On Sun, Nov 14, 2010 at 2:36 PM, Per Bothner <[log in to unmask]> wrote:
> On 11/14/2010 10:55 AM, Bruce D'Arcus wrote:
>> On Wed, Nov 10, 2010 at 3:05 AM, Per Bothner<[log in to unmask]> wrote:
>>> Probably most people benefiting from this spec wouldn't be typing and
>>> reading raw XML - but I don't know the target "market" that well.
>> To me, the obvious target formats are:
>> 1. HTML and XML
>> 2. RDF (in its various syntax forms; both XML and otherwise)
>> 3. JSON
>> So the details of syntax can't trample of any of that (and thus, angle
>> brackets are out).
> My question was: While HTML and XML are obvious target formats,
> how often would people be writing HTML/XML *by hand*, and how often
> would people by reading the raw HTML/XML, rather than using browser?
> If that is rare¸then we shouldn't *automatically* rule out using
> '<' or '>'. Especially in a rarely-used feature. OTOK if it's
> a rarely-used feature then it would also be OK to use a less
> "natural" syntax if it avoids issues with reserved characters!
It's very common in my world to both edit and read raw HTML and XML,
and most people who work with HTML and XML (particularly on the
standards' end) would agree.
You'd in essence have to say that what you want to be ">" is really
">" (or maybe only for some formats), make accommodations for just
that in parsers, etc., etc.
So yes, IMHO it's completely unreasonable to consider it. :-)