Elias Tzoc blog!

Last week (March 18-20), I attended this annual meeting at Purdue University. A page with all the details of the program can be accessed here. It was again great to get to see/meet colleagues that we often communicate with -electronically- …as well as enjoy some good food!

On the 18th, the day started with the keynote address by Geri Ingram from OCLC. We got some interesting updates and tips for the migration process to version 5.0 or 5.1 -to be released in May. On that day, I also did a 1.5 presentation/s:

a) Bulk uploading CONTENTdm images & metadata into Flickr by Jason Michel and myself

b) Re-Indexing CONTENTdm Fields for Searching and Browsing

Since we spoke quite a bit about PHP; for the last session, I chose to go and hear about the use of Python to move library catalog data into CONTENTdm -good and informative one!

On the 19th, I found the “Text-based projects in CONTENTdm” presentation very interesting! -especially now, when our own text-based collection -Miami Student Newspaper- has become the most popular among all current collections.

Last but not least, it was great to know that the gathering in 2010 will be in Iowa City!

…for the upcoming MidWest CONTENTdm Users Group meeting at Purdue University, I’ll be presenting a basic web application for re-indexing metadata fields and creating new browse and search features.  You’re welcome to try any of the tools by accessing this page.  This is a good example of how a PHP script can transfrom a regular CONTENTdm custom XML file and create:

a)  a tagcloud or a navigation list for browsing
b)  an autosuggest for searching

…to view an actual digital collection with these features, visit the Miami Student Newspaper Collection.

After helping to upload two image collections into Flickr and exploring some of the browsing features, I was re-motivated to experiment with a tagcloud feature for CONTENTdm collections.  After exporting and parsing an XML file including the subject field, I had to write a PHP script to generate the tagcloud.  For the presentation, I checked the tagcrowd site -very helpful!

For something like this, the use and manipulation of “arrays” was the key.  Some of the array functions I used are: count(), arsort(), array_slice(), ksort(), max(), min().  Of course, my favorite one is array_slice (not because of the pizza slice), it’s a handy way of limiting the number or top results to display.  A regular tagcloud shows 10-25-50-100 elements, but maybe no more than that.  Well, for this experiment, my array had more than 1,000 elements, …but with the array_slice function, now I can easily limit the top results to any number and even set the starting point (e.g. array_slice($array1, 10, 50) means get 50 elements of the array1 starting at the 10th element).

All right, here is the link to the current test file, …that’s it for now!