This post is a bit of a grab bag…three small entries, none of which really merits a post but together they nearly rise to an excuse for typing.
OpenURL / ILL
Spent most of Friday working up a web-based system to link our OpenURL resolver, our ILS (Voyager) and our ILL management system (CLIO as modified by Endeavor), enabling a user to move from the OpenURL resolver’s “Sorry, we have no full-text sources for this item” to an ILL request that would flow through our ILL management system without human intervention. The stumbling block was a value that our ILL program depends upon (PatronID) to automate NCIP links and the fact that this number is never displayed to users and appears only as an internal number to Voyager. Oracle DBI and Perl to the rescue (not to mention an SQL assist from one of my colleagues). Doubt there’s much interest outside our institution for the code to accomplish this (PERL and just a bit of PHP) but I’ll be happy to share it if there’s any expression of interest.
OA / IR
Sent a comment to Richard Poynder’s blog post on Institutional Repositories the other day–disagreeing with one of his suggestions about the use of an IR to provide Open Access services. I mention it here because 1) it’s an interesting and well-thought out (for the most part) analysis of the subject and 2) he’s asking for a PayPal donation from reader’s who find it worthwhile to support his continued writings in this area. Now that is a trend that would change the way blogs operate, isn’t it?
He responded to my comment and didn’t call me a library zealot or anything so that’s nice…still think he’s drawing too broad a conclusion from a very small sample.
A couple OS X sys admin commands:
There are some UNIX utilities you tend to run only when there’s some sort of problem. On Solaris, I find the “truss” command fits that role–it allows you to watch “real time” the execution of a job on the system. Feed it a process ID and you can view what a program’s doing (or not doing). I’ve really missed having this capability on OS X. Well, truss isn’t there but I just discovered a utility that’s quite similar: ktrace. Unlike truss, ktrace doesn’t dump the raw data to stdio (the screen) but rather writes it to a file which you can then view via kdump.
Another very useful command that I’ve used more than once over the years (typically when trying to determine if there’s an intruder running something on a system) is lsof (List Open Files). It’s available on OS X as well. lsof will show you all open files…and since UNIX treats most everything as a file (sockets, devices, inodes, etc), it’s quite powerful.