> >what do users of an HTTP binding need to get at w/o > >cracking the XML. If there is such info, then we should > >have HTTP headers. > > Good question. There are at least a few cases where having some > information in the HTTP header fields could be of help - if for nothing > else then to shut off messages unless they are signed or otherwise > verified: There is definitely value in having some amount of information at the HTTP layer, even if it merely duplicates some info already present in the XML-encoded header or payload. This means that existing load balancers, proxies and firewalls have a chance of being configured to do something potentially useful. I predict that in the long run, those network devices will have to understand and deal with XML at a number of layers -- just as switches had to learn to deal with HTTP in the first place -- but having the info duplicated allows for some limited amount of intelligent behavior by the existing network infrastructure. \\ Eugene Kuznetsov \\ eugene@datapower.com \\ DataPower Technology, Inc.Received on Wednesday, 9 May 2001 18:19:26 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:01 GMT