I think the main driver is to retain the illusion that everything is discoverable at runtime and that bytecode looks like source code in certain ways.
I personally am not bothered either way on this, but I know people often run up against the limitation. In fact, any generic warning or error that a programmer doesn't understand is often misattributed to erasure. Scala shows that you don't need reification to allow primitives as type parameters, however Scala also goes to lengths with its Manifests to let you get at the original types that I do wish it would present that as reification and not make you have to see any Manifest use in source. I wouldn't see reification as a disaster for Java if it did arrive, not sure whether Fabrizio is basing this on reason or feelings again. On Oct 3, 2012 6:04 PM, "Cédric Beust ♔" <[email protected]> wrote: > Fair enough, let us know if you reach any interesting conclusions! > > -- > Cédric > > > > > On Wed, Oct 3, 2012 at 1:50 PM, Martijn Verburg > <[email protected]>wrote: > >> Hi Cedric, >> >> I confess I am actually torn. I'm a big fan of polyglot and reified >> generics could throw a spanner in the works for this. We had a fun >> discussion with Charlie (JRuby) Martin (Scala), Jeff (Gosu) and Ola about >> this last year. >> >> Counter point is that I feel that type erasure generics breaks the >> principle of least surprise for the day to day developer. >> >> I always harp on about empirical evidence, so I really should set up a >> workshop and have developers try out both types and see what their feedback >> is. :) >> >> Food for thought - thanks for triggering it! >> >> Cheers, >> Martijn >> >> >> On Wednesday, 3 October 2012, Cédric Beust ♔ wrote: >> >>> On Wed, Oct 3, 2012 at 12:14 PM, Martijn Verburg < >>> [email protected]> wrote: >>> >>>> Tentatively scheduled for 9 or 10 - I'd prefer to see 9 personally but >>>> appreciate its a non trivial change ;-) >>> >>> >>> But why? I'm still struggling to find out why some people feel so >>> strongly about the importance of reified generics. If you spend some time >>> thinking about the implications and costs of reified generics, you actually >>> realize that the need is rare and that even in such situations, type >>> literals (or similar) can get you very far, and that erasure comes with >>> many more pros and less cons than reified generics do. >>> >>> I captured these thoughts in this >>> article<http://beust.com/weblog/2011/07/29/erasure-vs-reification/>a while >>> ago, I'd love to hear if your use case for reified generics is not >>> covered there. >>> >>> -- >>> Cédric >>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "Java Posse" group. >>> To post to this group, send email to [email protected]. >>> To unsubscribe from this group, send email to >>> [email protected]. >>> For more options, visit this group at >>> http://groups.google.com/group/javaposse?hl=en. >>> >> -- >> You received this message because you are subscribed to the Google Groups >> "Java Posse" group. >> To post to this group, send email to [email protected]. >> To unsubscribe from this group, send email to >> [email protected]. >> For more options, visit this group at >> http://groups.google.com/group/javaposse?hl=en. >> > > -- You received this message because you are subscribed to the Google Groups "Java Posse" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/javaposse?hl=en.
