[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: RDDL(2): new version up

• To: "Simon St.Laurent" <simonstl@s...>
• Subject: Re: RDDL(2): new version up
• From: Henrik Martensson <henrik.martensson@b...>
• Date: Sat, 24 Jan 2004 23:58:40 +0100
• Cc: XML Developer List <xml-dev@l...>
• In-reply-to: <r02000200-1028-C14FC8624BA311D8AD220003937A08C2@[192.168.124.11]>
• References: <r02000200-1028-C14FC8624BA311D8AD220003937A08C2@[192.168.124.11]>

On Wed, 2004-01-21 at 00:52, Simon St.Laurent wrote:
<snip>
> 
> I have a few answers to that one:
> 
> * no Microsoft support

Can be a factor, yes, but there are lots of stuff around that don't have
Microsoft support. RELAX NG, for example.

> * limited browser support generally

Other link models didn't have browser support when they were created
either. A link recommendation will get tool support if:

* Other XML recommendations use it, XHTML, for example.
* There are low level tools available, for example
  an XLink processing package for Java

The above are connected. If enough recommendations used XLink, the
incentive to write software would be greater. If there were a few
software packages around, this would be a strong incentive to use XLink
in other recommendations.

> * too many years in the making

True, and this has certainly been discouraging. On the other hand, XLink
has been around for a couple of years now, and there has been ample time
to make up for the slow start.

> * most people still don't get/want multi-ended links

On the other hand, Simple XLink is fairly, err, simple.

Actually, multiended links have been on the wish list of every major XML
project I have worked with since 1999. Major corporations are clamoring
for them, though I will admit that it is debatable whether they always
need them...

> * XPointer was slow to arrive, over-complex when it arrived

Also true. Yet, the few XLink applications I have seen (or built) do
very well without XPointer support.

> * standardization in linking isn't that exciting anyway

On the contrary, its very exciting. The same linking problems occur over
and over again in different systems. Each time, the problems are solved
in a slightly different manner, resulting in enourmous waste of
resources and money.

On the other hand, an XLink dialog I built for XMetaL several years ago
is used in several different companies without a single line of code
having been changed.

> * overlaps with RDF and Topic Maps for more abstract possibilities

XML Topic Maps use XLink, so I don't think XTM stole any thunder from
XLink.

> 
> That's a brief list.  Given more time, I can certainly find more.

Well, you certainly make some points that are worth thinking about. And
I do believe it is possible to come up with more. However, from a
technical point of view, I believe there are also valid counter
arguments to most (not necessarily all) of them.

/Henrik

• References:
  • Re: RDDL(2): new version up
    • From: "Simon St.Laurent" <simonstl@s...>

• Prev by Date: Re: Refreshed genx.h, plus some plans
• Next by Date: Re: Refreshed genx.h, plus some plans
• Previous by thread: Re: RDDL(2): new version up
• Next by thread: Re: RDDL(2): new version up
• Index(es):
  • Date
  • Thread