Re: Ada Reference Manual 2012 in info format

[Jerrid Kimball <jerrid@kimball.co>]

On 05/09/2012 09:09 AM, Dmitry A. Kazakov wrote:
> On Wed, 09 May 2012 15:56:32 +0200, Georg Bauhaus wrote:
>
>> On 09.05.12 15:00, Dmitry A. Kazakov wrote:
>>> On Wed, 09 May 2012 07:06:06 -0500, Nasser M. Abbasi wrote:
>>>
>>>> I think XML is not good for math.
>>>
>>> XML is not good for anything.  I don't care about Web but in automation it
>>> becomes a growing problem that precious bandwidth and human resources are
>>> spent on chewing XML mess.
>>
>> Writing math papers and making data traveling automation networks
>> seem rather different use cases.
>
> Yes, though XML is usually advocated as a universal solution for
> everything.
>
> Just two weeks ago, or so, XML was proposed to handle Ada source code.
>
> Wouldn't be great to have math in XML:
>
> <operation name='*'>
>     <number>
>       <digit position=1>3</digit>
>       <digit position=2>1</digit>
>     </number>
>     <number>
>       <digit position=1>5</digit>
>       <digit position=2>6</digit>
>     </number>
> ....
>
> no?
>

Oliver was hardly suggesting that XML is universal solution, good grief. 
  However, XML is a huge player on many fronts for a number of reasons 
including being easily readable by humans and being standardized.  There 
are, of course, instances where XML has been abused in very horrible 
ways--what hasn't been (Yes, even Ada has been.  You know it!)--but this 
in no way precludes it from being used for its intended purpose as a 
language for describing documents.  What an idea!  Obviously this will 
never change, but I'm sure we'll have good fun with the XML.

Keep in mind you are no obliged to come anywhere near it if you do not 
so desire.  So--and this is directed at Randy also--feel free to stay 
away from anything Oliver or I perceive as innovations on the front of 
community building ad if you don't have useful discourse, let your keys 
be silent because this sort of banter isn't helping him, me, CLA, Ada or 
anyone or anything else.  It's just trolling.  The limited bandwidth of 
your automation system networks and other difficulties you encounter are 
hardly relevant to RM formatting and frankly aren't even worth 
mentioning in this discussion.  George said it well with brevity. :)

That example is a bit far-fetched, too--almost obnoxiously so.  That's 
the sort of XML you might see on a late night infomercial for YAML. 
"Does your XML look this?"  "Don't you hate all those *ugly* tags?" 
"Always cutting yourself on those pesky brackets?"

I'm not entirely sure of the endianness of your numbers (you specified 
everything except the most obvious), but I imagine it could be reduced 
to a single tag.  It would depend on a lot of factors, but something 
like this doesn't seem out of the question:

<multiply>
    <left>31</left>
    <right>56</right>
</multiple>

or dare I suggest:

<multiply operands="31 56" /> <!-- multiplication is commutative after 
all -->

or have I gone too far?

Of course DSA has to represent things across the line.  I haven't done a 
whole lot with it, so far.  How does GIOP send data across the line?

That's my contribution for the day.

Have a good one.

Oliver ;)


-
Jerrid