[Rets-dev] Contemplations on RETS 2.0

[Matt Lavallee]

Hi Michael,

I think I can outline a "remedy" that won't kill anyone:

1. Prohibit the modification or redefinition of (complex) types defined in the
standard payload -- this does not affect the simple types of string, integer,
boolean, etc., that are defined by XML Schema.

2. Prohibit the modification or redefinition of Standard payloads.

3. Require that the payloads launched with the standard be delivered without
modification in order to be deemed compliant.

While I understand the argument of a "gentle transition", the problem is that
you can't put the poop back in the horse.  Tightening restrictions after 75% of
the systems are up and running would be pointless (look at all the RETS 1.5
systems that haven't gone to 1.7).  An overly-strict release specification that
is loosened gradually to resolve specific issues is the only way to keep the
Standard contained *and* backward-compatible.

-Matt

On Wednesday, August 01, 2007 at 10:24 PM, Michael Wurzer wrote:
> 
> This conversation has occurred before and I learned then that discussing
> the concepts of data standardization in the abstract is unlikely to be
> fruitful.  Both sides to the discussion have good points, in the abstract
> (basically, the positions boil down to those focused and usually remain
> unpersuaded of the others' position.
> 
> However, the payloads workgroup has shown that discussing the concrete
> detail can result in broad agreement.  The big question, though, is how we
> extend the good work done so far on member, team, etc., on to the breadth
> and depth of the listing data, which we haven't delved into too far yet.  I
> think what's there is quite good, but we need broader participation and
> input to get the kind of agreement that will allow for stricter
> requirements for compliance.
> 
> As we discussed at the last payloads meeting, we need to define a process
> for how the broad variety of constituents can provide input and then we
> need someone to harmonize the input.  As this work is being done, we'll
> have much more opportunity to discuss the strictness of the standard within
> the context of specific fields.
> 
> All that being said, I also don't think it is productive to delve back into
> the data types discussion right now.  We'll have a big enough hurdle
> discussing the fields without also requiring strict data typing.
> 
> To summarize: stick to specifics, establish a process, and set aside data
> strict typing for the moment.
> 
> Michael
> --------------------------
> Michael Wurzer
> President and CEO
> FBS Data Systems
> mwurzer at fbsdata.com
> www.flexmls.com/blog
> 800.437.4232 (office)
> 701.306.9341 (mobile)
>