Comments on: IRC: #boycottnovell @ FreeNode: September 9th, 2008

By: Baby In The Bath Water

Baby In The Bath Water — Wed, 10 Sep 2008 21:23:54 +0000

@Twitter: It's funny how whenever the BN fan-boys like you know their argument has lost, they immediately fall back to "Microsoft is guilty until proven innocent! I don't have to prove their guilt in this case, you have to prove their innocence!" Yea. Whatever. Luckily in the Real World, people are innocent until proven guilty. But good to know that in your fantasy land, everyone is guilty until proven innocent. Good luck with that. Note: this comment was posted from Novell's headquarters.

By: Gentoo User at Slashdot

Gentoo User at Slashdot — Wed, 10 Sep 2008 20:29:54 +0000

Not really relevant, but I noticed “twitter” from Slashdot is here. Anyone who is thinking of even listening to this person should read this:

http://slashdot.org/~willyhill/journal/205317

Note: comment arrived from a witch hunter that does not even use GNU/Linux.

By: AlexH

AlexH — Wed, 10 Sep 2008 16:55:59 +0000

Maybe it did, or maybe that’s jumping to conclusions.

By: Roy Schestowitz

Roy Schestowitz — Wed, 10 Sep 2008 16:46:14 +0000

Maybe a .NET downtime in London affected their “connectivity” with another SETS system.

By: AlexH

AlexH — Wed, 10 Sep 2008 16:43:29 +0000

They run SETS as I have already explained.

By: Roy Schestowitz

Roy Schestowitz — Wed, 10 Sep 2008 16:41:02 +0000

http://www.bloomberg.com/apps/news?pid=20601116&sid=ae0m7En9Oe98&refer=africa

I wonder what they run. It could be a synchronisation issue. To be fully convinced we must wait for proper analysis.

By: AlexH

AlexH — Wed, 10 Sep 2008 16:32:06 +0000

Circular reasoning, Roy. You said you didn’t know what caused the crash, so how does your comment about part of the system being based on .net bring any new facts to the table?

If it does demonstrate the “shoddy Microsoft system” as you claim, why did the JSE also fail? Was it out of sympathy?

By: Roy Schestowitz

Roy Schestowitz — Wed, 10 Sep 2008 16:20:54 +0000

See comment #6.

By: AlexH

AlexH — Wed, 10 Sep 2008 16:19:52 +0000

Since we don’t know what caused the crash, why did you say it “demonstrates a shoddy Microsoft system” ?

By: Roy Schestowitz

Roy Schestowitz — Wed, 10 Sep 2008 16:09:10 +0000

We don’t know yet what caused the crash and the system was down regardless. Claiming 100% uptime is false.

By: AlexH

AlexH — Wed, 10 Sep 2008 15:53:41 +0000

Sigh.

Roy, you wrote: “It demonstrates a shoddy Microsoft system [..] after those two collapses.”

You’re talking about “the system” as if it were all Microsoft’s doing. It’s not; SETS pre-dates the Microsoft system by many years and has crashed before.

By: Roy Schestowitz

Roy Schestowitz — Wed, 10 Sep 2008 15:37:45 +0000

Alex, who said anything about .NET? I’m talking about the system.

By: twitter

twitter — Wed, 10 Sep 2008 15:36:56 +0000

If a history of failure is not proof what would be, Mr. O’Brain? There’s the Hotmail debacle, Amazon search and many other examples of M$ pushing their buggy junk onto systems that worked well for years. The result is always failure. After a decade of such failure, the onus of proof is on M$ in these matters.

Reasonable doubt is not reasonable in the face of operational history. Saying M$ is not to blame for their IT failures is like saying it’s impossible to prove smoking causes cancer or that global warming is man made.

By: AlexH

AlexH — Wed, 10 Sep 2008 15:31:17 +0000

@Roy: it doesn’t matter if one part of the system was written in .net; that’s not the totality of what crashed:

“It is suspected that the problem has something to do with the LSE’s “SETS” trading platform, since the Johannesburg Stock Exchange, which uses the same platform, is also shut down.”

– http://www.clusterstock.com/2008/9/fannie-freddie-bailout-frenzy-crashes-london-stock-exchange

SETS is not .net based, nor does the JSE use the new .net-based TradElect: SETS is about 10 years old, and was responsible for previous crashes (e.g., Tokyo).

Putting the blame on .net seems a bit premature.

By: Dan O'Brian

Dan O'Brian — Wed, 10 Sep 2008 15:29:29 +0000

How does it prove that? Until you can prove that the bug was in .NET, you can’t prove either of your assertions – they both rely on the bug being in the .NET framework.

By: Roy Schestowitz

Roy Schestowitz — Wed, 10 Sep 2008 15:28:09 +0000

It demonstrates a shoddy Microsoft system and that Get the Facts campaign is false advertising after those two collapses.

By: Dan O'Brian

Dan O'Brian — Wed, 10 Sep 2008 15:24:41 +0000

It wouldn’t matter if Microsoft wrote TradeElect by itself, that doesn’t prove that the problem was in .NET.

By: Roy Schestowitz

Roy Schestowitz — Wed, 10 Sep 2008 15:22:12 +0000

“This is a custom set of C# and .NET programs, which was created by Microsoft and Accenture.“

By: AlexH

AlexH — Wed, 10 Sep 2008 15:18:12 +0000

@Dan: Indeed it’s more likely to a problem outside of .net than inside it.

@Roy: the problem is, LSE are trying to sell this system to other markets. It doesn’t make a genius to figure out why they would want to shift the blame for any problem in it.

As for SJVN’s commentary about whether or not .net can handle it: it seems that he’s made up his mind without any obvious qualifications or expertise in transactional processing. The fact is, the framework – be it .net, Java or whatever – is pretty irrelevant to this.

By: Roy Schestowitz

Roy Schestowitz — Wed, 10 Sep 2008 15:05:27 +0000

http://blogs.computerworld.com/london_stock_exchange_suffers_net_crash