James Wanless is a known sci.math loon, and can be safely ignored without missing anything of merit.
I'm sure there's plenty of case history on googlegroups. Oooh, I joust noticed that a googlegroups search for: "James Wanless" idiot brings up one of my posts as the top hit!
"If it were designed [badly, it would] not serve the purpose intended"
Which is exactly grandparents point. Almost all error popups I've seen have been badly designed. Even more annoying than the ones that just have "[OK]", are the ones that have "[OK]" _and_ "[Cancel]". These ones even pretend to be getting one bit of information from you, when in fact there's no behavioural difference between the choices. However, the first time you see such a dialog box you have to scratch your head and think "what might it possibly do differently if I select OK rather than Cancel". Humans should _not_ have to try and guess the outcome of their actions.
I have 3 machines in my linux compute farm that _share_ a single ordinary CDROM drive. I boot knoppix (yuck!) on each machine, pull all the binaries I'm likely to need (ssh, scp, ls, su, rm etc.) into the cache, start my client processes and the just yank out the IDE cable, and go to the next one.
"Also Times magazine elected Hitler as man of the year in 1938"/Time/ magazine did indeed nominate him man of the year, but if read the right way it was quite a damning article. Some of it is pretty unambiguous: """ Meanwhile, Germany has become a nation of uniforms, goose-stepping to Hitler's tune, where boys of ten are taught to throw hand grenades, where women are regarded as breeding machines. """
I fail to imagine an interpretation where that would be considered positive to US readers, for example.
He was undeniably one of the most influential people (in a bad way) of the decade, and having him nominated as man of the year was possibly one of the last great example of the 1900s US's mastery of irony. (Earlier, with Bierce and Mencken you had some of the finest ironic humourists in the world.)
Topshelf was countering the following: "Most of the electoral votes are in the heartland of the US" I believe his 4 examples accurately puncture that misapprehension. Nowhere does he indicate that those 4 on their own would be a majority. Also, PA is not a coastal state, but you'd _really_ be pushing the boundaries to call that a heartland state.
If you look at the ECV-proportional map at
http://synapse.princeton.edu/~sam/ev_prediction_1n ov.jpg you'll see that much of the "heartland" is if anything disproportionately _small_. IL and OH being the two biggest heartland states, I guess, but even together they're dwarfed by CA.
I know that one large telecommunications company, from Finland, still uses 1, 50MHz HPUX server as a login server and mail server for 150 people in one of the departments.
So stick with the 4 old 386es, it's about the same ooomph, and even has built in redundancy!
""" TCO numbers are computed using a specialized form of mathematics where the operations exist in a mathematical field called "The Bullshit Plane" So you have the field of integers, the field of complex numbers and then the field of bullshit numbers where TCO numbers, presidential polling, WiFi access point range and Best Buy extended service contracts are computed. """
I have nothing to add. It simply needed repeating, as it's a complete classic. Hmmm, it's a bit too long for a usenet.sig, but maybe not...
I remember: a) being the slap-happy "don't send around these without verifying" employee (drinking mates with the sysadmin, I was almost encouraged to do this, as it took the responsibility off him) to all the lamers in the company. b) Talking to the techies about quite how such things could be done. This was before HTML mails, and before JavaScript was well known. But we could see it coming, as a _feature_, not just as a bug.
Best of all - they enabled exploits that didn't even require you to read the mail to become infected - so no "don't open" warnings would even be any use!
They must reboot once every <49.7 days (2^32 * 1/1000s). Their instructions tell them to reboot once a month. Some new guy thought that the beginning of one month and the end of the following month would suffice, but alas it doesn't.
There was a great comp.risks article about the ATC systems, and their laughable fallbacks, about 2 months ago, ISTR. I.e. it's worse than it first appears; worse than just occasional windows lock-ups.
What about the 16000+ PPC cluster that was announced a short while ago? Oh yes, it was slower than this 8000+ I2 cluster.
You wouldn't be spouting bullshit, would you?
The original Itanium was crap, but the I2, through the injection of a fair bit of DEC Alpha brains, is an enormously capable processor, both singly and in clusters.
Slashdot Mirror
James Wanless is a known sci.math loon, and can be safely ignored without missing anything of merit.
I'm sure there's plenty of case history on googlegroups.
Oooh, I joust noticed that a googlegroups search for:
"James Wanless" idiot
brings up one of my posts as the top hit!
FP.
"If it were designed [badly, it would] not serve the purpose intended"
Which is exactly grandparents point. Almost all error popups I've seen have been badly designed. Even more annoying than the ones that just have "[OK]", are the ones that have
"[OK]" _and_ "[Cancel]". These ones even pretend to be getting one bit of information from you, when in fact there's no behavioural difference between the choices. However, the first time you see such a dialog box you have to scratch your head and think "what might it possibly do differently if I select OK rather than Cancel". Humans should _not_ have to try and guess the outcome of their actions.
FP.
I have 3 machines in my linux compute farm that _share_ a single ordinary CDROM drive. I boot knoppix (yuck!) on each machine, pull all the binaries I'm likely to need (ssh, scp, ls, su, rm etc.) into the cache, start my client processes and the just yank out the IDE cable, and go to the next one.
Some have been up for _months_ like this!
FP.
I'd forgotten this bit -
"the greatest threatening force that the democratic, freedom-loving world faces today."
It really wasn't a positive article at all.
FP.
"Also Times magazine elected Hitler as man of the year in 1938" /Time/ magazine did indeed nominate him man of the year, but if read the right way it was quite a damning article. Some of it is pretty unambiguous:
"""
Meanwhile, Germany has become a nation of uniforms, goose-stepping to Hitler's tune, where boys of ten are taught to throw hand grenades, where women are regarded as breeding machines.
"""
I fail to imagine an interpretation where that would be considered positive to US readers, for example.
He was undeniably one of the most influential people (in a bad way) of the decade, and having him nominated as man of the year was possibly one of the last great example of the 1900s US's mastery of irony. (Earlier, with Bierce and Mencken you had some of the finest ironic humourists in the world.)
FP.
Topshelf was countering the following:
n ov.jpg
"Most of the electoral votes are in the heartland of the US"
I believe his 4 examples accurately puncture that misapprehension. Nowhere does he indicate that those 4 on their own would be a majority. Also, PA is not a coastal state, but you'd _really_ be pushing the boundaries to call that a heartland state.
If you look at the ECV-proportional map at
http://synapse.princeton.edu/~sam/ev_prediction_1
you'll see that much of the "heartland" is if anything disproportionately _small_. IL and OH being the two biggest heartland states, I guess, but even together they're dwarfed by CA.
FP.
"Personally I'm predicting a blowout for Kerry. "
?!?!? If you're doing 65mph on the interstate and you get a blowout that's a really bad thing. Crash and burn...
The rest of your paragraph says "walkover" to me, so I imagine this is some slang I'm not familiar with.
FP.
Chrysler? You potatoer, you.
This guy's a true ricer!
FP.
Are the two '>' predicates needed in tri()?
Note, I've never even _seen_ any prolog before, I'm trying to work out how it might work (prolog, that is) from your code.
It just that you have those conditions in edge(), so it looks like they're redundant in tri()
FP.
_webmail_ is a real problem for anyone who doesn't want their email being read by others.
FP.
I know that one large telecommunications company, from Finland, still uses 1, 50MHz HPUX server as a login server and mail server for 150 people in one of the departments.
So stick with the 4 old 386es, it's about the same ooomph, and even has built in redundancy!
FP.
"""
.sig, but maybe not...
TCO numbers are computed using a specialized form of mathematics where the operations exist in a mathematical field called "The Bullshit Plane" So you have the field of integers, the field of complex numbers and then the field of bullshit numbers where TCO numbers, presidential polling, WiFi access point range and Best Buy extended service contracts are computed.
"""
I have nothing to add. It simply needed repeating, as it's a complete classic. Hmmm, it's a bit too long for a usenet
FP.
Amen.
I remember:
a) being the slap-happy "don't send around these without verifying" employee (drinking mates with the sysadmin, I was almost encouraged to do this, as it took the responsibility off him) to all the lamers in the company.
b) Talking to the techies about quite how such things could be done. This was before HTML mails, and before JavaScript was well known. But we could see it coming, as a _feature_, not just as a bug.
Best of all - they enabled exploits that didn't even require you to read the mail to become infected - so no "don't open" warnings would even be any use!
FP.
They must reboot once every <49.7 days (2^32 * 1/1000s). Their instructions tell them to reboot once a month. Some new guy thought that the beginning of one month and the end of the following month would suffice, but alas it doesn't.
There was a great comp.risks article about the ATC systems, and their laughable fallbacks, about 2 months ago, ISTR.
I.e. it's worse than it first appears; worse than just occasional windows lock-ups.
FP.
Unless your transformative hermeneutics tell you otherwise.
http://www.physics.nyu.edu/faculty/sokal/
FP.
Many a true word spoken in jest.
I dunno, on a scale from nought to two, I'd give her one.
*tish boom*
FP.
"Nintendo is the more "kid friendly" or "family oriented" than the other game systems"
/that/ kid friendly.
But I've just spent half an hour running round a cave murdering realistically-animated people on my N64. I
don't call
Nintendo are simply another bunch of litigious bastards.
They're not the first, and they won't be the last.
I'm glad that the US takes its first amendment very seriously, as I can't see this case even getting to a court.
FP.
"To beg the question" can be used as a transitive or intransitive verb.
Transitively, it's raising the question (that follows as the direct object).
Intransitively, it's committing the logical falacy.
Great-grandparent was transitive.
FP.
I'm perfectly familiar with the PPC lineup.
I'm also perfectly familiar with the English language.
Therefore I interpret:
"gosh, almost any processor except an Itanium would be even faster"
as being the uninformed comment. In particular given that a counter-example was so recent.
FP.
Finland 404s too.
What about the 16000+ PPC cluster that was announced a short while ago? Oh yes, it was slower than this 8000+ I2 cluster.
You wouldn't be spouting bullshit, would you?
The original Itanium was crap, but the I2, through the injection of a fair bit of DEC Alpha brains, is an enormously capable processor, both singly and in clusters.
FP.
Dang, that system's fast!
u fa cturing/Columbia_Mfg_5.jpg
ftp://shell.sgi.com/collect/ProjectColumbia/Man
Only one hand on the camera, perhaps?
and highly efficient drivers to utilise the capabilities of the hardware, of course.
(Though I'd agree the hardware is a far bigger job than the drivers.)
FP.