Re: [XCON] On encodings - observations

["Pete Cordell" <pete at tech-know-ware.com>]

This is interesting analysis, so I thought I'd try to add a bit more detail 
to it.

I've got some minor observations on the original post here, and some more 
detailed analysis in a separate post.

----- Original Message From: "Henning Schulzrinne" <hgs at cs.columbia.edu>> 
For XML, you have

> <tag  </tag>, thus, 4 bytes plus whatever length you make the tag

Should be <tag></tag> e.g. 5 bytes.

> Thus, the basic overhead is fairly similar if you choose small  element 
> names. Since numeric labels aren't self-describing at all,  this only 
> seems like a fair comparison.

My observation on this is that XSD does not lend itself to designing XML 
with small tags.  If everything is called X (or similar) the XSD doesn't 
describe the message very well.  So, although you can define an XSD with 
small tags, I think in practice you wouldn't!  (Aside: This is why Lumas 
allows for specifying a long descriptive name in the message definition, and 
specifying a short name for tagging.)

> Integers and floats are trickier: Unless you believe in ASN.1-style 
> variable-length encoding, it is quite possible that in many practical 
> cases, XML integers will be shorter since TLV always has to allocate  the 
> maximum range,

This depends significantly on the distribution of the numbers within the 
range.  If you have a 16 bit number, only the numbers 0-9 will be shorter 
than the binary encoding (not that I'm pushing binary, as you will see. 
Just trying to get the facts straight.)  This is 0.01% of the time _if_ the 
numbers are evenly distributed.  If they are evenly distributed (which is a 
big assumption) text will be longer in 98% of the cases.  For 32 bit values, 
text will be longer in 99.999% of the cases.  (again, distribution is a big 
issue here.)

> You do pay a price for very long tags, but only if you can't use gzip- 
> style compression, which essentially does the text-to-code  translation 
> automatically and without penalty.

gzip isn't free.  It costs on small devices and servers handling a lots of 
connections.  Hence I don't think you can say it is without penalty.

Pete.
--
=============================================
Pete Cordell
Tech-Know-Ware Ltd
                        for XML to C++ data binding visit
                        http://www.tech-know-ware.com/lmx
                        (or http://www.xml2cpp.com)
=============================================



_______________________________________________
XCON mailing list
XCON at ietf.org
https://www1.ietf.org/mailman/listinfo/xcon