AW: VMs: introducing myself and 2 questions :)

["Sebastian Unterreitmeier" <sebastian@xxxxxxxxxxxxxx>]

>Welcome to the group! :-)

thanks :)

>Or you can find one of the various mirrors to the no-ip.com scans - there
>was one on http://www.flamingjesus.com/voynich/ for example - and download

oh sh.... those who need to click at all these links will end with a carpal
tunnel syndrome at the end *g*
i downloaded the jpegs an did a short compressing test with photoshop:
its possible to make them about ~30% smaller without any notable loss in
quality. i would do the compressing and zip them in one file, but:
is it legal to offer them for downloading?

>There are two immediate barriers here:-
>(a) [tech-centric]: how can you transform a large mailing list archive into
>something more "useful"?

i thought about making a mysql-db with a entry for each mail. fields would
include author, date and so on. so each mail would become a single
databaseentry.
these entrys would be easily searchable so this would be an answer to the
question:
>(b) [user-centric]: what do we mean by "useful"?

>To achieve this, the first task needing doing would be to make the
>1991-2001 mailing list archives accessible as individual posts (rather than
>as huge flat text files) - unzipping these from

hm, its possible by coding a fetcherprogramm that cuts the textfiles at each
mails beginning and end. i have some experience with this, because a few
months
ago i needed to transform html-pages from a nonexisting db-system into a
mysql-db.
its not that hard, but you need a separatortext where you can see, that an
old
mail ends and a new mail starts. i had a short look at the txtfiles and
maybe
this could be "From ".

>to try it for yourself. Similarly, I'm not sure that a mySQL db would scale
>for complex queries on such a large dataset without grinding your machine
>*really* hard.

right, that could be a problem...

>set up, but just to ensure that everything is accessible via Google, and
>then to write a tiny wrapper around the Google API - or (simpler still) a
>JavaScript form in a top frame to compose search strings to send to Google
>(ie, returning results in a separate frame). This is no more than making
>Google power-user search tools a little bit more accessible - let someone
>else's PC take the strain. :-)

cool idea, but we need to feed google first.
we could cut the mailarchive in pieces like i described above and generate
a html-page for each email and than link it from several webpages so that
google
will index it, because afaik google dont index dynamic pages?

>(which wraps around Google Advanced Search features, or around the Google
>API), you could keep an accessible *history of searches* made by other
>Voynich researchers.
>You might think about developing that history of searches into a
>categorised search engine wrapper - search by page, search by name, search
>by location, search by topic, etc. This is like a kind of "passive
>WikiWiki" - gradually building up a search structure over time.

thats really a good idea...

>Anyway, despite having more questions than answers (as per normal), I hope
>this is of at least some help to you. :-)

well it is at least something that will keep me busy thinking about the
whole
weekend and my girlfriend will wonder why i take the dog for a walk 4hours a
day *g*

kind regards,
sebastian
--
www.sinandsoul.com

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