[OAI-implementers] Beta public interface up for playing

Alan Kent ajk@mds.rmit.edu.au
Thu, 4 Jul 2002 17:07:47 +1000

I am not sure how much use this will be, but I was surprised how useful
people found my old web page of errors I had trying to harvest people's

We are building an application on top of the system we sell here.
I will admit up front it is in part advertising ('buy our product').
So there is ("buy") some subliminal ("BUY OUR PRODUCT") advertising
on the first ("buy") page. I am sure it will be extrememly effective :-)
But the OAI application itself is not actually a product at this stage.

So I am really advertising (err - announcing?) this as a replacement
to the old page of harvesting results that I used to maintain.
This is still a work in progress, but there is nothing like
putting something in a real life situation to find the problems!

If you want a look try


I have defined one collection so far, and grabbed all the old OAI
sites I used to crawl (probably out of date now) and loaded them up.
I have only tried to crawl a small number of them so far though! (I will
try others incrementally). We don't have the automatic harvesting
going yet so I have to manually trigger a harvest for a site.
(If you want me to try your site in particular I am happy to do
so, after which you can look at the log messages yourself online.)
I am sure there will need to be lots of tweaking to make the log
contents more useful.

Any comments on presentation or what sort of messages to go into
the log would be greatly appreciated.

It will also allow harvesting of what I collect via OAI. The baseURL
will be http://inquirion.com:8123/oai/OAI. But its currently broken
(ListRecords does not even work for some reason). It was working
before, so its a transitory problem.

There is also a URL based query interface:


Parameters in summary are:
    ccl - ISO 8777 CCL query (%3F = '?' which means match everything)
    	title=science and creator=ed?
    recnum - starting record to get (from 1)
    maxrecs - how many to get (default 10)
    rs - Z39.50 record syntax (sutrs, grs1, xml)
    ss - stylesheet (summary and detailed are supported at the moment)

There is also a static HTML page at:

This page is basically a fill in form that jumps off to the above
query based URL.

There is also a Z39.50 server behind the scenes, but I am guess most
people on this list won't care that much about it.

I am going to be reinstalling, reharvesting, etc multiple times as bugs
are found. So don't plan to OAI harvest from this unless you are just
testing your code.

Note also there is no data that we create. I am just stealing everyone
else's data. This really is intended as one means of testing your
site - that is a public interface to the logs we are collecting when
we try to harvest your site, just like the old web page I used to

I am actually going on leave tomorrow, so any problems I may not
respond to immediately. Maybe I should not be releasing it yet,
but its fun anyway! :-)


ps: If you want me to add a new site or try your site, I will endevour
to do so. We plan to start automatically harvesting sites periodically
some time soon.