good performance tuning info is hard to come by and the tuning takes some time
Fact: Once you tune it right, it's blinding fast
Newer seen it happen in 1.5 years produktion use.
Hand-holding the query optimizer is not. Quite often, the optimizer gets the query plan wrong. Sending special commands to disable internal features is often the only resort.
The optimizer has had a major overhaul recently. I'd think you like the results.
Vacuum is a total non-issue. Nothing to see, move along.
I've personally put together a pg server that holds a 160 gig base that is used as backend for customer self-service in a Telco. Every 15 minutes it get a load of updates from the company main (oracle) db so there is no shortage of writes.
I'm currently (after the weekends upgrade party) on pg 8.1.3 and performance is blinding. Admin'ing that box is sooo boring it just chuncks away. Zero issues.
Bull.
Two-three years ago the was a really great open source VoIP platform named VOCAL from a silicon valley startup named Vovida. Then they got bought by Cisco. Guess what. There is absolutely zero activity on the project now. Sure I could fork, but then I'd have to restart the entire development effort.
g.711 encoding (ie. no compression) is 64kbit PAYLOAD. You'll end up with 77kbit on the WAN link, due to RTP and IP overhead.
g.729 encoding (ie. good compression) is 8kbit PAYLOAD. You'll end up with 21kbit on the WAN.
Assuming 30ms packetization time and voice activity detection(VAD) turned of.
These number are for routers not supporting header compression. Header compression saves you around 10kbit on both g.711 and g.729.
The GSM codec as implemented in your gsm cell phone is 13kbit PAYLOAD. Just checked the ETSI spec...
Even if the biplane could get it's shots in, then it's highly unlikely that they would have any effect. WW I planes fired riffle caliber ammo and with fairly low velocity. They were effective at the time because the planes were exeptional fragile.
During WW II planes got tougher and the guns shifted to around 12.7 mm. In Korea/Nam they shifted to around 20 mm again due to tougher targets. The targets (enemy planes) got tougher due to the higher loads associated with higher performance from higher speeds and higher g-loads.
Shooting down an old bi-plane would not be cost effective by the way. The missile would be far more expensive than the target.
I like JOnAS.
DOD tried that with Ada. Didn't do them a lot of good...
That's by-the-way a text book case of poor change management. They made the change gradually ...
Stable regions like ... Nigeria?
As far as I recal the major deposits are in: US, Australia, Former soviet union (can't remember where) and Nigeria. I guess both Australia and the US are stable enough ;-)
Sad but true. State-ot-the-art is having a hard time catching up with the 70's. So many great ideas from Lisp and SmallTalk is still trying to get a foothold within mainstream languages. Ruby is a promising example.
And no I couldn't even spell computer when those languages where invented, but I do try to learn from history. Languages like C and Basic are the true dead-ends, were just not all the way down the road yet...
Please, please stay away from Oracle Forms and in particular Oracle Reports. It's cruel and unusual punishment. Unless you're into B&D and is a sub: STAY THE FUCK AWAY !!!
Don't get me wrong. Oracle's DB is cool but the rest of their stuff really is a drag ...
Sorry. The TB's at work ARE EMC, fiberchannel, SAN and RAID5 (would have prefered RAID10 but that's a different story).
A two tera (partition, thats after raid'ing, hot standby disk etc.) on such a beast is less than ten grands these days. We are just putting in a storage system that's 20 times larger and we are an absolutely tiny telco. So my point stands: big ISP's have way more than that on hand.
Stop the press. An ISP with several TB of storage !!!
Get out of the 90's man. I've got half a tera at home and about five tera at the few box'es I run at work. And those are small piss-ant servers compared to what major ISP's run.
Give it a couple of orders-of-magnitude more. Then were talking major business.
Can anybody explain to me why they haven't put this into action ? They would be hailed as saviours of the (electonic) planet. Are there still a 'few kinks' to be ironed out or are we into tinfoil-hat-area ?
I am the VoIP provider.
Codec and transcoding is everything when it comes to Asterisk and CPU. Try running the same setup with g.729. Hint: My box with dual 3.6 Xeons max out at around 120 calls when it needs to transcode g.729 for pstn termination. If Asterisk only needs to pass the packets along without transcoding then it can handle thousands of calls.
They don't understand it and are afraid of things they don't understand.So there is little incentive to inform them. We only really need the managers when we need their approval to buy something. So they only ever hear about things that costs money and gets a distorted view of things.
I had the distinct pleasure to discuss exactly this topic with Date yesterday. Yes, that Date. To say that he's not pleased with the idea of XML in databases would be a very british understatement. In his words "it's a throwback to the hieracical database model, that has already been proven defect once and for all".
On top of that I would like to add that it's extremely rare that one encounters an international celebrity within the academic environment that is such a nice person.