[bop-devel] Python gateway module

[Ivan Kohler]

On Wed, Aug 23, 2006 at 04:43:12PM -0700, Jo Rhett wrote:
> On Aug 23, 2006, at 4:29 PM, Ivan Kohler wrote:
> >No, I'm talking about machine-parsable gateway descriptions that
> >programs in various languages can read, not English API documentation
> >humans can read.
> >
> >They're not equivalent unless you have a solution to the
> >natural-language problem hidden up your sleeve.  :)
> 
> I've seen dozens of attempts of this kind of centralization, and the  
> structure of the unifying language always fails to compensate for the  
> natural-language problems created by natural-human programmers who  
> are not always consistent in their usage, and even less so when  
> forced to be by inflexible and random business requirements.
>
> And jesus christos, you're talking about dealing with banks here.   
> Logic isn't even an acknowledged part of the equation.  Don't argue  
> that the input and output are fairly predictable, you'll be blown  
> away by how random the requirements are that these people have to  
> code against.
> 
> Given that BOP is roughly halfway done

Personally, I thought it was reasonably complete when I took it over 
four or five years ago, and I've always felt it needed maintenance, some 
refactoring/cleanup and refining rather than any sort of major 
development.  What did you have in mind?

> and lacking in unification for  
> much of the same reasons, I can't imagine why taking on a meta- 
> project like this is really useful.

1. I said it was pie-in-the-sky.

2. I don't personally think its anywhere *near* as intractable as the 
doom-and-gloom picture you paint, Mr. Little.  Half the supported 
gateways could supported with server/port/path, a straightforward set of 
field remappings, and a few other switches.  A week or two of busywork.

3. Nobody's asking *you* to pitch in.  Don't rain on my parade, dude.  :)
   Also see: #1

-- 
_ivan