[postgis-users] PostgreSQL/PostGIS and ArcGIS Server 9.3

[Paul Ramsey]

On Fri, May 23, 2008 at 11:59 AM, Tim Bowden <tim.bowden at westnet.com.au> wrote:

> wouldn't it make financial sense for a number of large SDE users to pool
> some resources to either make SDE - PostGIS multi-master possible (as
> per my previous post which I'm sure is about to get shot down) or
> re-implement the SDE API on PostGIS as an open source project?

Spatial databases are being commoditized, but SDE is not a spatial
database, it's a revenue enhancer for ArcMap, and ArcMap is *not*
being commoditized.

No government will ever spend money on reverse engineering the SDE
line protocol to create an SDE-API clone.

Any company foolish enough to do so, would lose their money because
having invested the money they would have to either (a) give it away
or (b) compete with ESRI in the shrinkwrap arena, and their product
would always be a "poor copy of SDE".  The cost of development of such
a difficult beast does not match the revenues available when finished.

Government may not *like* paying the money for SDE, but they *have*
the money for SDE, so given the choice of the expensive, non-risky
path and the cheaper, riskier path, they will always choose the
former.

Finally, because SDE does some things vanilla databases do not (like
versioning) a SDE cloner would end up, in many cases, recapitulating
the things that are annoying about SDE, like magic side tables.

In summary, your problem is ArcMap, not SDE.

On the multi-master replication front: there's no pgsql-to-pgsql
multi-master solution, you think someone's going to toss off one that
works between heterogeneous systems?

P.