07 May 2006
A year or so back I experimented with TiddlyWiki but didn’t really get beyond that.
I’ve since taken it up again as an attempt to become better organized. We use Confluence at work and I’ve experiemented with having a personal space where I can record my ToDo’s and other misc. pieces of information. It’s nice but requires me to be either on the network or at least vpn’d in. Basically, I want an information store that will _always_ be available, with or without an Internet connection.
Not ideal in all circumstances.
Enter TiddlyWiki
A free MicroContent WikiWikiWeb created by JeremyRuston and a busy Community of independent developers. It’s written in HTML, CSS and JavaScript to run on any modern browser without needing any ServerSide logic. It allows anyone to create personal SelfContained hypertext documents that can be posted to a WebServer, sent by email or kept on a USB thumb drive to make a WikiOnAStick. It also makes a great GuerillaWiki. This is revision 2.0.10 of TiddlyWiki, and is published under an OpenSourceLicense.
Prior to today, my strategy of organization has consisted of Jira (our bug tracker) tasks, Firefox bookmarks, Confluence entries, del.icio.us tags along with various email reminders.
My medium-term goal is to rid myself of most firefox bookmarks (the to-read ones), del.icio.us tags, and email reminders. My del.icio.us feed has become populated with hundreds upon hundreds of links and I’m finding it difficult to track things down.
TiddlyWiki looks like it could be a solution to the problem, I think the only barrier is me finding time to actually make use of it. It’s also nice that it doesn’t involve any server-side logic and thus I can copy and backup easily.
In closing, I must say that even if I don’t continue using TiddlyWiki, it’s at least an interesting application of recent (and not-so-recent) web technologies.
07 May 2006
Next weekend I’ll be in sunny San Francisco getting ready to take in my first JavaOne. We’re staying in the Sir Francis Drake, which was recommended by one of the guys travelling with us.
There’s no way I’m actually going to be able to sit in sessions from 8:30 -> 11:30 (counting the BoF’s) but I did make an attempt to sign up for anything and everything that looks interesting. Plenty of back-end sessions (looking for the Cameron Purdy’s talks) along with some client-side stuff from Karsten and Romain Guy.
Weather looks good, and it’ll be nice to finally get to a conference where I’m not actually performing booth duty in the exhibition hall.
18 Apr 2006
I just caught an announcement that Compass 0.9 was released. Not knowing what Compass actually was, I ventured over to the OpenSympony site.
Compass is a first class open source Java Search Engine Framework, enabling the power of Search Engine semantics to your application stack decoratively. Built on top of the amazing Lucene Search Engine, Compass integrates seamlessly to popular development frameworks like Hibernate and Spring. It provides search capability to your application data model and synchronizes changes with the datasource. With Compass: write less code, find data quicker.
Sounds pretty cool, basically an integration point between Lucene and Hibernate/Spring. It’s been an area of interest for some time, I’ve wanted to build a more googlish/search engine-ey interface to one of our applications at work. We’ve got rudementary searching in place, but it’s more pre-defined than ad-hoc. We provide the entities that you can search on and the relationships are mapped behind the scenes.
It works and it might even be more usable than something relatively open-ended like a traditional engine, but hey, we’ve utilizing hibernate on the persistence tier, I might at least try and incorporate something.
So I set out tonight to see how far I could get in a couple hours. Compass’ integration with Hibernate & Spring is fairly straight-forward, basically point it at an initialized SessionFactory and you’re set. Note: If you’re using Spring you must use the SpringHibernate3GpsDevice instead of the Hibernate3GpsDevice (it’s pretty funny what compass is doing behind the scenes to ensure that its event listeners get registered on the session factory, re: FieldInvoker’s on private collections, etc.).
The sticking point that prevented me from actually getting something real going was the compass equivalent of a mapping file. Currently there’s no easy way to convert an existing hbm hierarchy to compass. A tad unfortunate and there’s been some rumblings on the user forums but it’s not there yet. I’ve got an idea to whip a little templating solution together that’ll parse a simple text file and generate the approprate compass files. It shouldn’t be that difficult to hack together (easier than parsing the hbm.xml files) and it should give me an idea of whether or not there’s value in continuing down this path.
Needless to say it looks pretty interesting and the community seems active. I’ll post again a bit later in the week as I have time to work on this.