• From: Peter Flynn <peter@s...>
• To: xml-dev@l...
• Date: Thu, 13 Jan 2022 21:43:56 +0000

On 13/01/2022 16:16, Michael Kay wrote:
[...]
> I suspect those who argued that terseness was not important for XML
> were actually arguing that human readability is more important than
> message size.

Many SGML users will remember DTDs which defined all element type names
as two letters long, basically to save space and typing effort, so <TI>
was title, <AU> was author, <DA> was date, etc…and perhaps the sole
survivor, <P> for a paragraph. There were even complaints about the
excise of the markup characters themselves (< and >) being greater than
the length of [single-character] element type names.

Whatever about the length of markup, someone (Tim, I think it may have
been you) pointed out that "in future" memory and disk space would not
be as limited as they were in the mid-1990s, so terseness would indeed
be of minimal importance :-)

The absence of limits had a side-effect when XML became used as a
transport for rectangular data (eg databases, spreadsheets, etc), when
complex database stored procedures regularly constructed their own
concatenated names (presumably to represent the joins which gave rise
the the data values). This gave us element type names in some cases
hundreds of characters long, holding data values occupying a handful of
bytes. I believe some of these systems are still in existence, but
perhaps they now occupy JSON-space.

Peter

• References:
  • Why is terseness of minimal importance?
    • From: Roger L Costello <costello@m...>
  • Re: Why is terseness of minimal importance?
    • From: Michael Kay <mike@s...>