|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | RE: GISList: OGC and Standards |
| Date: |
01/07/2003 11:20:54 AM |
| From: |
Dimitri Rotow |
|
|
Chris,
Fair points, but I would respectfully disagree.
> First, I think it is true that the GIS user with desktop or even
> server-based GIS software available for their use is now in the
> minority of
> total GIS users, and this 'GIS Specialist' category will continue to get
> smaller in percentage. Most GIS users do not know what GIS is, nor should
> they know. They are using various enterprise-level applications,
> through the
I think you are making too big a leap from stating a) most people don't want
to hassle with GIS and b) therefore they are all using web-based GIS
applications. b) does not follow from a). For that matter, I don't grant
you a) as a proposition, but even if I did there are many ways to deal with
that other than web-based applications. For example, MapPoint is a way of
bringing GIS (of sorts) to the masses.
> Web or desktop or client-server configurations, and make use of spatial
> information and/or spatial operations as part of their use of that
> application. I am not talking about 'MapQuest users', I am talking about
> people at local government level, state agency, engineering firms, or
> private enterprise, who are using specific business-process applications
> that contain GIS. This use of GIS will continue to grow, and grow
> rapidly.
>
I think what you'll find over time is that *if* GIS is sensibly done as are
modern Microsoft applications, if it is sensibly priced and if it does not
take a year's course in GIS to be able to use, then many "average" people
take to GIS like ducks to water. I think what is going on is that you are
misidentifying people's dissatisfaction with clumsy, legacy GIS applications
with a general dissatisfaction for GIS.
The arguments you are using above remind me of those that were prevalent
when Borland brought out Turbo Pascal: it was said that people didn't want
to hassle with programming, etc, etc. It turned out there were millions of
people who wanted to program *if* they could do it in a cost-effective and
congenial way.
I don't disagree that there is a large contingent of people who are
perfectly content with "canned" GIS applications. However, there are many
people, millions of people, who would like to go GIS as a mainstream
horizontal application (like Access in DBMS or Excel in spreadsheets) *if*
they can do GIS in a cost-effective and congenial way.
However, even for canned applications that employ GIS you don't have to do
it in a low performance, technologically obsolete way as proposed by OGC.
> The OGC is involved in work that I feel will benefit this growing use of
> spatial information, and I applaud those OGC folks who are providing their
> time and talents in doing so. But, the OGC is not focusing on
> this audience
> of users at the expense of all others.
>
I'd disagree with that. The OGC stuff is focussed at low bandwidth, low
performance, over-the-web applications. How else do you explain things like
the ultra-low performance architecture of spatial DBMS's working at the
object level or GML?
>
> Second, GML has some promise for use in GIS. This promise is in
> inter-application exchange of spatial information. Should it be used as a
Why on Earth would you use a carrier pigeon technology like GML for
inter-application exchange of spatial information? Why not use something
more efficient? Seriously, if you are crafting a spec for interprocess
communication you don't use a texty thing like GML. That's not how
competant programmers do things.
> data format or as a distribution medium? Certainly not in its present
> iteration, even if the UK's OS has gone down that road. GML can be used to
> transfer points and lines representing a query from one application to
> another, however. I would hope that work continues in this vein.
>
Let's hope the work consists of a replacement for GML. If you abandon the
idea that GML makes sense as a storage format, and if you consider that in
the (hopefully) high speed world of binary communications between processes
you don't want horribly slow carrier pigeon interfaces like GML, what good
is it?
Regards,
Dimitri
To unsubscribe, write to gislist-unsubscribe@geocomm.com
________________________________________________________________________
GeoCommunity GeoBids - less than $1 per day!
Get Access to the latest GIS & Geospatial Industry RFPs and bids
http://www.geobids.com
Setup a GeoCommunity Account and have access to
the GISDataDepot DRG & DOQQ Catalog
http://www.geocomm.com/login.php
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
