RE: > in attribute values; decimal character reference

I think those are great work arounds that any good language designer worth their salt would come up with.

However, and again this is my very parochial view of things, anyone who uses xml or uxml to represent data is a language designer, whether they know it or not andnd whether they are any good at it or not. Ordinary users of XML do this sort of thing all the time.

I guess all I am doing is being a poster child for the "Oh uxml, that's' really neat... but I want all the old stuff too!" syndrome. 

Of course uxml can't have "all the old stuff too." but whenever it takes one of the old things away it should weigh in the how much it will take away from its acceptance.

Enough for philosophy for now, I've just got to get back to my regularly scheduled programming... :-)  I'm sure you all will work things out.


-----Original Message-----
From: Andrew Welch [mailto:andrew.j.welch@gmail.com] 
Sent: Wednesday, September 05, 2012 9:31 AM
To: Daniel Sullivan
Cc: David Carlisle; public-microxml@w3.org
Subject: Re: > in attribute values; decimal character reference

On 5 September 2012 14:23, Daniel Sullivan <dsullivan@danal.com> wrote:
> I guess I'm taking a parochial end users view and not looking at how complicated the spec will be.
>
> Having to do something like
>
> <box test="length&gt;1"/>
>
> When what I mean is
>
> <box test="length>1"/>
>
> doesn't make sense to me.

what about

<box test="length gt 1"/>

or

<box test="length greater-than 1"/>

or

<box test="is-greater-than(length, 1)"/>

etc

there's no restriction on the language designer to use < and >, it's not a reason (imo) to add complexity to microxml...


--
Andrew Welch
http://andrewjwelch.com

Received on Wednesday, 5 September 2012 13:49:40 UTC