OpenImpetus

Old News

10 Sep 1999
V0.9.6.4R is done and appears stable. We'll use this for the NEU comp, and we'll see if YCN can get it running. 28 Apr 1999
V0.9.5.7R is done! This version is a major internals rewrite, as it involves coding up a persistent server process that handles database requests. Accordingly, performance speedup is remarkable. I've also cleaned up the database internals so that pointers to structs, rather than structs themselves, get passed around.

V0.9.6.* will address the list of bugs and RFEs that were made after the MIT comp.

11 Mar 1999
V0.9.* is now in production for the MIT comp; as updates are made, I will update the version number.

2 Mar 1999
V0.6.0 is now in testing for the MIT comp. I am hoping to shoot forward to v0.9.0 when it (hopefully) goes live on 5 Mar 1999. When the Comp concludes, I'll hopefully have v1.0 ready for general release.

11 Jan 1999
Well, so much for getting v1.0 out by Jan 1. V0.4.0 is out and I've asked the MIT Comp Comm to help test it; we'll see how much assistance I get. I'm hoping that I can quickly move forward to v0.9 in terms of testing and confidence in the underlying engine; then I can concentrate on the rest of the bells-and-whistles and finally release v1.0.

At some point, I'll need to open this up to more general-audience testing, but I'm unsure how much response I'll get.

24 Nov 1998
v0.3.4 released. This release is the heart of the database functionality; everything that's left is just bells-and-whistles -- granted, stuff that makes the Registrar's life much easier, but it shouldn't affect the DB internals, which is where the real bugs can hide. I'm hoping to test this to stability between now and 11 Dec. If that occurs (namely, testers find the time to test and I find the time to develop) and no serious bugs are uncovered, then I hope that openimpetus can be rapidly promoted through its pre-release versions.

Probably the hardest part of what remains will be the site configuration perl script, as I haven't even attempted to do that. Number generation is a well-understood algorithm. Sending email will be annoying, but hopefully not too bad. The biggest problem will be interfacing with sendmail; the last time I tried this, it didn't work. If anyone has any tips on this, I'm all for hearing them. :)

I am hoping to release v1.0.0 by 1 Jan 1999.

11 Nov 1998
Woo-woo! V0.3.0 is out the door. Now to start collecting bugs. Hopefully the database is relatively stable. TBA still needs to be implemented.

29 Oct 1998
I'm almost done with v0.3; I need to test removing people from the database. V0.2.7 could almost be considered v0.3 except for this last detail. I think I've also finally figured out how to properly configure an apache webserver, which means that real testing can begin (i.e., testing done by others).

23 Oct 1998
Basic functionality is coming along: start registration, show registration, and testing change registration. Need to start the very horrible process of change registration; that's going to be a field day. Also, stats reporting isn't touched yet, but hopefully won't be too bad -- it'll be annoying, but hopefully not a show-stopper.

22 Oct 1998
Progress is smoothly coming along. Right now, it's still very much in development -- I'm hoping to complete process_change_registration() today, which means I can start logic testing soon. Maybe by the end of this week, I'll be able to release v0.3.1.

September 24, 1998
Well, upwards and onwards... now undergoing development. This is pretty disappointing, but oh well. :)


dave leung (dcltdw@mit.edu)