[mapguide-internals] RE: PATCH: Raster stability fixes, ticket#462

[Jason Birch]

I should add that if there are potential large code changes that are waiting on a branch before working in trunk, perhaps we should branch the current trunk as 2.1, apply the eventual solutions to these problems there before tagging for release, and then merge these changes back to trunk.

Jason

-----Original Message-----
From: Jason Birch
Sent: March-30-09 11:14 AM
To: 'MapGuide Internals Mail List'
Subject: RE: [mapguide-internals] RE: PATCH: Raster stability fixes, ticket#462

Bruce,

I understand your reluctance to introduce new code at this point, but I worry that releasing another version of MapGuide Open Source with the current set of defects will slow adoption even further and turn off folks who are currently implementing.  I've lost track of the people I've talked to who either elected to stay at 1.2, or gave up on the MGE/MGOS migration entirely and went to ESRI instead, solely because of stability issues.

MGOS 2.0 was not production ready and in my experience 2.1 is still not production ready.  Releasing it at this point and requiring users to wait an additional year for 2.2 will further decrease adoption and increase community frustration.  I personally see the inherent problems with the FDO connection manager, and the problem with not reusing TCP connections for the Windows (ISAPI?) client to be blockers for this release.  I recognise that perfection is the enemy of done, but I also feel that as a project we should not be beholden to a fixed release cycle and if there are critical issues they should be addressed prior to release.

Jason