[
Lists Home |
Date Index |
Thread Index
]
Rex Brooks wrote:
> CAP, for another example, which has not been tested
> as thoroughly as WSRP, and is not a web service standard
> although it can be implemented through a web service,
> had public tests in September at the Global Homeland
> Security Conference and again March 11, at a Congressional
> demonstration, the ComCARE-sponsored Intreroperability
> Now! workshop, holding the demonstration in the Rayburn
> House Office Building that evening.
Rex, if I were you, I wouldn't suggest that CAP [1] is an
example of any thing good in the OASIS standards process. If anything,
it should be an embarrassment to them. As you know, there are at least
some (myself among the noisiest) who would say that CAP is one of the
most poorly and ambiguously defined "standards" to be declared by any
group in recent history. This standard includes:
* An invalid XML Schema
* Factual errors concerning the facilities it provides (i.e.
it says it provides facilities for encryption and signatures, but does
not)
* Contains elements like <language> which are redundant with
XML's xml:lang
* Does not clearly identify all of its normative references
* References out-of-date or obsolete normative documents
* Departs from normal XML practice in many ways
* Contains many ambiguously defined elements (like the
name/value stuff, etc.)
* Contains elements whose meanings can only be determined by
processes external the spec itself. (like geocode, etc.)
* Allows a range of date formats which is so broad that it is
effectively not a definition at all. i.e. any of the gazillion formats
provided by ISO8601 are "supported" by CAP.
I could go on...
Also, at least one of the three "Certifications" of
"successful use" provided for CAP prior to voting was actually a third
party report! i.e. The USGS certification makes reference to the CA
EDIS group successfully using CAP but says nothing about USGS using it
nor if USGS was a principle in the EDIS work.
I contend now, as I have before, that I do not believe that
any substantive interoperabilty could result from two independent
development groups creating implementations without a great deal of
inter-group discussion between the two.
CAP is an exceptionally poorly written standard that could
only have been passed by an organization like OASIS with very soft
acceptance criteria. CAP wouldn't have stood a chance of acceptance by
the IETF, ISO, W3C or most other standards groups. It is saddening and
embarrassing that a standard which is as important as this one might
be (it involves life-and-death situations) should be getting such
shoddy treatment.
bob wyman
[1]
http://www.oasis-open.org/committees/download.php/5666/emergency-CAP-1
.0.pdf
|
