Am Freitag, 28.11.03 um 02:34 Uhr schrieb Tim Pritlove:
Ben,Note that I am using NetNewsWire and it works just fine with the fink feed for me.
"read it" I should. I did. I read the RSS 1.0 standard, not the output of a validator that doesn't get the point and doesn't test the necessary RDF structure completely. And I read the output of my RSS reader and it is confused by fink's feed.
I don't really understand why nobody understands that the RSS fink is deeply broken. I have pointed out the errors. The output of feedvalidator is no help.And I don't really understand why you don't understand that RSS 1.0 is not considered a standard by me :-). Just because somebody calls it a standard doesn't make it one. Our RSS feed is a RSS 0.9x feed, which is supported by many many news aggregators. RSS 1.0, despite its name, deviates in many substantial ways from 0.9x. RSS 2.0 is the true continuation of 0.9x. There are technical merits to both the 1.0 variant and the 2.0 variant. Problem is, if you comply to the one, you can't comply to the other, and vice versa. Personally, I am getting more and more fed up by the RSS community, which is apparently mostly about arguing which of them is the "true" RSS. Hopefully a true standard will emerge one day. At this point, however, there is none.
The most obvious bug and the one that made me do the bug report was that the <item> elements lack a unique resource URII already fixed that a couple days ago.... How about first checking? :-)
and that the <channel> element misses it's basic element: the <items> RDF sequence of resource URIs. In addition, all referenced URLs get forwarded to another URL.RSS 1.0 is the only of the multiple existing RSS specs which requires this. The others don't.
I already stated that I agree this should be fixed. Which is why I *already fixed it* some time ago.
The <lastBuildDate> content problem is not important and mainly useless anyway. Keep it or kick it: nobody will care. RFC 822 dates are neither user nor parser-friendly anyway (which the use of a unknown time zone shows), ISO-8859-1.
What RDF users _do_ care about is unique URIs. Maybe this is considered "pointless". I don't think it is.
Max
------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Fink-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/fink-devel
