|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | RE: GISList: Interior e-gov tack irks GIS vendors (Geospatial
One-Stop initiative) |
| Date: |
04/16/2003 12:05:01 PM |
| From: |
Michael Gould |
|
|
A few interjections below...eventually some OGC empoyee may chime in.
At 11:22 16/04/2003 -0400, Anthony Quartararo wrote:
>This is quite interesting. Who was it that recently claimed that politics
>had no role in OGC ?
must have been someone ignorant of the inherent consensus process within=20
which OGC works. of course there's politics, just as there is in W3C, ISO,=
=20
etc... but at least it's pluralistic and what comes out the other end is=
=20
more less what everyone is happy with.
>It would appear, at least superficially, that ESRI was
>at the least hedging it's bet.
>While this is not an unusual tack in this or
>any other industry, especially for the giant with cannibalistic business
>practices, but it does seem rather poignant in this instance. We can only
>muse why ESRI would not throw 100% commitment behind the OGC approach.
>
>The article seems to blame the OMB for "pitting" ESRI against OGC, but why
>wouldn't the OMB and it's customers want to have alternatives/competition
>for the technology strategy that will guide the Homeland "Security"
>department for years to come. There's a lot at stake, not the least of
>which is a lot of taxpayer money. I mean, OGC gets $450K to prototype an
>architecture? I can see a lot of extra 0's getting added to the end of th=
at
>figure once this is scaled up if it ever gets off the ground. I'd be
>curious to know if OGC doled out some or all of that $450K to certain
>members to do the actual work, and I wonder if ESRI received some of that,
of course members do the work. OGC has a small staff that just steers the=
=20
boat: the membership determines where to steer.
ESRI has been a paid participant in some OGC Interop Program activities:=20
don't know about this one.
>or was it all gratis ? Jeff Burnett's contention that OGC is "basically
>[we're] not the vendor" doesn't jive: they get a contract for $450K and
>they're not a vendor? I beg to differ.
OGC acts, in this case, like any other contractor...most of the money goes=
=20
to the "subcontractors" and they keep their management part.
It seems strange to say that a pilot project or testbed has a "vendor"
>Basically, OMB pays out almost a Million $ (the combined contracts are wor=
th
>$825K, but add to that the cost of OMB staffers doing the RFP and the enti=
re
>bloated federal RFP process, and I bet it's over a million), for a prototy=
pe
>to compare what is essentially the de facto standard within the
>Federal/State/Local government institutions (ESRI) with OGC, which certain=
ly
>has a large membership, of which ESRI is a major force, and has contributed
>to the technical development of the open specifications process, etc.
>Occum's razor right ? The most likely answer is the simplest one.
William of Occum said that you should by default opt for simplicity,=20
yes. However, since an "ESRI solution" is not on par with an "OGC=20
solution" then it is difficult to say the first is simpler. If you follow=
=20
the OpenGIS architecture-by-interfaces concept/methodology, you end up with=
=20
a conceptual system (call it GIS) based on your actual needs. Now, each=20
component of this system, say the web mapping component or the coordinate=
=20
transform component, has OGC-concensus-defined interfaces (=E0 la plug-n-pl=
ay=20
in the hardware world), meaning that ANY vendor (the biggest multinational=
=20
or your cousin the Delphi programmer) can offer you one or more components=
=20
to plug into your system, iff it conforms to the interface specifications.=
=20
This separates the inner workings (logic) from the overall=20
architecture...same as plug-n-play compatibility allows you assume that=20
most any CD-ROM drive will work in your Windows PC. The magic's in the=20
interfaces.
If you continue that pnp thread, ask yourself which is "simpler": an=20
all-IBM hardware configuration, or the ability to pick and choose and make=
=20
your own hardware configuration based on price/performance, not the brand=
=20
name!!
The OGC conceptual architecture assumes that most of us no longer live a=20
single-vendor fish tank, rather in open interaction with heterogeneous=20
distributed computing.
>On a slightly different note, the article purports that "much of the data =
is
>difficult to access", referring to the needs of emergency responders and
>others when it comes to spatial data. However, I'd argue that this
>difficulty in access is more of a policy matter among different government
>organizations within a jurisdiction and others at State and Federal levels,
>rather than a bits n bytes "access" issue. I find it hard to believe that
>if ESRI does indeed command a 70% market share
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
