Re: [xml-dev] Transformative Programming: Flow-based, functional, and more

From
Simon St.Laurent <>
To
"" <>
Date
2013-10-18T15:31:14Z
ID
<>
Thread
Re: [xml-dev] Transformative Programming: Flow-based, functional, and more
On 10/17/13 6:35 PM, Peter Hunsberger wrote:
> I'll buy into your view on REST, maybe a better fit is something like
> Web Sockets?

It could be.  I think my general thought is that it matters less how the 
modules communicate among themselves and more how they're assembled or 
orchestrated.  REST will certainly work for the communication, but I'm 
not sure it's central.

> However, I can't agree with you in regards to tighter
> specs inside applications and looser between them.  Are you suggesting
> that (for example) a Java method that has a single String as it's input
> and output needs a full XSD or some other more formal spec?

No, but - to use an example I have doubts about - programmers build 
types inside their code and expect pieces to fit together based on those 
declarations.  They don't need XSD, as there's already a type system there.

XSD is what you use when you want to inflict type systems on the outside 
world.  (Granted, 'outside' may have a variety of meanings.)

> I suspect
> you're indulging in a bit of hyperbole in that,

Actually, I was indulging in understatement.

> flexibility between
> applications in what they can accept might simplify the world, but you
> know as well as anyone that not everything can be mapped to name value
> pairs or tuples and left at that?

I think we've been through this conversation, but for a quick reprise:

* Brittleness works better inside of applications where you control the 
whole conversation.

* Flexibility, whether by means of multiple pathways, human 
intervention, or a combination of strategies, works better between 
applications where you don't control it all.

The tragedy of XML is that instead of building flexible systems for 
dealing with each others' expectations, we slapped schemas down as 
relatively simple tests, border guards who aren't very smart.  Tangling 
it further, we then try to talk through schemas when we should probably 
just talk with each other.

Yes, it does take effort to build systems without demanding prior and 
consistent agreement, and it can be hard to see the value when people 
come from legal or programming cultures that mistake brittleness for 
strength.

I'm happy to report that I keep hearing small stories of flexible 
structures built on markup, and hope that new best practices will climb 
out from under the current maze.

Thanks,
-- 
Simon St.Laurent
http://simonstl.com/