|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | RE: GISList: Re: GIS P2P * Market Research Question * |
| Date: |
01/24/2003 01:27:44 PM |
| From: |
Dimitri Rotow |
|
|
>
> A group of us in the UN Environment Programme (UNEP) are reviewing options
> for developing a p2p geo-refereneced data publication mechanism.
>
> The target is not so much the 'serious GIS practitioner' but rather the
> field researcher who may have a limited number of tabular datasets, with
> some form of spatial referencing, that they're willing and able to publish
> in a manner that makes them accessible for rendering as maps, charts etc
> via OGC-type interfaces. Essentially, data discovery - 'I have some such
> data'. Embelishments could incorporate allowing downloads of the data, or
> choosing to upload to central repositories should the custodians see the
> benefit of additional relaibility or availability. This obviously is not a
> commercial model, but a public good scenario.
>
> Our envisaged application include on-the-fly consolidation of data on e.g.
> marine mammals being gathered by scattered and disparate researchers often
> at the the thin end on internet connectivity.
>
> The requirement for high-quality metadata need not be re-iterated here.
> However, our potential contributors are those who probably have
> never heard
> of OGC, ISO/TC-211 or even GIS!
>
Mick,
First, let me commend you and your colleagues at UNEP for placing data
online. There's a vast array of UN GIS data that could be helping people
around the world if only it were made accessible, and the UNEP site is a
really ground-breaking step on the way to getting that data out of the file
cabinets and hard disks of the UN and into people's hands. The site looks
good and functions well and the data it presents is really useful. Well
done!
Second, I'd encourage you to consider the following points in any P2P
project:
1) Keep it simple. I think you are already focused on this but it bears
reinforcement. Don't slow things down to create the most perfect,
metadatable wonderful thing that anyone might ever need. Get it going in
the simplest way and then grow it.
2) Think in terms of real GIS, not just presentation or rendering. Modern
GIS that can do anything that was possible just a few years ago with $50,000
in GIS software per seat now costs less than $250 a seat. Anyone who can
afford Microsoft Office can today afford to do real GIS in a considerably
more sophisticated way than what was done just a few years ago when (quite
likely) most of those data sets were created. That means if the data
available is worthwhile in terms of content the main constituency to use
that data is people who will be able to work with it in sophisticated ways
within a real GIS environment, not just some dumb viewer or render-only web
site. Plus, people who work with the data in a real GIS environment provide
an important "multiplier" effect by adding value to the data, combining it
with new data sets, presenting it in many forms for those who cannot do GIS,
etc. So, encourage the availability and exchange of the actual data.
3) Encourage the use of whatever format the data is in - avoid citing OGC,
since OGC standards work against cost-effective usage and dissemination. To
take a specific case, the recent threads here on GML and Adena Shultzberg's
excellent reporting in GIS Monitor show what an utterly idiotic and
dysfunctional thing GML is, so don't encourage people to place obstacles in
the path of sharing data by adopting OGC obstacles. Encourage people to
present their data in whatever formats they have it in, right now. Any
truly modern GIS package will enable your users to work with data in any
popular format effortlessly, all for a cost much less than Microsoft Office.
The last thing you want is for all those scattered and disparate researchers
to think they've got to re-engineer their data into some
protect-the-dinosaurs, OGC non-standard for it to be worthwhile.
4) Encourage the use of modern, low-cost, high-performance GIS software.
Good data is worthless if a researcher has to spend thousands of dollars a
seat on legacy GIS software to do anything with the data. It's always been
especially important for people in the Third World to get good value for
their money and now the budget crunch affecting researchers in the First
World means that people need to avoid getting ripped off by legacy GIS
packages that cost ten or twenty times more than necessary. You can support
the use of high quality, high performance and high value software by
providing links to freeware packages for various formats and to those modern
packages that provide good value, perhaps by providing special emphasis on
packages that cost less than $300 a seat. I'll bet that if you wanted to do
a quick review to see what packages could work with the various formats
emerging in a P2P site you'd have no difficulty getting the modern, more
active vendors to send you free packages for your r
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
