Comments on: Mschedule ready for Fall 2006 http://blog.kylemulka.com/2006/03/mschedule-ready-for-fall-2006/ Thu, 02 Mar 2017 22:34:44 +0000 hourly 1 http://wordpress.org/?v=4.2.2 By: Kyle Mulka’s Weblog » MISchedule Ready for Winter 2006 http://blog.kylemulka.com/2006/03/mschedule-ready-for-fall-2006/comment-page-1/#comment-1359 Thu, 30 Mar 2006 05:05:48 +0000 http://blog.kylemulka.com/?p=357#comment-1359 […] o schedule generator have been fused together into one site. http://www.mschedule.com/ More info: http://blog.kylemulka.com/?p=357 MISchedule, the University of Michigan automatic schedule generator, is ready for Winter 2006 scheduling. I just l […]

]]> By: Kyle Mulka http://blog.kylemulka.com/2006/03/mschedule-ready-for-fall-2006/comment-page-1/#comment-1357 Thu, 30 Mar 2006 02:07:31 +0000 http://blog.kylemulka.com/?p=357#comment-1357 Hey Mike,

Please don’t attack the Mschedule.com server. I do have a cap on my bandwidth. If you are going to do that, either let me mirror the database on a different server and you can just do SQL select statements to the mysql database. Or… I can just give you a dump of the database and you can run whatever you want on your own server.

Most of the PHP source code released as open source I’ve thrown out because of the redesign of Wolverine Access and the lack of documentation for it. (I didn’t write it in the first place)

I’ll email you about meeting.

-Kyle

]]> By: Mike http://blog.kylemulka.com/2006/03/mschedule-ready-for-fall-2006/comment-page-1/#comment-1356 Thu, 30 Mar 2006 01:57:28 +0000 http://blog.kylemulka.com/?p=357#comment-1356 Hey Kyle,
I was interested in doing some modeling on demand for these classes and wondered if it was OK to do a full tree traversal around 4AM nightly for the next week or two. Just finished reverse-engineering the request API and response format, but I don’t want to contribute to the load if you’re on a capped bandwidth plan.

As a sidenote, the server I’m running is currently on the UM backbone (i.e. ~4ms latency for static responses and scaling logarithmically as concurrency doubles) and is decent enough to support load-balancing or failover for you, so if you’re interested, let me know. Also, been awhile since I’ve peeked at the source, but if memory serves, there was quite a bit of optimization that could be done on the caching mechanisms and gzip-compression (I noticed your hosting provider wasn’t running mod_deflate either).

And lastly, sorry about the miserable response time on your last email, but I’m still interested in meeting with you about that project. My lab of employment is in the process of moving in the coming month, but can you make a 1-ish lunch up here on North sometime in the coming week?

=m

]]>