(Haphazard collection of messages from Summer '94 through Summer '96, and even from '91-'94; plus a slightly less haphazard collection of later messages.)
IMPORTANT NOTICE TO SUBMITTERS: beginning 1 JAN 1996, all new submissions will be passed automatically to the postscript generator (tested on-line for the past six months) and rejected in the event of errors (virtually all errors are now user errors: misspelled macro names, missing style files, etc.). [users who supply only the final postscript output from tex source are again reminded that this is a poor and much less portable choice (produced at fixed resolution without hyperlinks), and later in january the www interface will offer other resolution (300/400/600 dpi) and font choices (unimbedded type 1 PS), as well as dvi and pdf (U.S. taxpayer dollars busily at work).] This change will mean in particular that the "figure" and "add" commands will no longer be supported, and multi-part submissions should instead be packaged as a single gzipped (or compressed) tar file. In addition, the internal storage formats here will be switched to flat .tar.gz files (and the entire archival database will be back-translated sometime during january). For much additional information (including pointers to the necessary utilities on various platforms), see neworder, or via e-mail, get neworder.txt (please read before e-mailing questions or comments)
The E-print archives at xxx automatically generate PostScript from TeX source. During the test phase, authors should confirm that their TeX source can produce PostScript automatically. This can be done either by e-mail or via the www interface. Either procedure will run an automated script on your TeX source (and attachments) to try to generate (hyper-)PostScript, and you will receive a message in response indicating success or failure. If the archive can't easily tex your source, then it is unlikely that anyone else can either (making your submission at best inconvenient, at worst useless). If you can't figure out what the problem is, send a comment to the archive in question using the `comment' command. If you don't fix your submission, then everyone who tries to get the ps version of your paper will receive a message stating that it is "author error", and that the author has been notified of the problem. Whenever you replace your paper, you should re-issue the psgenerate command to verify that your submission still works. Notes:More info on recently implemented auto (hyper-)postscript generation.
- At the time of the daily mailings, auto ps generation will be run on any submissions whose authors have not already done so, but by then there may be a longer delay before authors can correct trivial errors.
- There is a remarkable correlation between low quality content and unprocessability.
- Incidentally, if you `put' only PostScript thinking that it was more convenient than TeX for readers, please think again. TeX + attachments is a preferable format for archiving your paper [more information can be found in formats.txt, available via the `get' command] and allows automated hyperlinkages [see e.g. hypertex] for viewing either via hyperPS browsers [such as newest version of ghostview] or inclusion via distillation to pdf [viewable by adobe acrobat and soon to be incorporated into netscape and other www clients].) Rather than doing your readers a service, by submitting PostScript you instead do them a disservice by submitting something substantially less portable (and user-supplied postscript has more than its share of problems -- the postscript generated here is uniformly of higher quality than generated elsewhere).
- After the test phase, we will migrate (probably sometime in late december) to a simpler one-step submission format, which will allow gleeful rejection of submissions that cannot be processed automatically. The current awkward system is only intended as a temporary remedial measure to facilitate a smoother changeover to the more efficient target system. Watch this space for further details.
> Date: Thu, 2 Mar 1995 11:14:31 +0500 > From: ri@einstein.mps.nsf.gov (Richard Isaacson) > ... > (a) $$ This archive is based upon activities supported by the > U.S. National Science Foundation under Agreement No. 9413208 > (1 Mar 1995 thru 28 Feb 1998). $$ > (b) Any opinions, findings, and conclusions or recommendations expressed in > this archive are those of the author(s) and do not necessarily > reflect the views of the U.S. National Science Foundation.
NEW ARCHIVE: quant-ph@xxx.lanl.gov (fundamental aspects of quantum theory)
help text begins: "Theoretical, experimental and philosophical papers
announcing, explaining, clarifying or interpreting quantum phenomena and
their applications ..."
for continuation, e-mail To: quant-ph@xxx.lanl.gov
Subject: help
NEW ARCHIVE: q-alg due to unfortunate historical reasons, the e-print archive hep-th has long been something of a grab-bag for otherwise unplaceable submissions. as many have complained, this has left it less useful to its core consituency of formal high energy particle theorists than would be a more focused archive. with resources allegedly soon to emerge from gov'tl limbo, certain decoupled subconstituencies can at last be partitioned off to their own dedicated forums. first to go: a new math archive q-alg for "quantum algebra" (i.e. q-anything, including for the time being knot theory) will operate under the address q-alg@eprints.math.duke.edu (also available via msri.org). as described in more detail in the trailer attached below, papers written for mathematicians should be directed to q-alg . (if you are not sure whether you are a physicist or a mathematician, then it is appropriate to use q-alg and not to cross-list to hep-th). next: quant-ph@xxx (fundamental problems of quantum theory) and math-ph (watch this space)
The hep-lat archive has been moved to xxx.lanl.gov as the primary site -- for the time being it will continue to be mirrored to ftp.scri.fsu.edu.
Starting 1 Nov '94, submitters should be certain to use a "structured" title/author field, e.g. \\ Title: Recent Seminal Results Authors: Author One and Author Two Comments: 3 pages, latex, no figures Report-no: EFI-94-11 \\ abstract \\ body of paper
three notes:
1) there is a new version of uufiles (see attached message below)
to enhance the quality of your life
2) long little-known feature:
the www interface automatically translates appearances of url's of the
form http://hostname/path or ftp://hostname/path that appear as plain
text in titles or abstracts into hot links. for example:
"for associated mpeg file, see http://myhost.domain/file.mpg"
when accessed via www pages at xxx.lanl.gov (or babbage.sissa.it) views as
"for associated mpeg file, see <this URL>"
(for some examples, search cond-mat via www for the string http )
3) for the purposes of automatically compiling references and citation lists
(e.g. a significant aid to the slac spires-hep database), it is useful
to include the citation to the arch-ive/papernum in your reference list
when known (e.g. a string such as hep-ph/9409201 is easily harvested).
this will also facilitate automatic network hyperlinks of references
from within papers (see e.g. http://xxx.lanl.gov/hypertex/ ).
fourth note:
it is possible to retrieve just the past week of title/author listings
a) via e-mail, Subject: list pastweek
b) via anon ftp, as the file arch-ive/listings/pastweek
c) via www, linked from arch-ive home page or from overall form page
Due to many reliability complaints, and the difficulty for Los Alamos personnel to administer the astro-ph cond-mat and funct-an archives properly from remote, these archives have been moved from babbage.sissa.it to xxx.lanl.gov as the primary location. They will continue to be mirrored to babbage.sissa.it. This will at long last ensure the same level of functionality and reliability as the rest of the archive system.
For physics Job/Conference announcements, point your www client at
http://xxx.lanl.gov/Announce/
This system has long been inundated by requests to disseminate physics job
and conference announcements, but has consciously restricted to
research dissemination alone.
But with the widespread usage of WorldWideWeb clients, automated short-term
archival systems are now significantly easier to establish and maintain.
Posting to the above is via www and can either be full plain text or can
include hyperlinks to further information maintained at remote servers.
(Entries can be removed at any time by original submitter.)
There are no plans at present to establish a parallel e-mail interface
(we could do it, but it would be wrong) since form compliant www clients
exist for all platforms (for more info see the www faq,
http://sunsite.unc.edu/boutell/faq/www_faq.html ,
also available via e-mail e.g. To: hep-th@xxx.lanl.gov Subject: get wwwfaq ).
Note also that the above may serve as well as the prototype for an eventual
public commentary system associated to the e-print archives (with full author
control over attachments to any given submission).
note that there is a new version (10/94) of uufiles (used for tarring,
compressing, and uuencoding files for e-mail transmission on unix systems).
why tar?
because it is a convenient way for people who receive a file with many
parts to unpack it automatically into its component parts properly named
(without messy cutting and pasting).
why compress?
because filesizes are typically reduced by at least a factor of two
why uuencode?
because compressed files are binary and will be corrupted when sent via
e-mail. (uuencode translates to ascii specifically for e-mail transmission.
note that raw postscript as well can be corrupted by e-mail transmission
due to occasional long lines or entry into bitmap mode --- uuencoding
[after compression, since .ps tends to compress well] eliminates any problems)
what's new:
the new version of uufiles solves two long-standing nuisances:
older versions of uuencode that use ' ' instead of '`' (i.e. space instead
of backquote, as still found on most
suns) now have this automatically corrected, and versions of uuencode
(e.g. on sysV hp) that assign mode 0 to stdin have this corrected to 644.
uufiles no longer tars single files, instead just compresses.
uufiles has two new options:
uufiles -gz
uses gzip compression, and
uufiles -gz -9
uses the slowest compression method (i.e. the optimal compression).
evidently to use this, you must have the gnu gzip/gunzip utilities installed
on your system (.gz compression frequently does better than a factor of two
better than .Z compression on postscript files -- for example, the
56 .ps files associated with hep-th/9409195 [a total of 1.7Mb] compressed as
figs.tar.Z = 542kb , whereas figs.tar.gz = 196kb . [of course the recipient
must have the gunzip utility installed to unpack, but this should by now be
commonplace since these utilities have been freely distributed for years.])
the new version of uufiles is 100% downwardly compatible in usage, so there is
no reason not to download and install even if you don't yet have gzip/gunzip.
use get uufiles via e-mail interface (also available in papers/macros
via anon ftp, or via www under macros).
more frequently asked questions:
why am i instructed to submit first the .tex file as a `put', then submit
the uufiled figures afterwards via the `fig command', rather than simply
submit the .tex (or .dvi) together with any .ps figs (or other attachments)
all uufiled together in the first place?
this is recommended as a compromise for the electronically disadvantaged.
those who for whatever reason cannot unpack the uuencoded compressed tar
files (despite the existence of these utilities on most platforms
[vax/vms, mac, pc included]) at least get to see the tex file.
of course any author is free to do as he/she pleases
(happily complaints go directly to submitter, not to here),
but better solutions can be implemented as soon as modernized network clients
are uniformly in use (allowing these systems to discard artifacts of their
e-mail based origins).
but why can't i already just POST a mime-formatted multipart transmission
to the server via a form page from my www client?
sadly, because most current www clients are either too primitive
or are broken. the next version of mosaic may at least properly support a
mailto: form which will facilitate pre-formatting submissions.
other clients are on the way, but it will be (at least) many months
before the situation stabilizes and suitable standards are in place.
for information on "hypertex", point your www client at
http://xxx.lanl.gov/hypertex/ . Some background:
It is easier to add hypertext capability to TeX than to simulate the TeX
typesetting environment within www browsers. (And for scientific text,
total author control over formatting and fonts is frequently preferred.)
A simple standard is to insert \special commands that add the
necessary structure to the .dvi file. For example,
\def\href#1#2{\special{html:<a href="#1">}{#2}\special{html:</a>}}
allows one to surround text with the necessary specials, so that
Here is a link to \href{http://xxx.lanl.gov/hypertex/}{hypertex}
will insert the desired anchored text with a hotlink to an arbitrary
network URL (including e.g. photos, movies, sound files, mathematica files,
etc.) The object is not only to provide hyperlink functionality to external
(network) resources, but also to provide a means of transmitting much of
the important structural and contextual information (sections, subsections,
equations, references, ...) already contained in the underlying .tex for
suitable gui postprocessing, whether directly by .dvi previewer or after
higher level translation (e.g. to .pdf [Adobe's modification of .ps that
supports a hyperlink overlay --- for which viewers will soon be generally
available for unix/pc/mac]).
The above hypertex URL contains info on dvi previewers that already implement
this functionality (for nextstep and X), contains pointers to modified
macropackages (latex and harvmac/lanlmac) that *automatically*
insert the \specials for internal eqns, refs, etc., and also has links
to some sample hyper .dvi files compiled from pre-existing source
(including random submissions to hep-th et al.) that provide edifying demos.
(There's also a haphazard collection of messages from '91-'94; plus a slightly less haphazard collection of later messages.)
www-admin@xxx.lanl.gov