Perl 6 - the future is here, just unevenly distributed

IRC log for #webwork, 2012-07-13

| Channels | #webwork index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
02:14 CIA-105 joined #webwork
15:12 mgage__ joined #webwork
20:35 djun joined #webwork
20:36 djun hello
21:28 djun joined #webwork
22:24 mgage__ joined #webwork
22:26 mgage__ hi djun
22:27 djun hi mike
22:27 djun how's it going?
22:27 djun are you still on the West Coast?
22:28 mgage__ ok -- yeah -- I'm in los altos hills for the next couple of days
22:28 mgage__ then back to berkeley and we'll be headed home on Wednesday
22:29 mgage__ looks like things are going ok with the library -- and peter and david have been working on the classlist editor.  so looks like good progress.
22:29 djun hope you're managing to enjoy it :)
22:29 djun i've got some time to look at things over the weekend
22:30 mgage__ yes -- it's been good to catch up with old friends and to see family
22:31 djun nice, so holiday time and not business, I guess?
22:32 mgage__ a little of both.  I went around to Stanford this morning and also to AIM --- but mostly it's holiday time
22:32 djun I've skimmed your email re: releases, will go over more carefully tomorrow
22:33 djun The last couple of weekends I've been tied up trying to rent out our house. Playing landlord seems to take an excessive amount of energy sometimes
22:34 mgage__ ok -- I'd like you comments and perhaps we can discuss things more toward the end of next week or next weekend.
22:34 mgage__ Yeah -- landlord is a job like anything else unless you get really lucky
22:35 djun Sounds good. It's always good to have a plan, and IMHO also good to publicize these things.
22:35 mgage__ I'd like to set up a more structured procedure for releases than we use at the moment.  It will make it easier for more people to participate.
22:35 djun I guess we've been lucky as far as tenants go - just the changeover is exhausting :)
22:35 djun yes, I agree
22:36 djun Perhaps something like Trac might be useful - have actual milestones with tickets against them
22:36 mgage__ ok -- I'll plan to catch up with you about that next week.  Also about whether to have a separate -dev repo or not -- I still see advantages each way.
22:36 djun I notice that the domain openwebwork.org redirects to the MAA site.  Are there any plans for this?
22:37 mgage__ I own it -- I can redirect it where ever we decide to.  For a while it was directed to the source forge site -- the maa address seems the best for now.
22:37 djun sounds great
22:39 mgage__ the other thing is how often to release things.  I've always liked frequent releases -- particularly since we don't have really good testing mechanisms/ and people -- that way the bugs are found while there are still be people around who remember how to fix them.  Others are much more conservative - there is probably some reasonable compromise.
22:39 djun I'm booking tickets to Winona today.  I guess the nearest major airport is Rochester, MN? Are people renting cars, or there some convenient way to make that last leg of the journey?
22:41 djun What you mentioned at the Codecamp re: users installing over the summer is a pretty big driver for an annual cycle of major releases.  I think having lots of releases leading up to such a thing is a good idea - scheduled preview, alpha, beta, RC releases, etc.
22:41 mgage__ Car rental is probably best -- you might look at LaCrosse, WI also -- it's slightly closer .  The one time I flew in I went to Rochester and rented a car, but I would at least consider LaCross next time. Because of ticket prices when I bought them it was better for me to fly in and out of Madison and rent a car to go from Mathfest to winona.
22:42 djun I'll explore the options :)
22:42 djun I gave a workshop on WeBWorK authoring at one of our local colleges this morning.  Lots of interest.  They are doing an online placement exam.
22:43 mgage__ that's ok with me -- particularly if we can convince a reasonable number of people to actually try them in situations realistic enough to find bugs.  --- to often everyone waits for the bugs to go away --- but if no-ones testing the bugs don't go away -- and mean time the programmers have lost interest or disappeared entirely :-)
22:43 mgage__ very nice.
22:44 djun Also ran into a strange bug with includePGProblem().  It doesn't like BEGIN_SOLUTION.  Seems like that should have an easy fix...
22:44 mgage__ -- haven't noticed that -- can you describe it?
22:46 djun Basically, when I includePGProblem a file which has the "BEGIN_SOLUTION" keyword, the processor bails out with an "Error in file" message.
22:47 mgage__ ok -- I'll take a look at it.
22:48 djun no worries - i can have a closer look, file a bug report and probably a patch over the weekend
22:49 djun I looked briefly at the source, and it looks like there's processing of similar cases
22:50 djun just thought I'd mention it in case you'd run into this earlier
22:50 mgage__ don't see what it should go wrong on just that command -- just looking at the source in PG.pl
22:55 djun looks like in includePGtext, strings like "BEGIN_TEXT" are substituted with the heredoc version
22:55 djun line 83 of IO.pm
22:56 mgage__ that is done everywhere    BEGIN_TEXT,  BEGIN_SOLUTION  are just syntactic sugar for the here doc version.
22:56 djun looks like includePGtext doesn't know about BEGIN_SOLUTION.  I just added it
22:58 mgage__ in which file?
22:59 djun pg/lib/WeBWorK/PG/IO.pm
22:59 djun are there other instances?
22:59 djun it's possible that I'm looking at a quite old version of this..
23:00 mgage__ the substitution actually takes place in PG/Translator.pm
23:00 mgage__ let me look at IO.pm
23:01 djun sorry, I need to leave in a minute.  Department BBQ :)
23:02 djun Plus I don't want to keep you from friends and family esp on a Friday on vacation :)
23:02 mgage__ you might be -- Arnie reported a problem with solutions in the webwork2 repo -- I suggested that he use the webwork2-dev repo instead since I think I have that in reasonably stable state, but he decided to go back to the svn version instead.
23:03 mgage__ np.  The includePGtext macro was somewhat specialized --- meant for CAPA file compatibility only -- there is some refactoring that could take place here there are at least 3 versions of include files and some of them can be consolidated
23:03 mgage__ ttyl
23:04 djun thanks!  Talk soon
23:55 djun joined #webwork

| Channels | #webwork index | Today | | Search | Google Search | Plain-Text | summary