[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Gallowspace
On 28 Jul 2001, at 19:39, Giddy Landan wrote:
>
> Gabriel wrote
> > The function of p and f has been discussed many times, I think that
> > originally related to the gallows as shown in Cappelli's dictionary.
>
> Ref./URL ?..GD
Have a look in the mail archives. There used to be a link to a scan
from Cappelli somewhere, but I haven't got it now.
> > I don't think that p and f are the spaces and the rest aren't. Of
> > course I cannot prove it,...
>
> On the other hand, can we prove that spaces are spaces? Have anybody
> tried to identify the spacer character statistically? ..GD
I think that the spaces are spaces. Last year I did some spectral
analysis of the vms and other texts without spaces and that
analysis seems to indicate the modal token length (in the absence
of spaces!). The results for the vms in EVA were as follows:
For the spaces coded vms the modal token is 5 characters,
>From the "line length distribution", there is a peak at 6 characters
(those are isolated strings, the tokens in the text).
For the space-removed vms analysis, there is a peak is at 5.9
characters.
A preprint of the paper is at:
http://web.bham.ac.uk/G.Landini/evmt/cryptol.zip
In that version of the document, the tau peak is wrongly stated as
5.6 in the legend of figure 8b, but it is correct in Table 3 and in the
text.
> Gallowspace will transform it into:
>
>
> +-O--+ +-R-+
> +--+ +--+ +----+
> | +-SO-+ +-E-+ |
> + |
> | +-C--+ |
> | | | |
> | +-CC-+ | +-O--+
> | | | |
> --+--+-SC-+--9--------++-++-4O-+--.-->
> | | | |
> | +-S--+ | +-SO-+
> | | | |
> | +-Z--+ |
> | | | |
> | +----+ |
> | |
> | +-AE-+ |
> | | | |
> +--+-AJ-+-----------+
> | |
> +-AM-+
> | |
> +-AN-+
>
Perhaps the +-40-+ string should be at the left hand side of the
diagram, since 4 is almost always word-initial.
> Which is much more neat for my visual perception of a written
> language.
Why?
Cheers,
Gabriel