The proper answer would have been "test the theory of evolution."
I talked to professor of genetics of the local Uni here, University of Hohenheim about this and he said he is not very interested in evolution since there is *no way* to prove it scientifically. For that you would have to be able to reproduce the whole thing. He is much more interested in theories that can be proved.
On the "triumph of reason": "Reason" is not to accept the most popular theory. But that's what many ppl do. They read it in Nature and Scientific American and./ and boing! it's a fact.
Anyway I urge ppl to read read and read stuff about this from both sides, only that can enlighten you and lets you make an informed decision. If you jump into discussion make sure your discussion partner understands your semantics.
If you stay in.de,.ch or.at, the most likely clearing point is DE-CIX in Frankfurt. The backbones here are quite well connected, so IMHO we wouldn't notice it THAT much as long as we stay European.
I'm a bit late, but I was fighting the same problem. A few hints:
you almost certainly are using ipmasq'ing. Set the timeout for tcp connections with ipchains higher than the keep-alive time of sockets in the kernel. Standard for ipmasq tcp timeout is 15mins or so, but the keepalive time is 2 hours! (see two posts above on how to do this.
if your target machine runs a 2.0.x kernel: I still have problems with that one. I don't know why. Upgrading to 2.2.x is probably the best solution.
Does the dev software matter? No, it's closed source, WinNT only. Very stupid license (AFAIR). Costs heaps of money.
Does the client software matter? No, reference running on WinCE *only*. Costs on "case-by-case" basis.
Does WAP matter? No, in 9 months there will be phones that have the displays and the computing power to do HTML. IP space is private to the provider who masqs them to the internet or they use ipv6.
Above all: Why is this story on slashdot ? The news is old. The software is available for years now. Nothing about it is open-source or connected to Linux or *BSD. Did I miss something ?
I spent last christmas hacking a websystem together in java, with PostgreSQLas db.
PostgreSQL may have come a long way, but it's definitely lacking. Esp. a working and complete JDBC driver.
The manual tells you about how easy lobs are with getAsStream() but when you use those functions you get a "Not implemented" error. GREAT! Fortunately there was a LOB handling through custom classes, but it didn't work as advertised and a manual that lies to me is very frustrating.
All data coming into JDBC are strings. Fine. But the date conversion doesn't work if you live in ECT/CET. So you have to parse the string yourself. You tell the author of the JDBC driver and he says "Oh didn't know that, I'll look at it." Haven't heard from him since. Last jdbc driver update was September 14 1999, jdk1.2 only.
Lots of stuff is not implemented in the JDBC driver, like the meta stuff.
Then postgresql itself.
Limited tuple size, limited query size (8k each) in 6.5.3.
One failed operation (say, an insert) will abort the whole transaction, all the previous work also.
LOB migration (to other db installations) was only possible by writing a whole application.
The docs suck, the mailing list interface even more, searches don't work.
Slashdot Mirror
These umbrellas (http://www.tenarafabric.com/medina.html) seem to be working quite well in Medina.
No, even Linus got it: /usr/src/linux-2.4/kernel/exit.c:658
panic("Aiee, killing interrupt handler!");
(FC1, 2.4.22-1.2197.nptl)
Now you understand why it was necessary to write
"He became flesh..." (John 1)!
The proper answer would have been "test the theory of evolution."
./ and boing! it's a fact.
I talked to professor of genetics of the local Uni here, University of Hohenheim about this and he said he is not very interested in evolution since there is *no way* to prove it scientifically. For that you would have to be able to reproduce the whole thing. He is much more interested in theories that can be proved.
On the "triumph of reason": "Reason" is not to accept the most popular theory. But that's what many ppl do. They read it in Nature and Scientific American and
Anyway I urge ppl to read read and read stuff about this from both sides, only that can enlighten you and lets you make an informed decision. If you jump into discussion make sure your discussion partner understands your semantics.
uhm, not really. I saw a 0:thoor account in /etc/passwd seconds before they pulled the plug.
Somebody rooted the box. Too sad. Why did they use SunOS ? OpenBSD would have been a much better choice. (I use primarily Linux, ok ?)
If you stay in .de, .ch or .at, the most likely clearing point is DE-CIX in Frankfurt. The backbones here are quite well connected, so IMHO we wouldn't notice it THAT much as long as we stay European.
Does the dev software matter?
No, it's closed source, WinNT only. Very stupid license (AFAIR). Costs heaps of money.
Does the client software matter?
No, reference running on WinCE *only*. Costs on "case-by-case" basis.
Does WAP matter?
No, in 9 months there will be phones that have the displays and the computing power to do HTML. IP space is private to the provider who masqs them to the internet or they use ipv6.
Above all: Why is this story on slashdot ?
The news is old. The software is available for years now. Nothing about it is open-source or connected to Linux or *BSD. Did I miss something ?
One action fails -> transaction should be rolled back. Everything else would violate ACID.
thx
yes, Mysql user. And MiniSQL. And Oracle. And Solid.
PostgreSQL may have come a long way, but it's definitely lacking. Esp. a working and complete JDBC driver.
- The manual tells you about how easy lobs are with getAsStream() but when you use those functions you get a "Not implemented" error. GREAT! Fortunately there was a LOB handling through custom classes, but it didn't work as advertised and a manual that lies to me is very frustrating.
- All data coming into JDBC are strings. Fine. But the date conversion doesn't work if you live in ECT/CET. So you have to parse the string yourself. You tell the author of the JDBC driver and he says "Oh didn't know that, I'll look at it." Haven't heard from him since. Last jdbc driver update was September 14 1999, jdk1.2 only.
- Lots of stuff is not implemented in the JDBC driver, like the meta stuff.
Then postgresql itself.- Limited tuple size, limited query size (8k each) in 6.5.3.
- One failed operation (say, an insert) will abort the whole transaction, all the previous work also.
- LOB migration (to other db installations) was only possible by writing a whole application.
- The docs suck, the mailing list interface even more, searches don't work.
Mysql may suck, but it's a good at that.