[Top] [All Lists]

Re: [xsl] Adding entity declarations to DOCTYPE in xml output

2019-02-26 19:22:56

Agreeing with everything Eliot is saying, but looking at the glass
half full. Just because a feature is there does not mean we need to
use it - or use it for/as intended. I am not above using a parsed
entity for an easy hack. But I am not, these days, expecting to
recommend it as a feature in a system. For all the reasons Eliot says
and more.

On topically, it is a demonstration in itself that XSLT plugged the
hole left if, in 1999 or so (about when it became usable), you needed
an inclusion mechanism, and couldn't or wouldn't use entities for
whatever reason. Yet at the same time, there were systems up and
running including precursor systems of Michele's EAD stack, which
availed themselves of this feature even to the point of depending on
it for key functionality -- because there was no XSLT and no other
recommended way to turn syntax into boilerplate or perform other sorts
of easy and necessary substitutions of text or even markup. Because
SGML had no XSLT (we leave aside the four or six brave visionaries who
did this all in DSSSL, which would include Eliot Kimber maybe one or
two other readers), perforce it needed entities. Because SGML offered
entities, EAD had entities, and because EAD was SGML and XML carried
that legacy forward onto the web, it was easier for EAD to become XML
-- especially as it did not have to retool its entire stack, or move
to something that didn't exist yet (XSLT) or wasn't 'proven' (DSSSL
even if Eliot had already proven it, or any number of other solutions
that might have been offered then or since).

So because it made it easier for EAD (or name-your-vertical SGML
application) to migrate to XML, entities came along with it. Tail
wagging dog? Thus evolution lurches forward. It is not at all clear in
retrospect that this was not a correct choice strategically, even if
it has left us with technical debt of some nature.

See the thing is, Michele's application is still running. I call this
a success. It has probably paid for itself, is one way of putting it,
plus dividends. Had external parsed entities or internal DTD subsets
been dropped from XML -- who knows what else we might have lost along
the way?

On the other hand, if you are designing a system today and you need
boilerplate expansion? Just look at DITA or any of a half dozen other
improved approaches.

I think it was a help that XML could support named character entities,
e.g. & ldquo ; & mdash ; and the like. Since people were already used
to them from HTML.

Cheers, Wendell

On Tue, Feb 26, 2019 at 5:47 PM Liam R. E. Quin liam(_at_)fromoldbooks(_dot_)org
<xsl-list-service(_at_)lists(_dot_)mulberrytech(_dot_)com> wrote:

On Tue, 2019-02-26 at 21:53 +0000, Eliot Kimber 
That general requirement (simple string macros) can be satisfied
using XInclude, which is implemented by most, if not all, of the
modern XML parsers.

Pretty much. It's unfortuate that we never added control over XInclude
processing to the doc() function in XPath/XQuery/XSLT/...

I could go farther and say that the original SGML design of DTDs was
entirely misguided as well and should never have been done that way
and certainly shouldn't have been carried into XML (again, I
certainly argued *for* them at the time)

It's questionable as to have far from SGML we could have taken XML and
had it go anywhere. As it was for a while i was getting daily calls
from someone called Charles opposed to making DTDs optional :-)

 but that's easy for me to say now. At the time that SGML was being
defined and implemented the DTD syntax seemed perfectly sensible and
it took a long time for us to recognize the inherent problems with
DTDs as they exist in SGML and XML.

In particular, because they are a purely syntactic mechanism DTDs are
a security risk and provide no reliable declaration of the actual
semantic document type of the document that exhibits the DOCTYPE

You're right - public identifiers do *not* do what many people think,
and neither does a DOCTYPE declaration. A combination of a namespace
URI and a version attribute seems better, but is still not perfect.

It always bugged me that SGML DTDs didn't guarantee a particular root
element. Yuri’s take was to point out that he could write a book
starting with a single paragraph and working outwards, and have the
document remain DTD-valid at all times. But that just says that SGML
needed a separate concept of document-valid alongside what-you-have-
here valid.


Liam Quin, https://www.delightfulcomputing.com/
Available for XML/Document/Information Architecture/XSLT/
XSL/XQuery/Web/Text Processing/A11Y training, work & consulting.
Web slave for vintage clipart http://www.fromoldbooks.org/

Wendell Piez | wendellpiez.com | wendell -at- nist -dot- gov
pellucidliterature.org | github.com/wendellpiez |
gitlab.coko.foundation/wendell  - pausepress.org
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
EasyUnsubscribe: http://lists.mulberrytech.com/unsub/xsl-list/1167547
or by email: xsl-list-unsub(_at_)lists(_dot_)mulberrytech(_dot_)com

<Prev in Thread] Current Thread [Next in Thread>