[odf-discuss] Mars: XMLisation of PDF - opportunity for ODF?

[Jerry Askew]

You are probably looking for one of these two documents:

A Mathematical Theory of Communication
http://cm.bell-labs.com/cm/ms/what/shannonday/shannon1948.pdf

The physical limits of communication:
http://www.santafe.edu/research/publications/workingpapers/99-07-054.pdf


-Jerry

-----Original Message-----
From: odf-discuss-bounces at opendocumentfellowship.org
[mailto:odf-discuss-bounces at opendocumentfellowship.org] On Behalf Of
Daniel Carrera
Sent: Thursday, November 09, 2006 12:37 PM
To: ODF Discussion List
Subject: Re: [odf-discuss] Mars: XMLisation of PDF - opportunity for
ODF?

On Thu, 2006-09-11 at 10:55 -0800, marbux wrote:
> There is also a very long-term data storage issue of the data not 
> being stored in ASCII. If you look ahead 1,000 years from now, today's

> documents stored in PDF will not be in human-readable form absent 
> software that can render PDF. Of course the same is true of ASCII, but

> the more we can standardize on a single human readable format, the 
> fewer burdens we impose on our archaeologist descendants. ("Virtual 
> archaeology," anyone?)

Absolutely. ASCII has a much better chance of being readable to our
descendants. Not only that, but to read PDF you need to know ASCII
anyways (the text is ultimately in ASCII).

I hope our descendants can also figure out zip compression. That one
will be more difficult. The best we can do is to use a very common and
standard compression format (zip). If everyone uses zip, there's a
better chance our descendants will figure it out.

>From an archaeological point of view, using compression is a bad thing.
And the more advanced the compression system, the worse it is for our
descendants. There is a paper titled, "any sufficiently advanced
communication is indistinguishable from noise". As compression improves,
the data looks more and more like random noise. Think about it: suppose
that there is a pattern that makes the data look different from noise.
You can use that pattern to compress the data even further.

Dang... I've been trying to find that paper online. It's from Claude
Shannon from the 1940s...

Cheers,
Daniel.
--
"I AM in shape. Round IS a shape."