||April 28, 2003 - Todd
SFU has been running Citaiton Manager for the last 6 months or so now with
much positive feedback. Currently there's about 500 users and 2000+
citations stored. No major problems to report, just a few minor bug fixes
and interface tweaks. Bulk importing is now working as well, with filters
for EndNote, Procite, RIS, Web of Science, Proquest, tab delimited, and XML.
Mark Leggott successfully imported over 1000 citations, though displaying
that many citations is slow. I'll be adding paging to the interface to deal
with large citation sets as soon as I get a chance.
|Test Server Updated
||June 18, 2002 - Todd
I've updated the test server to match the current development version. The big change
in this one (aside from bug fixes) is the move to a citation list format which scales
to different screen resolutions much better than the old style. All the data was maintained,
so you can log in with an old account if you set one up before.
|LITA Open Source Pre-Conference
||June 17, 2002 - Todd
I've just returned from speaking about Citation Manager at the LITA Open
Source Pre-Conference in Atlanta. Thanks very much to Jeremy for inviting
me down and Emory and OCLC for facilitating (and financing) some informative
sessions. I've added a download link since I've been able to package
everything up into something reasonably easy to install. I'll be automating
the installation system more down the road, but for now it involves a bunch of file
editting, but nothing too confusing.
There's been a number of updates to the Citation Manager interface which are
not evident from the test server at this point. If you want to check them
out, feel free to play with the development
server. The current version download will usually be the version that the
development server is running while the test server may lag behind somewhat.
|New Test Server!
||October 19, 2001 - Todd
I've updated the test server to match my development server as of this morning. It
now has the ability to edit records and uses the new Citation object internally.
It's not running with persistant database connections right now, so you'll probably
see short delays before each page is displayed. If I get around to it, I'll try
turning on persistant connections to test that it's working properly in a "production"
||October 16, 2001 - Todd
I added a Feature List page today which I will
try to keep updated with features that I plan to add to version 1 (and future
versions). Let me know if there's something you want to see added to the list.
|Site/User Selectable Fields?
||October 15, 2001 - Todd
I've made the switch to using objects to represent the citation on my development
server. Field editting is close to working, but I've found lots of places in the
code where it's complicated by me trying to add support for sites (and users?) being
able to select which database fields they can use. Unfortunately, it means that
I have to do a bunch of extra checks against the site config table, as well as
maintaining a seperate list of the default site fields. It would be much easier
to strip this out and just go with all sites/users having the same fields. I think
this is a feature which I thought would be nice to add, but wont actually end
up being used and due to the complications in supporting it, I'd rather strip it
out for now. In the future, sites will still be able to customize what labels
they have on all the fields, however I'll be taking away the ability to say that
(for example) SFU students never see the "MLOG" field. Any objections? email me!
||October 10, 2001 - Todd
I now have my development server running under mod_perl with persistent database
connections. This speeds up server response substantially as the Perl scripts
driving it do not have to be recompiled for each request, and the somewhat
"expensive" connect/disconnects to PostgreSQL are minimized.
I've been stripping out redundant code and trying to have objects carry around
a little more data which saves me from passing parameters around. Some common
code was turned into functions, error reporting was cleaned up, and a new footer
with more run-time information was added. Hopefully tomorrow I'll have enough
back-end changes done to implement field editing...
|Missed Export Filters
||October 9, 2001 - Todd
Sorry everyone, I missed putting the XML and EndNote export filters into the
CVS tree so they weren't working in the demo server -- this is now fixed.
I'm going to do a pass through the existing code and clean up a bunch of
stuff I put in there just to get the demo working for Access. Might be a
couple days before any visible changes make it in, but there's lots going on
in the background.
|Field Editing, Object Changes
||October 4, 2001 - Todd
I've almost got field editing working properly, but I'm taking the time to
objectify the citation while I'm at it. There are lots of places I've found
in the code which could be simplified if the citation knew a little more
about itself, plus I can help protect the programmer (that would be me) from
field errors by switching to accessor functions. I'm also taking the time
to do some documentation, which is badly needed.
||October 2, 2001 - Todd
You're looking at the new project home page for Citation Manager. Unfortunately,
content is almost non-existent -- I'm working on it! There's a link to our CVS
repository which contains all the Citation Manager code as well as all existing GODOT
code (both the old and new "versions"). There's also a section containing Perl database libraries which I have written and
use heavily throughout the new GODOT code and Citation Manager itself. Hopefully I'll
get to documenting the whole thing soon.|
I've split off the development tree and will maintain a separate test/demo
server which people can use to test features, find bugs, etc. It is
separate from any live production servers, and is also separate from my
development server. Please go ahead and use it as much as you want. If you're running
GODOT you can use it to get data into Citation Manager, otherwise you'll have to use
the blank form until I get generic "Import Filters" finished. User keys are arbitrary,
you can use whatever you want and it will create an account on the fly. The default
is '1234567890', but use whatever you like if you'd like to keep your data separate for
Thanks to everyone who attended Access and gave me many useful suggestions which I'll
be working to implement as soon as possible. Your kind comments were much appreciated despite
my lack of presentation skills. Thanks again to Mark Leggott and the University of Winnipeg
for hosting another great Access conference! Mark seems to understand the best way to get
free software out of us is to provide free beer...