eprints_tech messages

From: Margery Tibbetts <margery.tibbetts AT ucop.edu>
Date: Fri, 13 Jul 2001 10:14:44 -0700

Chris,

         Do you have plans to support version 1.1 of the OAI Metadata 
Harvesting protocol in the next release of EPrints?

Thanks,
Margery


Margery Tibbetts
California Digital Library
University of California
1111 Franklin Street
Oakland, CA 94607-5200
(510)-987-0581
fax - (510)-763-2471

From: "Faine, Mark" <Mark.R.Faine AT msfc.nasa.gov>
Date: Mon, 23 Jul 2001 07:21:04 -0500

Ok, I have a nice fresh clean Solaris 8 install with all the tools and apps
I'll need installed and the box is ready to go.  I plan on using it for a
test of the eprints software for our NASA tech reports. I haven't yet
installed Apache/mod_perl yet nor a few other things.  Does anyone have any
ideas or suggestions, perhaps a checklist to add based on their own
experience?  I'd like to benefit from the mistakes you have made if I could.
 
Any suggestions/advice would be appreciated.
 
-Mark 

--

ATTACHMENT: eprints-tech.50.2.html!
--

From: support AT ecs.soton.ac.uk
Date: Mon, 23 Jul 2001 13:27:14 +0100

Regarding your message:
[EP-tech] starting new server
(full message included at end of email)

Your message has been submitted into the problem & suggestion
tracking system. 

No action is required on your part.

If you are interested, your message has been assigned the 
following URL:
https://www.ecs.soton.ac.uk/cgi-bin/jbug/systems?findid=1309

Please contact webmaster AT ecs.soton.ac.uk if you need more 
information about this system.

> Subject: [EP-tech] starting new server
> From: "Faine, Mark" <Mark.R.Faine AT msfc.nasa.gov>
> Date: Mon, 23 Jul 2001 07:21:04 -0500
> 
> This message is in MIME format. Since your mail reader does not understand
> this format, some or all of this message may not be legible.
> 
> ***ATTACHMENT: sub1
> ***ATTACHMENT: sub2
> 
> 
> 

From: Ed Sponsler <eds AT library.caltech.edu>
Date: Fri, 27 Jul 2001 14:55:16 -0700

Hello, 

I'd like to make two suggestions:

-- Include file size next to the document download link on the main document
(service / front matter) page.

-- Allow the citation style used on the result list / browse pages to differ
from the style which appears at the top of the main document (service /
front matter) page because they serve two different purposes. Only minimal
information is desirable for browse/search result lists (author, title,
maybe year). But the main document page, we think, should contain a 'real',
model citation style suitable for cut/paste by users into other documents
(in which the work is cited). We would, for example, want to include
institution, department and a persistent URL on the main document page
(citation). This is because the document download page is accessed outside
of the context of EPrints by Google and OAI service providers. However, the
extra information we'd put on the document page citation is unnecessarily
redundant in the citation style used within an Eprints browse / search
result list. 

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Ed Sponsler, Computing Analyst
Caltech Library System

From: ePrints Support <support AT ecs.soton.ac.uk>
Date: Wed, 1 Aug 2001 16:44:18 +0100

As announced last month at the London OAI meeting:

August 1st 2001:
EPrints 2 Alpha-1 has been released and is available from:
http://www.eprints.org/files/eprints/

Features Include
 - International language support (UTF-8 and multiple "phrase" files)
 - Much improved database back end - real text indexing and normalised data.
 - Improved installation (no longer uses ./configure & make)
 - Full support for multiple archives on one server
 - Object Oriented Design and database abstraction

Caveats
 - This is not a stable (or even fully working) release.
 - It is not an upgrade for EPrints 1 series.
 - Many bits remain to be done.

If all goes to plan, from tonight, nightly builds from CVS will be available ↵
too.

See http://www.eprints.org/ for more information on the eprints project.

--
"The EPrints Team"
Christopher Gutteridge & Mike Jewell
support AT eprints.org
+44 23 8059 4833

From: Bob Kemp <robert.kemp AT strath.ac.uk>
Date: Thu, 09 Aug 2001 11:31:37 +0100

Hi All,

I've been building a new subjects config file for our archive here at 
Strathclyde, and it isn't rendering correctly. I suspect this is because 
I've used numeric subject tags, although the documentation doesn't actually 
exclude this. Anyone know if this might be the case?

Thanks,

Bob

Dr Bob Kemp
Information Officer, Digital Information Office
Centre for Digital Library Research
University of Strathclyde, Glasgow G4 0NS

Phone: +44 (0)141 548 2379              FAX: +44 (0)141 548 2102

From: Bob Kemp <robert.kemp AT strath.ac.uk>
Date: Mon, 13 Aug 2001 15:37:25 +0100

Hello again,

Some of you may think this should have been obvious! My problem was nothing 
to do with numbers, and everything to do with the spaces I'd inserted 
between sections in my new subjects config file. They made for much clearer 
reading, but basically broke it.

Cheers,

Bob

At 11:31 AM 8/9/01 +0100, Bob Kemp wrote:
>Hi All,
>
>I've been building a new subjects config file for our archive here at 
>Strathclyde, and it isn't rendering correctly. I suspect this is because 
>I've used numeric subject tags, although the documentation doesn't 
>actually exclude this. Anyone know if this might be the case?
>
>Thanks,
>
>Bob
>
>Dr Bob Kemp
>Information Officer, Digital Information Office
>Centre for Digital Library Research
>University of Strathclyde, Glasgow G4 0NS
>
>Phone: +44 (0)141 548 2379              FAX: +44 (0)141 548 2102
>

From: Bob Kemp <robert.kemp AT strath.ac.uk>
Date: Mon, 13 Aug 2001 15:38:38 +0100

Hi All,

My experience in setting up the eprints software is that getting things 
like the subject tree right involves a certain amount of iteration. This 
would be much easier if it was possible to change the metadata fields and 
the subject hierarchy without deleting the database and recreating it every 
time.

Would this make a useful version 2 improvement?

Bob

Dr Bob Kemp
Information Officer, Digital Information Office
Centre for Digital Library Research
University of Strathclyde, Glasgow G4 0NS

Phone: +44 (0)141 548 2379              FAX: +44 (0)141 548 2102

From: billn <billn AT yelmtel.com>
Date: Mon, 13 Aug 2001 10:32:24 -0700

> useful version 2 improvement..

I think it would be *essential*. The first implementation might not be
able to save/move any attached information, but the ability to make
changes over time as the subject list grows is important.

Bill Nicholls
Wrriter & Consultant

Bob Kemp wrote:
> 
> Hi All,
> 
> My experience in setting up the eprints software is that getting things
> like the subject tree right involves a certain amount of iteration. This
> would be much easier if it was possible to change the metadata fields and
> the subject hierarchy without deleting the database and recreating it ↵
every
> time.
> 
> Would this make a useful version 2 improvement?
> 
> Bob
> 
> Dr Bob Kemp
> Information Officer, Digital Information Office
> Centre for Digital Library Research
> University of Strathclyde, Glasgow G4 0NS
> 
> Phone: +44 (0)141 548 2379              FAX: +44 (0)141 548 2102

From: Christopher Gutteridge <cjg AT ecs.soton.ac.uk>
Date: Tue, 14 Aug 2001 00:39:34 +0100

On Mon, Aug 13, 2001 at 03:38:38PM +0100, Bob Kemp wrote:
> My experience in setting up the eprints software is that getting things 
> like the subject tree right involves a certain amount of iteration. This 
> would be much easier if it was possible to change the metadata fields and 
> the subject hierarchy without deleting the database and recreating it ↵
every 
> time.
> Would this make a useful version 2 improvement?

I'm working on the subject tree right now!

I'm not planning a way to change the metadata fields just yet - that's
HARD. Maybe I could provide instructions for the brave on how to do it
with SQL'ing. I'm afraid the SQL in version 2 is much more complex, though.

I plan to make the Subject Tree alot more editable. You will be able to
add, delete and modify subject nodes. Although I may disallow, or at least
discourage, deleting nodes which have items in the archive having that subject.
comments?

As part of my internationalisation (i18n) effort, the subjects will also be
able to have alternative names for each language you wish to support. Don't
worry, if you only care about english then this won't make life harder.

I have also redesigned the subject tree to be more of a subject lattice, that
is a node may have more than one parent. Eg. Biochemistry may have both
Biology and Chemistry as parents and the same subject will appear in both
sub trees, although it will be illegal (and very messy) to have recursion.

I have also removed subjects as a system field. It is now an ↵
"archive" field
which you can use or not. Also a "subject" field need not have ↵
"ROOT" as it's
top level. For my own department I want 3 subject style fields ↵
"projects",
"themes" (research areas) and "Research Group".  The ↵
"research group" will
be required, the others optional.

These 3 areas will appear as seperate fields and be seperately searchable 
but will all be the same overall heirarchy for the purposes of the ↵
"view" 
static pages and the OAI sets.

I'm also working on the concept of "subject editors" users with less 
permission than admin or full editors, who can only approve submissions
attached to certain subject nodes. In my departments case there will be one
or two per research group, although there could be an editor for large 
projects too.

If an item is editable by two editors then either can approve or reject it. It
has been suggested that the system should require some kind of consensus(sp?)
between the editors but I am considering this a human issue, out of the scope
of automated legislation. (for now).

I still plan to have an import (and export!) method for the subject tree so 
that you can share it (or bulk modify it). I'll probably make a bulk importer
too so you can dump it, edit it, and load it back. This will also be handy if
you want to set up with the same subject tree as another eprints archive.


-- 
Christopher Gutteridge -- cjg AT ecs.soton.ac.uk -- +44 (0)23 8059 4833
----------------------------------------------------------------------
Take care of the cojones and the frijoles will take care of themselves. Try 
to have getaway money -- but don't be frantic about it.
  -- From "The Notebooks of Lazarus Long" by Robert Heinlein

From: "Christopher D. Green" <christo AT yorku.ca>
Date: Mon, 13 Aug 2001 20:21:06 -0400

My version of e-prints (see http://htpprints.yorku.ca/) seems to have a
little bug in the search engine. If I search an author's last name, I
get his or her documnets. If I search both the first and last names (in
whatever order, or with a comma between, or with an "and" between ↵
them,
etc.), however, I get nothing. I also get nothing if I search the first
name alone. Has anyone else had this problem? Is there a way to fix it?

Best,
--
Christopher D. Green
Department of Psychology
York University
Toronto, Ontario M3J 1P3

e-mail: christo AT yorku.ca
phone:  416-736-5115 ext. 66164
fax:    416-736-5814
http://www.yorku.ca/christo/

From: Bob Kemp <robert.kemp AT strath.ac.uk>
Date: Tue, 14 Aug 2001 14:16:14 +0100

We have the same problem with our archive here at Strathclyde too, so you 
could be right to suggest a bug. It must be fixable though - you can search 
with a comma between names OK in CogPrints (although not if you try their 
second example - "Harnad,Steven"....).

Bob

At 08:21 PM 8/13/01 -0400, Christopher D. Green wrote:
>My version of e-prints (see http://htpprints.yorku.ca/) seems to have a
>little bug in the search engine. If I search an author's last name, I
>get his or her documnets. If I search both the first and last names (in
>whatever order, or with a comma between, or with an "and" between ↵
them,
>etc.), however, I get nothing. I also get nothing if I search the first
>name alone. Has anyone else had this problem? Is there a way to fix it?
>
>Best,
>--
>Christopher D. Green
>Department of Psychology
>York University
>Toronto, Ontario M3J 1P3
>
>e-mail: christo AT yorku.ca
>phone:  416-736-5115 ext. 66164
>fax:    416-736-5814
>http://www.yorku.ca/christo/

From: ePrints Support <support AT ecs.soton.ac.uk>
Date: Wed, 15 Aug 2001 04:20:15 +0100

I'll dig out the fix on cogprints next week, unless someone needs it urgently.
I'm on hols. this week....


On Tue, Aug 14, 2001 at 02:16:14PM +0100, Bob Kemp wrote:
> We have the same problem with our archive here at Strathclyde too, so you 
> could be right to suggest a bug. It must be fixable though - you can ↵
search 
> with a comma between names OK in CogPrints (although not if you try their 
> second example - "Harnad,Steven"....).
> 
> Bob
> 
> At 08:21 PM 8/13/01 -0400, Christopher D. Green wrote:
> >My version of e-prints (see http://htpprints.yorku.ca/) seems to have ↵
a
> >little bug in the search engine. If I search an author's last name, I
> >get his or her documnets. If I search both the first and last names ↵
(in
> >whatever order, or with a comma between, or with an "and" ↵
between them,
> >etc.), however, I get nothing. I also get nothing if I search the ↵
first
> >name alone. Has anyone else had this problem? Is there a way to fix ↵
it?
> >
> >Best,
> >--
> >Christopher D. Green
> >Department of Psychology
> >York University
> >Toronto, Ontario M3J 1P3
> >
> >e-mail: christo AT yorku.ca
> >phone:  416-736-5115 ext. 66164
> >fax:    416-736-5814
> >http://www.yorku.ca/christo/

-- 

 Christopher Gutteridge                   support AT eprints.org 
 ePrints Technical Support                +44 23 8059 4833