|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | GISList: SUM - GIS P2P * Market Research Question * |
| Date: |
01/20/2003 08:21:54 PM |
| From: |
Anthony Quartararo |
|
|
Ok, here it is folks, even if it is premature, since no doubt there are
several others that have not yet responded, but I am traveling for a bit and
wanted to get this back out before I leave.
The original post that I sent can be found on the gislist archive at
http://spatialnews.geocomm.com/community/lists/daylist.phtml?month=3D2003-0=
1&l
ist=3D2
To this, there were many responses on the list. Many people focused on
metadata.
Paul Ramsey was first to point out the need to metadata in any solution
resembling Napster GIS. He also stated that such a system, if implemented
correctly would be orders of magnitudes more useful that traditional FTP
systems. Paul later responded that firewalls would be an issue that would
need to be negotiated, especially where public agencies are concerned, and
since he suggests that this community [public agencies] would be among the
largest users of this Napster GIS model, downloading would not be a problem,
but uploading would be.
Holly Glaser however responded in private to say that it would be a civil
offense and outright theft of intellectual property, and I should in no way
be encouraging the development of such technology. Here concerns seem
centered around accountability and liability in passing data back and forth,
and either through intentional malicious or unintentional ignorance, the
data ends up cause harm in some way. Add to this the versioning control
problem and the spurious and dubious quality and pedigree of a given file
over time. Magnify that by X users, and that could pose a real problem.
The publisher could assert the right to sue for actual and punitive damages
should this scenario play itself out in the market place.
Robert Heitzman reinforced Paul Ramsey's comments and added that an
acceptable level of Quality of Service (QoS) would have to be guaranteed by
the client [machine] in order to stay on the system, a continual qualifier
of participants if you will. He also makes the point of suggesting that all
technical obstacles aside, revenue is key, and this would be challenging.
He suggests some form of online service that would allow participants to
"rent" space that they would then be willing to share with others in the
system. This might find a niche in the small-medium corporation market
since many might not be able to afford a full-fledged FTP download site. He
further suggests a model where owners or stewards of a defined geography can
offer their data in the public domain along the same lines as Google allows
a restricted/defined search of a particular newsgroup, in this case search
only within the organizations data and not necessarily lump it into the
world wide web, etc. If I understand his comments correctly, the search
function would also be limited to those belonging to that jurisdiction (ie.
A county).=20=20
William Howell commented that in Napster, there was no real centralized
server, rather client machines that download the software actually speak to
each other individually, and that Napster only stored registered users of
it's software, log in account info, etc. but no central repository of data.
He suggested that Kazaa would be a better model because of the various forms
of technical discrimination that the software allows when conducting a
search. He mentions the Fast Track protocol that any Napster GIS could
leverage in it's implementation.
Pat Waggaman joined the discussion to disagree with others assertion that
metadata would be crucial. The comments focus on volume of data and choice,
that is, the person doing the dowloading would either have choices or not,
putting metadata aside. If the only source for a given topo sheet does not
have metadata, the user still has a choice whether or not to download it.
If there are multiple choices, then all things being equal, the user would
select the file with the most metadata that suits their specific needs. Pat
also goes on to say that rather than charging people who load data into the
system, a revenue model can be derived from charging downloaders, either a
la carte, or some form of subscription. Pat echoes Robert's vision of a
"distribution channel" for public organizations that is free.
Paul Ramsey responded to Pat's email suggesting that metadata, or the lack
of it, especially real-world coordinates, would be a real show stopper for
this type of system.
Sonny Parafina contributed with a post suggesting that the current OGC
specifications actually addresses metadata requirements. Also, Ron Lake
provide an URL for the specific GML metadata requirement for coordinate
systems.
Paul Ramsey commented to both Sonny & Ron that many of the potential users
of the NapsterGIS would not necessarily part of the OGC movement, and by
implication, not really benefit from the OGC specifications. Sonny came
back to Paul's comments and argued that the OGC provided a ready-made
[vehicle] to accomplish some form
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
