Monday, April 22, 2002

There's a big ruckus on the OpenACS boards about MSSQL Server support. There are a few people who have found themselves in a situation where the client they are pitching is completely reliant on a MSSQL Server architecture. This doesn't bode well for suggesting OpenACS as a solution. I would like to make it a more realistic option for these folks, and so a while back I started porting the kernel data model (including pl/pgsql functions) to MSSQL Server. I didn't get to where I could run the SQL-level regression testson the data model, but I figure that's no more than a few weeks away. My fingers are crossed that someone would like to see this happen, and (more importantly) could locate some financing. I think a big wildcard is the aD release. What they're doing could potentially be an easy port to the dotNet platform. I have no doubt that MS went all out of the Java-2-C# conversion wizards, and that might be the closest path to getting MS shops interested in (open)ACS. Database support and licensing, remain the big TBD's. 6:24:03 PM      comment []  trackback []

Why was I extolling my e-smith gateway box? Well, I just set up Radio's mail-to-blog feature. This feature lets you submit a blog entry via email - the Radio client just scans a POP account once a minute. This is nifty because: It's another way for me to write my blog entries with Emacs It is way easier to make an asynchronous architecture like this fault tolerant. Userland could have slapped a front end onto their Radio servers (actually they do, it's called Manila), but instead they gave you a gateway from e-mail, which you'll pretty much have everywhere. Okay, to connect this back to e-smith, it has POP accounts and so it was really easy for me to set it up to catch my blog postings on my gateway box, which are read and forwarded by my Radio instance whenever it connects to the network. Slick. 6:15:16 PM      comment []  trackback []