Next Round

james anderson James.Anderson at
Thu Jan 21 17:08:07 GMT 1999

Don Park wrote:
> >First, we will need to define a new type, namely Name.
> Turning names into first class objects does solve the namespace problem as
> well as providing an opportunity to solve the 'intern' problem by adding
> name comparison methods.  Implementations that maintains 'interned' name
> within Name 

For which application domains is it necessary to intern the string itself?
Have "cross-namespace" name comparisons turned out to be significant enough to
make this "second level" interning pay off?

>         will obviously be very efficient past the tree building process.

Doesn't the "first level" intern of the name object suffice?

> It is not a very well known fact that Docuverse DOM SDK uses 'interned'
> names and it has been very useful to many of our customers although it does
> cost about 10% speed penalty during tree building.

It would be interesting to hear if any of them use functions which include the
prefix among thir arguments and, if so, what for? Perhaps there's something
peculiar about our approach, but 100% of our coding has been in terms of the
universal names.

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at
Archived as:
To (un)subscribe, mailto:majordomo at the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at

More information about the Xml-dev mailing list