+1 on this. For everything they've gotten wrong, I think in this regard Microsoft got naming of interfaces correct. There's no ambiguity about the evolution of a given interface.
-- Kevin On Tue, Feb 10, 2009 at 10:50 AM, Howard Lewis Ship <[email protected]> wrote: > The problem with "ExtendedAsset" is that in Tapestry 5.2 I might have > to add "SuperExtendedAsset" or something. I'd rather see Asset3! > > The number on the end is ugly, but reflects the evolution of the > framework. There's also a ModuleDef2 and a ServiceDef2 now. > > This is the best approach to backwards compatibility. > > For service interfaces that are not expected to be extended by the > user, just injected, I'm more willing to break "perfect" backwards > compatibility and just introduce a new method. It's a judgment call. > > > On Tue, Feb 10, 2009 at 1:14 AM, Massimo Lusetti <[email protected]> wrote: >> On Tue, Feb 10, 2009 at 9:54 AM, Andreas Andreou <[email protected]> wrote: >> >>> I feel there should be a better name for this. I could propose one >>> ExtendedAsset or VariableAsset but i dont know if more methods will >>> end up there... Is Asset2 meant to be directly used by users? >> >> It's not 'internal' so i guess yes. >> >> -- >> Massimo >> http://meridio.blogspot.com >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: [email protected] >> For additional commands, e-mail: [email protected] >> >> > > > > -- > Howard M. Lewis Ship > > Creator Apache Tapestry and Apache HiveMind > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
