Version française
Home     About     Download     Resources     Contact us    

This site is updated infrequently. For up-to-date information, please visit the new OCaml website at

Browse thread
[Caml-list] Array Indexing
[ Home ] [ Index: by date | by threads ]
[ Search: ]

[ Message by date: previous | next ] [ Message in thread: previous | next ] [ Thread: previous | next ]
Date: 2004-07-31 (17:10)
From: David McClain <dmcclain1@m...>
Subject: [Caml-list] Array Indexing
I would like to offer up some blasphemous ideas in the world of

Array indexing has forever been dominated by the accountant's view that we
own only so much memory real estate and by golly your indexes into that
array better be within bounds.

But scientists realize that we live on a torus in this universe. Our arrays
are but a snapshot of the basic unit cell of a mapping that extends
infinetly in all directions. Hence array indices that extend beyond that
basic unit cell are merely referencing an identical twin array in the next
cell over. FFT's are used like pencils in this world, and our digitized
versions impose this torus mapping on us. Would that we could perform
non-digitized analysis and get perfect Fourier Transforms, but that isn't

I would like to suggest an additional option in BigArray, one of addressing
mode -- Accountant, and Scientist.

Lacking the Scientist mode means much excessive code for users, having to
check that arrays are within bounds, splitting the accesses that cross these
boundaries, etc. etc. The computer ought to be doing this tedium for us.

David McClain
Senior Corporate Scientist
Avisere, Inc.

+1.520.390.7738 (USA)

To unsubscribe, mail Archives:
Bug reports: FAQ:
Beginner's list: