Re: decoupling XSLT streaming recommendations from the main spec

If I had a few months to spare for writing, I think that writing books or tutorials would be a far better use of that time than refactoring the specifications
.
Sadly, I don't, so the decision doesn't arise.


> The Streaming specification must list explicitly the available implementations (seems there is just one?), 

It's not the role of a specification to list or endorse implementations. Consider the practicalities: we've just learnt of new Python implementations of XSD 1.1 and XPath 2.0. What would a working group need to do to satisfy itself that these were worthy to be listed in the spec? Equally, one of the implementations of XSLT 3.0 (Exselt) and its creator (Abel Braaksma) no longer have an active presence on the internet; what would a working group need to do to remove the product from the spec?

The XSD working group tried to maintain a web page (separate from the spec) listing implementations, but it's a nightmare to maintain such information.

Michael Kay
Saxonica

Received on Thursday, 22 September 2022 17:06:09 UTC