Re: The ultimate luxury ?

From: Jesse F. Hughes (jesse_at_phiwumbda.org)
Date: 08/03/04 

Date: Tue, 03 Aug 2004 14:38:43 +0200

jmfbahciv@aol.com writes:

> In article <87n01co5pm.fsf@phiwumbda.org>,
> jesse@phiwumbda.org (Jesse F. Hughes) wrote:
>>jmfbahciv@aol.com writes:
>>
>>> In article <87u0vlk7st.fsf@phiwumbda.org>,
>>> jesse@phiwumbda.org (Jesse F. Hughes) wrote:
>>>> .. As far as I
>>>>can see, there's no reason to have a special term "sort" meaning "sort
>>>>alphabetically or numerically" and use a distinct term "collate" for
>>>>all other arrangements of data according to some order on a field of
>>>>the data. The distinction isn't particularly useful to me.
>>>
>>> I noticed and I've been thinking about this. I fear it's a side
>>> effect of having too much memory, too much disk storage and CPUs
>>> that are too fast. There isn't much advantage, ala wallclock
>>> time, to sort then process data records.
>>
>>None of those are relevant to my background, which is only about
>>theoretical CS largely unencumbered with worries like feasibility.
>
> [jarred emoticon hearing fingernails on blackboard] ;-)
>
>>Sometimes, a theorist is interested in modeling a problem without
>>considering time or space constraints.
>
> I understand. However, those models don't print paychecks which
> is how most CS grads make a living.

Sometimes they do.

> In the early 80s, if a resume had CS degree printed on it, we didn't
> even bother to set up to interview the guy. We didn't have the time
> or money to untrain him.

Whatever. I'm not applying for a job. But one of us is laboring
under the misapprehension that newsclients regularly fetch new
articles via Message-ID and References and one of us is not laboring
under that misapprehension.

Or, if it's *not* a misapprehension, perhaps you can give me some
evidence? I can show you the file my client keeps regarding which
articles I've read. It consists only of article numbers, not
message-ids.

>>No doubt.
>
> Before this post, I was going to point you at a site. It doesn't
> have anything to do with what were talking about. But you
> probably won't be impressed with what the guy did :-).
> I'll type in the address anyway for the lurkers out there.
>
> http://home.ecn.ab.ca/~jsavard/comp/cp01.htm
>
> I haven't checked it out but the guy's reliable and knows his
> stuff. He just announced it; there have been no bug reports.

I like John Savard, I think. At least, I have vaguely positive
memories of the name.

But, no, I don't care to read this article. Sorry.

>> ...Sadly so. Other times, the old distinctions were dropped
>>with good reason. In this case, I still don't get the importance of
>>the distinction you're keen on drawing.
>
> Specs and how people implementment them. If I collated a deck of
> cards, I could get different results than if I sorted them.

Then *what the **** does collate mean? You have never said. Here's
a deck of cards. Right here. Take it. Now collate it. What's the
result? Okay. Now sort it. What's the result?

Want me to sort it? Okay. According to what order? Give me the
order on the deck, and I'll sort it. (If the order depends on the
initial state of the deck, then I may not be able to give an explicit
description here, but that isn't relevant to our discussion.)

>>Call me a stupid optimist.
>
> Nope. I'd call you foolish and dangerous around code. Bit
> gods, the best software developers in the world, are good only
> because they are the most paranoid people in the world.

Bit gods wouldn't spend a week debating the distinction between words
like "collate" and "sort" and they wouldn't be paralyzed if their
correspondent used the "wrong" word. 

-- 
Jesse F. Hughes
"And a journal can beg me for the right to publish it [...] because
I'd rather see it in "People" magazine [...]" 
     --James Harris on his simple proof of Fermat's last theorem
Quantcast