Hey guys, I just checked out the new UI, and it looks really great. Definitely a large improvement over the struts stuff, way more usable.
Most everything seems to work decently, so I'm wondering why we're calling the next release alpha2? Alpha to me implies little more than a working prototype. Trunk right now is definitely better than that. If we call it an alpha then we won't get that many people checking it out and giving feedback.
Is there any big objection from calling it beta1? I think it's at the point where we'd like people giving real feedback on it, no? And when we fix any bugs they uncover we can point them at the nightlies.
Chris
--
Chris Holmes
OpenGeo - http://opengeo.org
Expert service straight from the developers.
Chris Holmes ha scritto:
Hey guys, I just checked out the new UI, and it looks really great. Definitely a large improvement over the struts stuff, way more usable.
Most everything seems to work decently, so I'm wondering why we're calling the next release alpha2? Alpha to me implies little more than a working prototype. Trunk right now is definitely better than that. If we call it an alpha then we won't get that many people checking it out and giving feedback.
Hmm... the rationale is that we did so many changes and so little
hands on testing that it's very likely to break in unexpected
ways. The official wikipedia classification is here:
http://en.wikipedia.org/wiki/Software_release_life_cycle
Alpha's are meant for internal testing... which we don't have.
If you look at Ubuntu they released 9.04 beta one month before
the final release, after 4 alphas.
One thing that I also seemed to know about alphas are they are not
feature complete. And 2.0 is not, given that community schema is
not working right now.
My vote is to call it still alpha. Also consider that we're
discussing switching from jdbc datastores to jdbc-ng ones
for this release, and the postgis-ng datastore has received
zero testing, solid, besides the unit test harness.
I don't think it even passes the cite tests (due to some
weird tests comparing dates with strings, but anyways).
Cheers
Andrea
--
Andrea Aime
OpenGeo - http://opengeo.org
Expert service straight from the developers.
Hmmm... Ok, if we are switching to postgis-ng then that's sufficient reason for me to go with alpha.
Andrea Aime wrote:
Chris Holmes ha scritto:
Hey guys, I just checked out the new UI, and it looks really great. Definitely a large improvement over the struts stuff, way more usable.
Most everything seems to work decently, so I'm wondering why we're calling the next release alpha2? Alpha to me implies little more than a working prototype. Trunk right now is definitely better than that. If we call it an alpha then we won't get that many people checking it out and giving feedback.
Hmm... the rationale is that we did so many changes and so little
hands on testing that it's very likely to break in unexpected
ways. The official wikipedia classification is here:
http://en.wikipedia.org/wiki/Software_release_life_cycle
Alpha's are meant for internal testing... which we don't have.
If you look at Ubuntu they released 9.04 beta one month before
the final release, after 4 alphas.
One thing that I also seemed to know about alphas are they are not
feature complete. And 2.0 is not, given that community schema is
not working right now.
My vote is to call it still alpha. Also consider that we're
discussing switching from jdbc datastores to jdbc-ng ones
for this release, and the postgis-ng datastore has received
zero testing, solid, besides the unit test harness.
I don't think it even passes the cite tests (due to some
weird tests comparing dates with strings, but anyways).
Cheers
Andrea
--
Chris Holmes
OpenGeo - http://opengeo.org
Expert service straight from the developers.
One thing that I also seemed to know about alphas are they are not
feature complete. And 2.0 is not, given that community schema is
not working right now.
My vote is to call it still alpha. Also consider that we're
discussing switching from jdbc datastores to jdbc-ng ones
for this release, and the postgis-ng datastore has received
zero testing, solid, besides the unit test harness.
I don't think it even passes the cite tests (due to some
weird tests comparing dates with strings, but anyways).
If postgis-ng is the line that divides alpha from beta I would say let's keep postgis-ng in unsupported/incubation for a while longer. I know I just voiced a +1 for switching all to jdbc-ng but I think postgis is a special case due to the care and maintainence it has received over the years. While the performance issues can be addressed in the short term it will take us more than a few months imo to catch all the special cases that the current one does. I fear that if we put postgis-ng on the critical path to a beta release then we should not expect that beta any time soon.
Cheers
Andrea
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
Justin Deoliveira ha scritto:
One thing that I also seemed to know about alphas are they are not
feature complete. And 2.0 is not, given that community schema is
not working right now.
My vote is to call it still alpha. Also consider that we're
discussing switching from jdbc datastores to jdbc-ng ones
for this release, and the postgis-ng datastore has received
zero testing, solid, besides the unit test harness.
I don't think it even passes the cite tests (due to some
weird tests comparing dates with strings, but anyways).
If postgis-ng is the line that divides alpha from beta I would say let's keep postgis-ng in unsupported/incubation for a while longer. I know I just voiced a +1 for switching all to jdbc-ng but I think postgis is a special case due to the care and maintainence it has received over the years. While the performance issues can be addressed in the short term it will take us more than a few months imo to catch all the special cases that the current one does. I fear that if we put postgis-ng on the critical path to a beta release then we should not expect that beta any time soon.
On the other side, if we wait last minute to do the switch we'll never
get to RC. Bigger changes, the sooner the better imho.
Cheers
Andrea
--
Andrea Aime
OpenGeo - http://opengeo.org
Expert service straight from the developers.