in reply to More About Memory -- case study

If you want to know what all this is for, how would you describe (as COMPLEX AS YOU WANT OR NEED to) how the alphabet is stored in your brain, in Perl terms as much as applicable.

(You obviously don't store it as an array...)

$_="goto+F.print+chop;\n=yhpaj";F1:eval

Replies are listed 'Best First'.
RE: RE: More About Memory -- case study
by Blue (Hermit) on Oct 19, 2000 at 23:43 UTC
    The best way to describe it would be to say that I store the alphabet as a singly linked list, with a few pointers to "well used" parts so that I don't always start at the beginning. Pointers to "A" (the song), "L" (el-em-en-o-pee), "T" (I guess because of the hard sound), and "X" (xyz).

    When going backwards, I did a section, went back to my closest pointer, went forward until I found my last backwards, keeping as much as I could in a small short term memory space (say an array of 5 to 7 elements).

    As a side note, doing the alphabet forward was not traversing each element as I needed it, but a rather efficient block load which pulled in a "mouthful" of elements, spitting them out autonomously so that there were no I/O blocking.

    Damn I feel geeky. 8)

    =Blue
    ...you might be eaten by a grue...

[reply]
      <cite> When going backwards, I did a section, went back to my closest pointer, went forward until I found my last backwards, keeping as much as I could in a small short term memory space (say an array of 5 to 7 elements). </cite> Yes, this is because you know the proximity of some of the letter. The node for 's' might look like: 
      s => {
  next => 't',
  near => 'r',
}

      [download]


      $_="goto+F.print+chop;\n=yhpaj";F1:eval
[reply]
[d/l]
        I agree. I'd put it a little more like:

        s => {
  next => 't'
  nearPrev => 'qr'
  nearNext => 't'
}

        [download]
        With the size of the nearPrev and nearNext dependant on the letters. M would be in the whole "lmnop" cluster.

        For numbers, I would also have a "prev" which is undefined for letters.

        Now, I've found all of this facinating (really!), but how can use something like this? We've talked about associations for numbers, letters, and words.

        One interesting project would be a Perl program that plays word association, and learns from playing with others. Instead of just a "next", it would be most interesting to me if it had a "context" it tried to stay in (usually) also.

        Well, I've got a weekend in front of me...

        =Blue
        ...you might be eaten by a grue...

[reply]
[d/l]
RE: RE: More About Memory -- case study
by japhy (Canon) on Oct 20, 2000 at 00:23 UTC
    I'd think it's a hash that looks like a linked list: 
    %alphabet = (
  a => 'b',
  b => 'c',
  c => 'd',
  # ...
);

    [download]
    This is because I can start ANYWHERE in the alphabet and continue to the end. But the structure is really a bit more complex than that, since I do know the POSITION of some of the letters, as well as the letter that comes before it. So it's a partially linked list. You'll see my report soon enough...

    $_="goto+F.print+chop;\n=yhpaj";F1:eval
[reply]
[d/l]
RE: RE: More About Memory -- case study
by AgentM (Curate) on Oct 19, 2000 at 23:33 UTC
    How about a doubly-linked list with "hash characteristsics"? The back links are "slower" than the forward links. I say "hash characteristics" since it is unneccessary to count to 5 in order to use the number...Any other ideas? Update: maybe even a hash with linked list characteristics?
    AgentM Systems nor Nasca Enterprises nor Bone::Easy is responsible for the comments made by AgentM.</h6>
[reply]

In Section Meditations