Re: Pycoon?

On Nov 15, 2003, at 7:17 AM, Bruce D'Arcus wrote:

>
> On Nov 14, 2003, at 7:14 PM, Kimbro Staken wrote:
>
>> I'm not trying to build a general XML management system because  
>> there's a lot of junk that is considered part of XML, in particular  
>> DTDs are completely incompatible with what I'm doing.
>
> Yes, I see this.
>
>> Actually any schema language that constrains an XML document instance  
>> is incompatible. I rely heavily on being able to arbitrarily mix XML  
>> fragments together however you want.
>
> I'm having trouble understanding this part.  Can you give an example  
> of where you would not want to have a schema (I'm a Relax NG fan  
> myself)?  I suppose in my case the data I am interested in is regular  
> enough that I'd want to constrain it for consistency's sake (though it  
> does mean writing my own schema).

Constraint annoys me. :-) Basically since I'm targeting this as a  
personal data management tool, if I want to put x where x doesn't  
belong I don't want the system telling me no. Over constraint to me is  
actually a fundamental problem of computing and leads to tremendous  
usability problems with computers. It's a large part of what makes  
humans have to bend to the ways of the computer rather then the  
computer accommodating the way we work.

Basically constraint is like getting a 2x4 and having it slap you  
across the face with a stern no when you try to cut it at a 47 degree  
angle when the designers only intended it to be cut at 45 degrees. This  
is why we suck so bad at software reuse, we just can't figure out how  
to build software that allows you chop of a corner, round off the edges  
or cut it at a 47 degree angle unless we design that in.

Anyway, that's my philosophical argument, the technical one is that I  
slice and dice XML, pull bits from here, bits from there and jam it  
into any old place within a document. There is simply no way a system  
can do that if there's a schema standing in the way.

Note: to understand at all what I'm talking about you have to  
understand that I'm only looking at Syncato. I'm not interested in  
global interchange of data. I'm also not necessarily talking about data  
that comes into the system from untrusted sources. There are very  
legitimate uses of constraint, it's just over used in my opinion.

There is also nothing that prevents using a constraining schema  
language around the edges. For instance you can certainly validate the  
document before you put it in Syncato. Of course it will probably never  
validate again because syncato will stick an id attribute in the root  
node and a pubDate element as a child of the root. So if you really  
want  a system that will enforce the schema Syncato won't be it.

>
> Bruce
>
>
>
> -------------------------------------------------------
> This SF. Net email is sponsored by: GoToMyPC
> GoToMyPC is the fast, easy and secure way to access your computer from
> any Web browser or wireless device. Click here to Try it Free!
> https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/ 
> g22lp.tmpl
> _______________________________________________
> Syncato-general mailing list
> Syncato-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/syncato-general
>
>
Kimbro Staken
Software, Consulting and Writing http://www.xmldatabases.org/
Apache Xindice native XML database http://xml.apache.org/xindice
XML:DB Initiative http://www.xmldb.org



-------------------------------------------------------
This SF. Net email is sponsored by: GoToMyPC
GoToMyPC is the fast, easy and secure way to access your computer from
any Web browser or wireless device. Click here to Try it Free!
https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/g22lp.tmpl