[comp.ai.neural-nets] Getting papers through the net

magi@polaris.utu.fi (Marko Gronroos) (11/08/90)

        If what I'm asking is already available, I'd appreciate a
            personal email answer. Thanx.
        If not, you might want to comment this.
            If not, just dump this article and the World will be happy.. :-)

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

I already posted this to the comp.ai.neural-nets and Neuron Digest,
but there appears to be some difficulties with the news getting
outside of Finland sometimes so I'm reposting this.

? ? ? ? ? ? ? ? ? ? ? ? ?  Here is the question.. ? ? ? ? ? ? ? ? ? ? ? ? ? ?
?                                                                           ?
?    I (and propably many others) would like to be able to get all kinds    ?
?   of research papers about NNs in electronic form from a listserver or a  ?
?   FTP site. Could this be arranged somehow?                               ?
?                                                                           ?
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?

 (If there is such a possibility already, I surely haven't heard of it, sorry)


                          Why would we like it..

 It often takes a long time for papers and books to arrive to our
libraries, if they arrive at all, which is the most usual case. Also
it is very hard to find older papers anywhere. Of course we could
order them directly, but undergrads (like myself :-)) and other
lower-educated people are not supposed to do that.. And in many
universities there just are no people doing research on neural nets.
And they don't subscribe things that they don't need in their
CS Dep. libraries. Conventional CS is still 'in' in some places..
..So a possibility to get some of the material from the net would be
welcome.


                         Why they might like it..

  Many researchers write here in news about their new papers which will appear
in some future conference and offer preprints if someone wants to get a copy.
(lots of thanx to them). It would be easier for them too to send the papers as
text file to just one FTP site than to send it via real mail to everyone.
Some of them already do it, thanx to them too.


                   If you happend to like the idea...

  About the text format.. I think that the (La)TeX format would be the best
for most of the people, PostScript can't be transformed to other formats
as easily and printing and pre-viewing utilities are more
common for TeX than for PostScript (not everyone has lasers, etc.).
  Normal text might also a good choise but the formulas might be hard
to convert.

  The best FTP site might be the one that keeps the Neuron Digest
archives (@hplm.hpl.hp.com). Comments, Mr. Marvit?


                        About simulating, etc. software...

  A one FTP site would also be a good for all kinds of PD network
simulators, etc.
  I will propably put some software available to funic.funet.fi so
that people here wouldn't have to ftp to USA.

  Any information about already existing archives would be appreciated.

Thank you.

------------------------------------------------------------------------------
  Nothing that I write can't be used as an evidence against me, or the
Department of Computer Science of the University of Turku, which opinions I 
don't and won't represent - they would kill me, by hanging me from my b..
  Nothing that I write should not, must not and cannot be taken seriously.
							- Megadisclaimer