Re: [Nbd] Back to the options parsing debate
- To: Goswin von Brederlow <goswin-v-b@...186...>
- Cc: nbd-general@lists.sourceforge.net
- Subject: Re: [Nbd] Back to the options parsing debate
- From: Alex Bligh <alex@...872...>
- Date: Sun, 31 Jul 2011 19:59:23 +0100
- Message-id: <0D967B89BFD241132E35120B@...873...>
- Reply-to: Alex Bligh <alex@...872...>
- In-reply-to: <87y5zee4lk.fsf@...860...>
- References: <3606CBE1A3E3A79634940EB3@...873...> <87y5zee4lk.fsf@...860...>
--On 31 July 2011 19:34:47 +0200 Goswin von Brederlow <goswin-v-b@...186...>
wrote:
Strike 1.
XML is basically nothing more than a tokenizer. You then still need to
define a grammar to make sense out of the data contained in the xml.
So all you did so far is add an obscuring encoding.
I know what XML is. Actually you do not need to define a full grammar,
(in XML terms an xsd / schema / whatever). It's perfectly possible
to have an extensible xml format without a schema, and that's
particularly useful if you want to add (e.g.) vendor specific bits.
Strike 2.
Last the options would be a back and forth with queries and replies. A
negotiation of capabilities and so on. XML does nothing for this.
Indeed it does not, and I was not suggesting it was a complete solution.
I don't think I've seen any complete solutions.
Strike 3. You're out.
"Send code"
--
Alex Bligh
Reply to: