Yes, I am familiar with Marceau's work (at least the published papers), it is very cool. One of very few systematic research-based approaches to error messages. I haven't used WeScheme, might be something to look into (although our target audience is college students and other adult beginners, not middle school).
Elena On Tue, Apr 12, 2016 at 11:13 AM, Matthias Felleisen <matth...@ccs.neu.edu> wrote: > > Yes, all of our releases are always announced. > > If you are studying error messages, please check out Guillaume Marceau’s > work, supervised by Kathi Fisler and co-advised by Shriram Krishnamurthi. > He worked out an even better version of his ideas for WeScheme, which is > still the starting point for Bootstrap, our educational spin-off project > for middle schools. > > — Matthias > > > > > > On Apr 12, 2016, at 11:58 AM, Elena Machkasova <ele...@morris.umn.edu> > wrote: > > Sounds good, thanks! Will the new release be announced? We are looking at > error messages as a part of a research project, so it would be good to use > the new one if we can. > > Elena > > On Tue, Apr 12, 2016 at 10:53 AM, Matthias Felleisen <matth...@ccs.neu.edu > > wrote: > >> >> Thanks, I have pushed a fix and additional tests for error messages. This >> will be included with the next release, to appear in a couple of weeks. — >> Matthias >> >> >> >> >> > On Apr 11, 2016, at 9:25 PM, Elena Machkasova <ele...@morris.umn.edu> >> wrote: >> > >> > Hi All, >> > >> > we have discovered that in 6.4 DrRacket (Advanced Student) seems to >> have a bug generating error messages for foldl/foldr: >> > >> > (foldl + (list 2 3 4) 0) >> > >> > gives an error message "foldl: 2th argument must be a list, given 0" >> > (in addition to the incorrect "th" for "2nd", the 0 is actually the >> third argument, not second). >> > >> > There is also a somewhat misleading error (for beginners) in 6.4 and >> earlier versions, shown with a somewhat artificial example: >> > >> > (map (lambda (x y) (+ x y)) (list 2 3 4)) >> > >> > gives and error: >> > >> > the given procedure's expected number of arguments does not match >> > the given number of lists >> > given procedure: #<procedure> >> > expected: 2 >> > given: 1 >> > >> > This is technically correct, but not understandable to a student since >> most likely they mistakenly used a two-argument function instead of a >> one-argument one. In fact, students are never shown a map with multiple >> lists. A better message would say that a one-argument function was >> expected, but a two-argument function was given. >> > >> > The second error is somewhat of a design choice, but the first one >> (with foldl) is a bug. If there is a way to patch it in 6.4, that would be >> greatly appreciated. >> > >> > Cheers, >> > Elena >> > >> > >> > -- >> > You received this message because you are subscribed to the Google >> Groups "Racket Users" group. >> > To unsubscribe from this group and stop receiving emails from it, send >> an email to racket-users+unsubscr...@googlegroups.com. >> > For more options, visit https://groups.google.com/d/optout. >> >> > > > -- > Dr. Elena Machkasova > Associate Professor of Computer Science > Division of Science and Mathematics > University of Minnesota, Morris > Office: Sci 2325 > (320) 589-6308 > http://cda.morris.umn.edu/~elenam/ > > > -- Dr. Elena Machkasova Associate Professor of Computer Science Division of Science and Mathematics University of Minnesota, Morris Office: Sci 2325 (320) 589-6308 http://cda.morris.umn.edu/~elenam/ -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
