iNODE

Just a short note to help me remember the next time this comes up—I decided it was worth a note once I entered the second hour of trying to figure out the problem…

If you want to enable INNODB support with MySLQ 4.0.x, you must install an extra package: mysql-max. Here’s a note off the mysql website that explains both the problem I was seeing and the solution.

I’ve found quite a few people having issues initializing MySQL 4.0.x with InnoDB support under Suse 9.x Pro. Server startup will fail with a message like:unknown variable ‘innodb_data_home_dir=/var/lib/mysql’

If this happens, make sure you’ve installed the mySQL-Max package (after installing the base mySQL server). Then invoke the server like:

$ /usr/sbin/mysqld-max

You should then see your InnoDB tables initialize.

Yes, that was the problem. After grabbing the mySQL-Max package and installing it, suddenly the innodb configuration statements in my.cnf started working properly. My thanks to Brian Krispinsky who posted this last April.

http://metaatem.net/words/

Basically, you enter a word and images of each letter are pulled from Flickr…click on any one letter in your word (or short phrase) and that letter will be refreshed with a new image. Source code is included on the site for those wishing to extend things a bit.

It’s not new but it is certainly an interesting way to present digital versions of manuscripts and worth a look if you haven’t yet visited the site. The British Library offers “turnable” pages (somewhat like what you get with the Zinio reader) for historic manuscripts—requires Shockwave but it is slick. Fifteen great books are offered: Sketches by Leonardo, Mozart’s musical diary (shown above), the original Alice in Wonderland, and more. The magnifier is a really nice addition to the interface.

http://www.bl.uk/onlinegallery/ttp/ttpbooks.html

A year or so ago when I switched away from Windows, about the only thing I continued to miss was the ability to run Access on my local PC and via an ODBC connection to a couple of Solaris servers, manage MySQL databases. Yes, it built up my MySQL command line skills, ssh’ing into the server to make little corrections and whatnot, but it wasn’t as easy or as productive as the old way.

Finally, today, I learned about a product that I suppose most other MySQL users already know well—but I just discovered that they make a Mac OS X version too: Navicat

Navicat began life as a shareware product (MySQL Studio) but migrated over time to a commercial package. It offers a number of features you might well expect (create tables, edit data in tables, etc.) and many that may surprise you: import/export data; server administration, data migration and more. It’s an impressive package and if it fills your need as well as it has mine, it is easily worth the $99 ($75 for education users).

There’s also a Postgres version of Navicat (of interest to DSpace installations, perhaps).

I spent some time last night exploring the new Pages 2.0 (part of the new iWork ‘06 suite) and I’m really amazed by one feature—you can do your initial pasteup/layout using Pages (which is quite easily done) and then export the resulting document as a Word file. When you call the document up under Word, you can further edit the text and the tables and charts and columns are still there and functioning in Word. It is so-o-o much easier to do these layout tasks in Pages than in Word.

Keynote 3.0 has at least one nice feature I didn’t see in earlier versions—the ability to export your presentation as a set of web pages (PNG or JPEG) along with navigational controls. Very nice. Here’s a link to an HTML export done from a LOCKSS presentation I prepared with Keynote 2.0 some time back.

http://furbo.gmu.edu/LSO/lockss.html

Endeca’s ProFind is the “engine” behind things like World Book, the Barnes & Noble “Bookfinder” service, and a number of other systems. Today I visited a library OPAC (North Carolina State University) that’s using Endeca’s search software (a first?) and it really does raise the bar on how an OPAC ought to function and what it ought to look like:

http://www.lib.ncsu.edu/catalog/

Beyond the really nice (clean) look of this OPAC, what I immediately notice is that you can actually browse around for things—hey, that’s different. Clearly the current generation of OPACs was designed for “known item searching” and browsing was just an afterthought. This Endeca-powered OPAC shows a great deal of context around a browse and with a few clicks you can make much better progress through the collection.

Of course, the power of this “browse” feature also reveals some problems with LC classification. For example, I thought I’d browse for computer books so I clicked on “Technology” but that didn’t work. The “T – Technology” meant, the T classification, so lot’s of engineering books but nothing on computers. Knowing a bit about LC classification, I backed up and selected “Q – Science”. Finally got there but only because I knew I was really looking for QA 76 (and knew to click on the QA1-QA939 Mathematics link which then revealed the QA71-QA90 “Instruments and Machines” section). Bad Metadata In—Bad Browsing Out.

Those problems aside, an example of this sort of power in an OPAC should help move the genre forward…

I’ve noticed that ever since I moved this site to WordPress 2.x, I get comment spam (from online casinos) but only on two posts—two entries that mention WordPress in the title.  I’m sure there’s a reason for this (perhaps the spam bots just crawl post titles or something) but I’ll be more careful about how I title these entries.  For example, I didn’t title this one “WordPress Comment Spam.”