On October 1, 2001 04:51 am, Andi Gutmans wrote: > At 12:46 PM 10/1/2001 +0200, Stig Sæther Bakken wrote: [snip] > >I still think Zeev's suggestion (HTML::Table) is very good, if it > >doesn't impose too much runtime overhead. > > I don't like Zeev's suggestion because it does impose an extra hash > lookup (usually it'll be two) and probably some more logic too. I > don't think that new features we add today should suffer from > run-time overhead some constructs from the past have had to suffer. > By the way, it also has some ambiguities. Is foo::bar() a function in > class foo or in namespace foo. No one said you can't have a namespace > foo when you have a class foo. > I think the best way to go is to go with ':' and live with the > ambiguity. Many languages have such ambiguities and it's probably not > such a big deal. In light of :: also having ambiguities why not go > with :? > > Andi
This may be a ~very~ daft question, but what effect would having classes behave as read-only namespaces have? Would that alleviate any problems? Just my one cent (not willing to put a whole two cents on this one. ;) -- Zak Greant PHP Quality Assurance Team http://qa.php.net/ "We must be the change we wish to see." - M. K. Ghandi -- PHP Development Mailing List <http://www.php.net/> To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] To contact the list administrators, e-mail: [EMAIL PROTECTED]
