Re: VMs: words shrinking

[Gabriel Landini <G.Landini@xxxxxxxxxx>]

On Wednesday 02 February 2005 10:12, Marke Fincher wrote:
> It's been said once or twice that the VMs word lengths tend to get
> shorter towards the end of a line (or something similar).

If you count each eva or FSG character as a letter, yes, they are shorter. 

> Is this true and widely accepted?  

I think it is true because I did it in 1997 :-) (maybe others have done this 
earlier or later than that). This is easy to reproduce.

> Does anyone have any figures/data on this effect?

It was originally in my page (Zipf's laws in the vms) but this is not 
accessible anymore and I have not set a new vms home yet).
A plot of the vms token and word length distributions in eva is in my 
Cryptologia paper, but there are no other languages to compare there (the 
intention was to show the differences between the vms and a 'random' text 
with the same character and space distributions as the vms). I can send you a 
preprint of that and a couple of graphs from the original page. Just contact 
me directly, because the list does not carry attachments.

Cheers,

Gabriel

______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list