ejournals, authentication, etc.

[John W. Felts, Jr.]

Hi all,

I created a Microsoft Access database that is accessible 
through a WWW interface using Active Server Pages 
scripting.  This gives users the ability to browse 
or search for full-text electronic journals real-time, via 
the WWW (point your browser to 
http://library.uncg.edu/ejournals/), and allows me to 
maintain one record for each ejournal title, rather than 
having to update one very long or several HTML pages.  Of 
course, this solution works only for those running Windows 
NT as the NOS, and IIS 4.0 as their WWW server software.  

I presented on this recently at LITA, and put my Powerpoint 
screens online if you'd like to see an outline of the 
process, and how I created this interface 
(http://library.uncg.edu/depts/eir/jwf.asp). I'm 
also further elaborating on ejournals at NASIG in June, if 
anyone happens to be there (sorry for the self- 
endorsement:).

Stacy raises some valid questions, especially pertaining to 
our old ejournals friend, Lexis-Nexis.  Unfortunately, I've 
found no way to automate incorporating their new source 
list into our database.  Not only are there existing titles 
w/ different full-text inclusion dates and a slew of new 
titles, but as Stacy points out, many of the titles aren't 
full-text.  I'm painfully (*very* painfully) going through 
each and every title and determining full-text 
availability, ranges of coverage, etc.  Jimmy Gaphery at 
VCU and I have also been trying to figure out a way to 
script a URL for title-level access to L-N ejournals.  This 
actually works in principle, but the various L-N indexes 
lack authority, which means you get false drops when 
scripting to the title.  It would be a colossal chore 
to tweak each and every URL and test for accuracy, but it 
can be done.  Also, the source lists, even though 
considerably better than two months ago, leave much to be 
desired.  If anyone would like a copy of what I come up 
with for these L-N titles, I'm happy to share -- just let 
me know (jwfelts at uncg.edu).

For those ejournal titles that originate from aggregate 
databases such as ProQuest, SearchBank, etc., I send users 
to a local help page with information on how to locate a 
specific title in a specific database.  Since sessions are 
dynamically created at the vendor's server, this is the 
best I can do.  There are obvious security issues that 
arise, too, if a vendor started allowing users to create 
scripts to be run on their servers.  Sounds like fun, 
though :)  I provide title-level access to any titles that 
have a dedicated URL, such as those in Muse, JSTOR, 
Academic Press, etc.  We use a couple of Perl scripts that walk 
users through the authentication process. 
 
We are currently using MS_Proxy for remote access to these 
resources, but hope to convert to EZ-Proxy soon since its 
easier for our patrons to use. We're currently using 
EZ-Proxy for our database authentication, but, at least as 
we see it, each URL has to be in a master WWW file, and 
given that there are hundreds of URLs for these ejournal 
titles, it'll take us a little time to get to this.  I'm 
also not sure how such a long list might impact EZ-Proxy's 
functionality.  If you have any questions, I'm happy to share ideas.


On Sat, 25 Dec 1999 09:03:58 -0800 (PST) 
web4lib at webjunction.org wrote:

> 			    WEB4LIB Digest 1673
> 
> Topics covered in this issue include:
> 
>   1) Re: journals online
> 	by spober at manhattan.edu
>   2) Re: ejournals, authentication, and the OPAC
> 	by spober at manhattan.edu
>   3) Elegant solution to login requirements?
> 	by spober at manhattan.edu
> 
> ----------------------------------------------------------------------
> 
> Topic No. 1
> 
> Date: Fri, 24 Dec 1999 22:19:05 -0800
> From: spober at manhattan.edu
> To: web4lib at webjunction.org
> Subject: Re: journals online
> Message-ID: <386461D9.157C at manhattan.edu>
> MIME-Version: 1.0
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
> 
> Earlier this year I did a consolidated browsable list of links to the 
> journals we had access to through the various vendors (UMI, Lexis-Nexis, 
> etc.) at the time: 
> http://web1.manhattan.edu/library/journals/jinfo.html
> 
> That was done by putting the journal lists from the vendors into
> spreadsheet program, doing various sorts, then adding HTML to link to
> the specific vendor sites.  It was a little tedious to do but not
> particularly challenging from a technical standpoint.  
> 
> However, some of the links posted here were searchable journal databases
> and very impressive.  Could some of the creators of the searchable lists
> tell us a little about just how they created those?  
> 
> Also, a related question for those of you who create and maintain these
> journal databases.  What are you doing about the journal lists from
> Lexis-Nexis Academic Universe?  The downloadable source database does
> not distinguish full-text titles from those for which only abstracts are
> availabl (or at least it didn't do this the last time I downloaded it.) 
> Their own searchable source list is also somewhat buggy, in that the
> links to "Search This Title" sometimes go to the wrong database segment. 
> 
> I went to some of the library sites that had searchable lists from the
> URL's posted on Web4lib and I notice that for the individual journal
> records, the links usually go either to the main gateway page into the
> vendor's site or else they link to the vendor's journal list.  I know
> that for selected titles, a few libraries have chosen to write URL's
> that link directly to a search for the individual publication, but AFAIK
> those links are usually done by hand-coding.  Still, that would be ideal
> for most users - to click from the library's list of journals available
> in full-text directly to the search page that specifies a search in that
> specific publication.  Has anyone managed to automate the creation of
> those links for any of their vendors?  
> 
> Best,
> Stacy Pober
> Information Alchemist
> Manhattan College Libraries
> spober at manhattan.edu
> raingoddess at juno.com
> 

----------------------
John W. Felts, Jr.
Networked Information Services Librarian
University of North Carolina Greensboro
jwfelts at uncg.edu
336-334-4238