Brinild

Cloud?

2011-03-22T11:13:00.000-07:00

Cloud: Dumb it down!
And I mean that as a good thing :-)

All good things must come to an end...

2004-12-17T14:41:00.000-08:00

Should have known better than to get too excited about WS-Addressing. Should have known that such a simple spec would never remain that way. Remember when the "S" in SOAP stood for "Simple"? Those days are long gone.

Recent changes made, under the guise of removing ambiguity, have in fact headed us down the path of turning a very nice and terse document into another "design by committee" mess. Instead of simply making things clearer, by removing the ambiguity, they added uncertainty and even more ambiguity to things. Take the wsa:ReplyTo and wsa:FaultTo changes made. Why not just be more explicit about what it means to be present or absent? Too simple I guess. Instead they removed its meaning all together - leaving us in the position of not knowing for sure where the heck our messages will go (if anywhere at all).

The original spec was only a couple of dozen pages - something most people could easily manage. Now its split into three - or is it four since we now have an "adjunct" document? Sigh

Sometimes I wonder if going to a standards body is really the best thing for a spec. While I don't want them all to be written by just MSFT and IBM, I have to admit that watching one of the better specs take a turn for the worse like this does make me wonder if it really would be that bad. Could be the lesser of two evils. See what an ugly mood it put me in! I need some comfort food. :-)

Changes should clear things up...

2004-12-14T13:51:00.000-08:00

Current WSA spec says:

[fault endpoint] : endpoint reference (0..1): An endpoint reference that identifies the intended receiver for faults related to this message. When formulating a fault message as defined in 3.2 Formulating a Reply Message, the sender MUST use the contents of the [fault endpoint], when present, of the message being replied to to formulate the fault message. If this property is present, the [message id] property is REQUIRED.

So, what does no wsa:FaultTo mean? No fault should be returned? Doesn't seem to clear things up.

Par for the course...

2004-11-10T13:55:00.000-08:00

Micrsoft lies? Never! Very interesting story. Anyone who has dealt with these folks can easily believe this.

Thatcher vs Blair

2004-11-09T08:18:00.000-08:00

[1] I miss Maggie.

Table for one!

2004-11-08T09:16:00.000-08:00

Martin Gudgin wrote: [1]
> It is a rare spec that makes everyone happy. I'd love to produce
> such a spec in this WG. But I think a more realistic view is that
> the spec we f inally ship will NOT be such a beast. In terms of
> broad adoption, I have some idea of who I need to make
> happy in order for the spec to be widely implemented.

LOL Must be nice to be know that how your company goes everyone
else will follow (for be forced to follow). :-)

WS-Transfer

2004-10-15T07:20:00.000-07:00

So in WS-Transfer MSFT is passing a SOAP message with no Body, the entire purpose of the message is encoded in the wsa:Action header. Now we have two places in the message (the Body and Action) which contain the true purpose of the message. What ever happened to SOAP Headers being extensibility elements???? Its one thing for the wsa:Action header to contain duplicate info (of the Body) but to replace it is an entirely different thing.