Re: The ultimate luxury ?
From: Jesse F. Hughes (jesse_at_phiwumbda.org)
Date: 08/03/04
- Next message: Esnesqw: "The Gravity Transformations and the Creation of a Universe"
- Previous message: markd_at_toad-net.com: "Re: 'APPROACH ASTROLOGY WITH AN OPEN MIND'"
- Maybe in reply to: Kenneth Doyle: "Re: The ultimate luxury ?"
- Next in thread: mitch perkins: "Re: The ultimate luxury ?"
- Reply: mitch perkins: "Re: The ultimate luxury ?"
- Messages sorted by: [ date ] [ thread ]
Date: Tue, 03 Aug 2004 15:54:00 +0200
jmfbahciv@aol.com writes:
>> .. but my introduction to
>>binary trees was in an out-of-sequence assembler course (not really a
>>course in assembler, but in OS fundamentals or something -- it's been
>>a while).
>
> Ah, that's where you might be stumbling. This isn't about binary
> trees. It's about putting data in order vs. putting collections
> of data in order and the rules used to order them.
Which do you think is which?
Message-IDs are a means of identifying posts, but it's not the means
by which either the server or the client primarily identifies them.
That's the article number.
>>> Specifications aren't precise. I'm finding lots of youngsters who
>>> can't think in terms of bit flows. I'm really disturbed that you
>>> hadn't heard of collating. There used to be an IBM machine that was
>>> a stand-alone collator.
>>
>>I'm *still* not at all clear what you meant by collating that is
>>different than "sorting on some order on a field of a record." (To be
>>fair, the order I defined required two fields, but that's
>>inessential.)
>
> I was just told that the hardest thing one of guys has to do is
> teach the kids about sorting on sub-fields.
Well, kids today. What ya gonna do? Meanwhile, it is sorting on
sub-fields that seems to be the issue at hand, and you've spent
considerable time arguing whether it's a "sort" or a "collation".
>
> Collating takes two decks of cards which contain data and
> making one deck based on an arbitrary set of rules that
> examines a piece of the data contained in each card.
>
> Merging takes two card decks and makes one card deck based
> on a set of rules that have nothing to do with the data stored
> on the cards; there doesn't have to be anything stored on
> the cards.
>
> Sorting takes one card deck, and arranges that deck using
> the data stored on each card in the same relative postition,
> in an ascending or descending order, alphameric, numeric, or
> alphanumeric.
Fine. Whatever. I don't really care about your distinctions much,
aside from making sense of this conversation. Use whatever jargon you
prefer, but as far as I'm concerned, sorting data is essentially
mapping a set to a linear order. That's the intrinsic problem to be
solved. That mapping may or may not be created by lifting a mapping
to the set from some field or set of fields. All of this really isn't
consequential.
Especially, these distinctions which are so stunningly important to
you have nothing to do with the fact that Relf's subject changes do
not trouble those of us who sort/collate/flib-flub on references
rather than subject.
Let's drop this tedious aside.
-- "I'd step through arguments in such detail that it was like I was teaching basic arithmetic and some poster would come back and act like I hadn't said anything that made sense. For a while I almost started to doubt myself." -- James S. Harris, so close and yet....
- Next message: Esnesqw: "The Gravity Transformations and the Creation of a Universe"
- Previous message: markd_at_toad-net.com: "Re: 'APPROACH ASTROLOGY WITH AN OPEN MIND'"
- Maybe in reply to: Kenneth Doyle: "Re: The ultimate luxury ?"
- Next in thread: mitch perkins: "Re: The ultimate luxury ?"
- Reply: mitch perkins: "Re: The ultimate luxury ?"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
