On Tue, 2009-02-17 at 03:55 -0800, alan wrote: > Following up on this discussion: > http://groups.google.com/group/django-users/browse_thread/thread/83f7b9b93973671f/e4b40ec5a482d155?hl=fr&lnk=gst&q=jsi18n+cache#e4b40ec5a482d155 > > How about a solution where the js i18n file would contain all the > translations for all the languages (in a dictionary indexed by > language)?
That could be a potentially huge file. You'll only be using 2% of it, even with the current number of translations. Lots of unnecessary information sent over the wire there. > This way, the file could be cached by the browser (because non > language cookie dependent). > Of course it would require more bandwidth to transfert but: > - it would be transfered only once per client You're sending literally 50x more data per page than you would at the moment. We already send a bit more information than we have to, since we send all translations, rather than just those needed for a particular page (that's kind of hard to solve). It's survivable because web pages are reasonably regular (and repetitive), but a bit fragile. It also has to be checked for updates on the same frequency as the page data itself (and will change anytime the page changes). You'll get some reuse, but I'd want to see some more concrete numbers on potentially real-world-style data to be convinced about this. Hand-waving doesn't work here. You have to make some real measurements and carefully control for all the variables. Regards, Malcolm --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django users" group. To post to this group, send email to django-users@googlegroups.com To unsubscribe from this group, send email to django-users+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/django-users?hl=en -~----------~----~----~----~------~----~------~--~---
