No, the ANSI SQL way to do this is to use MERGE. Unfortunately, pgsql doesn't support that yet. It's on the todo list so I'm sure if someone got out their checkbook and wrote the pg developers a check we'd see it soon enough.
Zing! Hole in one!
The big problem with those chips (the 8xxx series AMDs) is that they have an 800MHz memory bus, and you just can't pump enough data into and out of them at that speed to keep them all busy. I've got a quad 12 core magny cours, and it can pump around 80 to 100Gigabytes per second into / out of memory. If they want to borrow some of my down time to benchmark I'll gladly cooperate.
I would expect the French to be outraged at this after that incident. Requiring no un-breakable encryption allows foreign govts to snoop on French companies and sell secrets all over again. Lots of packets pass through networks in other countries that have the technological capability to take your breakthrough idea and run with it. Suddenly North Korea is the largest producers of hoverboards and we're all a decade behind suddenly.
Correction, For some reason it took Microsoft decades to CARE...
No, it goes like this:
Stupid (L)user: It's not running automatically! See it's broke!
You: No, you just do this (dblclick my computer) and this (dblclick removable drive) and this (dblclick application)
Stupid (L)user: Why do I have to do that?
You: Because autorunning things when you put in a USB key is a security hazard and can cause your computer to get infected with a virus or trojan.
Stupid (L)user: OK.
There. 5 minutes tops.
How about the Internet Coke Machine?
Yeah, nobody uses it for anything useful or anything.
Except skype.
And Afilias (the guys what run the
And Cisco.
And the USGS.
Wait, is that really a released, supported, and ready for production version? I thought mysql was still working on getting 6.x out the door. Cool if it is.
Enterprise DB implements a 95% or so solution to the oracle compatibility thing, including plsql.
http://www.enterprisedb.com/exposure/oracle-postgres_wp-1.do
eMachines D620, just bought a new battery.
acpi -i
New Battery:
Battery 0: Unknown, 99%
Battery 0: design capacity 6600 mAh, last full capacity 6514 mAh = 98%
Old Battery:
Battery 0: Charging, 38%, 11:43:18 until charged
Battery 0: design capacity 4500 mAh, last full capacity 3564 mAh = 79%
Runs ~1.5 hours on old, about 3 hours on new.
I run a small server farm that takes up all but about 2U of a full sized rack. In this rack we have lots of different drives. They're all rated for use in servers etc against vibration, for all the good that does. Here's the failure rate over time for most of them
32x Seagate 15K5 147G SAS drives 2 have failed in ~2 years
4x Seagate 2.5" 73Gig 10krpm server drives 1 failure in ~3 years
8x WD 7200rpm 2TB SATA drives (black models) 2 dead in ~1 year
2x WD 7200rom 1TB SATA drives (black models) 0 dead in ~1 year
8x Seagate 1TB SATA 5400 RPM drives. 2 failures in 6 months.
10x Seagate 500G SATA 5400 RPM drives 4 failures in 6 months.
Lots more different drives, but none have been as much problem as these shit-tastic 500G and 1TB Seagates of late. Those drives were trash and should have never left the factory.
That's like a car that comes with the airbags in the trunk and if you install them properly you'll likely not die in a collision. Is it really that hard to make it do that all the time? It's easy enough to be disabled by the client on mysql as well. Reminds me of Bill Gates blaming most Windows crashes on users.
Now, if you are a GPL purist this won't worry you, since you'd rather shave your own head with a cheesegrater than use proprietary code (and you're probably using PostgreSQL anyway)
If you are a GPL purist, you are almost certainly not running the BSD-licensed PostgreSQL...
Either is fine by me, as long as it keeps me from having to use Cisco.
8 Catfish = 1 Octo-puss
