Hi Shriram Nice to hear from you!
In a course for 14-18 year olds where we are rigorously enforcing the design recipe, the overwhelming majority of errors are, indeed, syntax errors. (Except for missiing-else conds, of course) So that was what I meant. Pyret syntax errors are reported confusingly. Infix just messes things up. And in my view prefix does not present a problem for students Conversion: Indeed we did find an old version of Reactive in Racket, by which time we had already translated the newer Pyret version into Turkish and also introduced some experience based changes of our own. So what we had in our hands had diverged in several ways from the old Racket version. So we were merging the changes we had made and also translating the Turkish langauge documentation from Pyret to Racket. I understand the critcisms for using lang racket rather than advanced student in the second week , after using beginner in the first. Given that advanced student also accepts else-less conds (the worst source of hard-to-find errors) and we wanted to use the new Racket struct syntax (and struct-copy which can make HtDP universe code much shorter) we were pushed in that direction. I dont think we paid any other penalty for using lang racket. If we had had struct and struct-copy in advanced student, we would have used it. I am sure that is something we could have solved ourselves ... but ....time... Thanks for your views. Chris On Wednesday, August 21, 2019 at 2:38:55 PM UTC+3, sk wrote: > > >> Pyret was a pain. Error messages were not clear and the whole change >>> confused students. >>> >> >> Can you elaborate more on this? My personal experience with Pyret is that >> it has an exceptionally good error message, except when an internal >> error occurs (which should not happen). >> > > IMO, Pyret's *parse* errors are terrible. To some extent they are an > unavoidable consequence of infix, but it doesn't matter: parse errors just > suck, even for me, even today, even as one of the three longest-running > Pyret programmers. > > The run-time errors are (again IMO) much better than Racket's. > > Chris, FYI, Bootstrap:Reactive *was* in Racket to begin with, so there > may not have been a need to "convert" it… > > Shriram > -- 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. To view this discussion on the web visit https://groups.google.com/d/msgid/racket-users/bae23873-6837-415c-97ff-d7c0879873ad%40googlegroups.com.
