Re: Important for POC Registry Implementors: Bug FixesinRegistryServices v0.8

[Sam Hunting <shunting@ecomxml.com>]

----- Original Message -----
From: "Farrukh Najmi" <Farrukh.Najmi@east.sun.com>
To: "Sam Hunting" <shunting@ecomxml.com>
Cc: "David RR Webber" <Gnosis_@compuserve.com>; "Farrukh Najmi"
<Farrukh.Najmi@east.sun.com>; "ebXML poc" <ebxml-poc@lists.ebxml.org>;
"Burdett, David" <david.burdett@commerceone.com>; "ebxml repository"
<ebxml-regrep@lists.ebxml.org>
Sent: Friday, October 06, 2000 8:23 AM
Subject: Re: Important for POC Registry Implementors: Bug Fixes
inRegistryServices v0.8


> OK. I get the hint. We will retain ebXMLError. I assume the
> Attribute.type=>Attribute.attributeType change is acceptable unless I hear
> otherwise.

Well, since you bring it up...

Maybe I've missed a thread, since I am so new, but in general I distrust
dot-style naming in tags unless such naming supports a formalism that is
documented.

If the dots are there because they feel "comfortable" to a procedural
programmer, then that worries me, because the "notation" suggests a
formalism that is in fact not present, and may not be present consistently
across the ebXML family of DTDs.  In that case, why not camelcase such tag
names to "AttributeAttributeType"?  Another alternative is to use dashes or
underscores for "word separators" since they do not have the formalist
connotations that dot-style notation does.

Alternatively, if a formalism is present, there may be other ways to express
it...

I realize this is a controversial matter; I am not, repeat not, trying to
start a flame war, just to share some lessons from my own practice.
Thanks...

Sam Hunting
EComXML
XML Evangelist