|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | RE: GISList: Compressed Terrain Data |
| Date: |
01/07/2003 02:35:40 PM |
| From: |
Dimitri Rotow |
|
|
Let me respond to two, somewhat related posts:
> I have the same question after reading this discussion. I think
> people are not used to the open source profit model yet, where it
> is not the software that makes money, but the support of said
> software. OSS (Open Source Software) companies, including Red Hat
and Cameron Crum's post:
>
> So I guess my obviously rhetorical question would be after all
> this....why don't
> you make the Manifold product open source and give it away? Like
... and there were some other posts asking why I would criticize OGC and
then suggest a new proprietary DEM format was not a good idea.
I think it is important to avoid confusion to resolve the discussion into
separable issues. The original posting asked what was thought of the idea
of introducing a new, proprietary format for DEMs that claimed some
advantages over existing formats, principally a higher degree of compression
in storage.
There are a variety of business and public policy issues in that proposal.
From a business perspective, it's up to you what you do, but my guess is
that going into the business of selling formats can be financially
successful only to the degree you can bottle up widely-disseminated data
within your format (to force people to use your format to get at the data)
or to convince people to use your format within their own installations to
get the benefits it offers for their own operations. Since a format by
itself is worthless you have to recruit makers of GIS programs or other
applications that interact with DEMs to use your format. Obviously, you
don't have a business unless you can get third party software developers to
use your format.
Here the interests diverge. The success of Manifold or any other GIS
depends upon access to lots of GIS data. Data, after all, is the "fuel"
which a GIS burns (except for that tiny fraction of the market that actually
creates new maps). That's why Manifold reads 80 different formats.
However, it is very much in our interest for GIS data to be available for
free to our customers. We therefore tend to resist proprietary formats that
might make it difficult for customers to get free access to public data. We
are forced by the prevalence of some proprietary formats, like EO0, to
import them but that doesn't mean we think that the proliferation of such
proprietary formats are either good for us or good for our customers.
One way to finesse that is for the format proprieter to provide free SDK's
(Software Development Kit) for software vendors to use that build in support
for his or her format. If such SDKs are free and sensibly licensed, that
takes away the sting of a proprietary lock upon public data. A good SDK
license is like that used by Earth Resources for their ECW compressor, which
allows vendors in a free of cost, no-strings attached way to embed support
for their formats. A bad SDK license is like that used by LizardTech for
MrSID, which the last time I read it included provisions that we believed
essentially required the software developer to forswear any competition with
LizardTech. That is, they just didn't want the developer to not use the SDK
to directly compete, they wanted as a condition of licensing the SDK to
prevent the developer from ever competing with LizardTech even using the
developer's own independently-developed products. For a company like
manifold.net, that might one day provide zero cost compression as just
another one of few hundred more new features, that's not something we could
agree to.
From a public policy perspective, the last thing we want would be for public
agencies to adopt a proprietary format for data distribution if such a
format required a restrictive license provision like the MrSID one in order
to allow third party programs to access the data. That not only would
bottle up the public data in a proprietary format, it would also allow the
format licensor to use his or her control over the format (and thus the
public data) to erect artificial barriers to prevent competition in the data
compression market.
So, my original response to the request for comments 'way back when was to
point out that no, the last thing we want is yet another proprietary format
and to advise Cameron that if he wanted to maximize his chances of getting
it adopted to use a sensible license in the style used for ECW and to avoid
a pernicious license in the paranoid style used for MrSID.
On Openness
Somehow this discussion has gotten off the main track of the sense or lack
thereof of introducing yet another proprietary format for DEMS and onto a
track as to whether "Open" source software is good or bad. I've read
another posting, in addition, that wonders why on the one hand I would
criticize OGC (the so-called "Open"GIS consortium) efforts while arguing
against a proprietary fo
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
