Re: [xml-dev] A market view

From
Didier PH Martin <>
To
Bart Schuller <>,
Date
2002-04-30T14:20:55Z
ID
<002701c1f052$3db967f0$6401a8c0@didierhome>
Thread
Re: [xml-dev] A market view
Hi Bart,

Didier said:
> On Tue, Apr 30, 2002 at 08:20:37AM -0400, Didier PH Martin wrote:
> > If the last three market segments are limited to a few thousand or less
than
> > 100 000, then its obvious that the architecture promoted by the java and
> > Microsoft world leads the world. They create the mainstream trends.
Bart replied:
> The millions of users of some technology didn't *create* that
> technology. You are also lumping together programming languages,
> frameworks and operating systems. Most of the interesting XML
> technologies are actually patterns that can be and have been applied
> succesfully to all modern and some ancient programming languages.

Didier replies:
Off course, but I didn't said that they *create* technologies simply that
they create the mainstream trends. A trend is created when you have a
critical mass of people doing something or making similar choices. It a
matter of supply and demand. The suppliers *create* and develop
technologies, the clients "create" trends by making choices and thus, in the
process, favor some suppliers to the detriment of others. So, do you mean
that a trend is created by the suply side? My point is that the trend is
driven by the demand, by the client (i.e. developers) and the choice they
make.

Bart said:
> Are the architectures that different in the different camps? SAX, DOM,
> object serialization can be found everywhere.

Didier replies:
Yes most of the languages offer tools to process XML documents. But this
never resolved the impedence mismatch between the language constructs and
the XML API. For instance, in Javascript XML constructs would be better
integrated if we could refer to an element's content as
html.body.div.paragraph = "hello world". Everytime you have an impedence
mismatch between a universe and another, you have a potential competitor
with what can resolve it.

Bart said:
>
> But you may have a point, for some people a task can only be solved if
> the vendor has handed them the appropriate API for solving that type of
> task.
> Others will look further, to add-on libraries or even to new ideas that
> they implement themselves.

Didier replies:
yes

Bart said:
> What would be more interesting to know is what makes the few people
> responsible for the frameworks in the different camps decide to pick a
> particular technology for inclusion.
> Or how many of those technologies truly originate with that particular
> framework, instead of just standardizing things like "this is how you
> create a SAX parser from a file on disk".

Didier replies:
I guess that for commercial entities its simply the demand. For instance I
remember Denis Gilbert (from Microsoft) that developers want tools that help
them. The intellisense stuff make sense if that reduces my memory overload.
So the demand is for tools that will:
a) reduce my memory overload
b) leverage what I already know
c) augment my productivity by providing parts/components. "just for
assembly" elements.
d) Give to my skills a better position in the employment marketplace
e) reduce my stress
f) reduce my work load
g) give me some fun. Provide to me a good experience
h) other factors....

cheers
Didier PH Martin