W3C | Architecture | Addressing
Notes on implementations and how they're described and specified -- i.e. shared context in the developer community. Also bugs.
Hmm... it would be nice if URI access libraries supported A DNS RR for specifying the location of services (DNS SRV), wouldn't it?
scheme specific processing: removing :80 from http:, lower-casing domain names in news:...@foo.bar.
processing that should be scheme-specific but isn't: looking for hostnames when doing "subtraction". lowercasing domain names.
URI scheme was originally referred to as the "access scheme" -- failure to realize the myth of names and addresses. Now called simply the URI scheme (see comment in HTParse.c.
Hmmm... anchors survive relative combination? e.g. x:foo#bar + blot = x:blot#bar?
Surround the hyperlink with angle brackets, because it contains spaces Using Microsoft Access 97 to share static data to the Web (Aug 17, Technical Article)
Wild! Try this one: an IRC channel.
NS fails to recognize it as an absolute URI!
See also: the bug involving fragment ID's and HTTP redirection. (seems to be fixed in NS 3)
Cites only NCSA docs! Not IETF RFCs at all!
terminology is not abstract at all: protocol, host, port, file, file.
monikers in the com spec
relativepathto, ala HTRelative