Firstly I must say that I'm NOT criticising anything - the recent release of 1.3.0 Alpha was a great milestone, this is just a suggestion I'm putting out there for discussion...
Don't shout at me for this suggestion!
Now, I think the release of 1.3.0 Alpha highlights a weakness in the process of releases, in that there are several 'killer' bugs IMO. So much so that I've dropped back to the previous release as [for me] 1.3.0 was unusable.
(Just to be clear - I've not submitted anything on these 'bugs' as they've all been reported by other already...)
Of course I realise that this is an ALPHA release and the bugs are kind of the point but the delay in releasing this seems to have been because of bugs which caused it to be unusable (obviously I'm not a member of the team so this is a huge assumption on my part) - in my case there were enough errors to make it unusable anyhow = little/no benefit for the delay (again I'll point out that I may be wrong in this assumption).
Also I think that the Alpha releases are viewed more as the next release of MP and not as a 'test' build.
So can I suggest that for future releases the community is strongly warned of the status of the release and it's put out there for community testing ASAP (once the current bug list has been addressed). Similar to what is done now but using the power of the great community here to harness the testing power of the many different set-ups in use for testing...
This test version would only be installable in a parallel manner (somehow detect that MP is already installed and uses many of the existing resources) so as to not disrupt the existing HTPC setup (so the significant other can still watch the soap operas!). Akin to the Chrome Canary idea - can only be installed as a secondary solution.
This should result in a faster, smaller release, another bug list being generated, another milestone to hit and a final release that has been thoroughly tested in many different scenarios.
Would this not speed up the development? It's fairly obvious that all scenarios for MP cannot be properly tested so why take the time?
Just release little and often in the test stages rather than the big list approach currently used.
Play nice please and discuss...
Don't shout at me for this suggestion!
Now, I think the release of 1.3.0 Alpha highlights a weakness in the process of releases, in that there are several 'killer' bugs IMO. So much so that I've dropped back to the previous release as [for me] 1.3.0 was unusable.
(Just to be clear - I've not submitted anything on these 'bugs' as they've all been reported by other already...)
Of course I realise that this is an ALPHA release and the bugs are kind of the point but the delay in releasing this seems to have been because of bugs which caused it to be unusable (obviously I'm not a member of the team so this is a huge assumption on my part) - in my case there were enough errors to make it unusable anyhow = little/no benefit for the delay (again I'll point out that I may be wrong in this assumption).
Also I think that the Alpha releases are viewed more as the next release of MP and not as a 'test' build.
So can I suggest that for future releases the community is strongly warned of the status of the release and it's put out there for community testing ASAP (once the current bug list has been addressed). Similar to what is done now but using the power of the great community here to harness the testing power of the many different set-ups in use for testing...
This test version would only be installable in a parallel manner (somehow detect that MP is already installed and uses many of the existing resources) so as to not disrupt the existing HTPC setup (so the significant other can still watch the soap operas!). Akin to the Chrome Canary idea - can only be installed as a secondary solution.
This should result in a faster, smaller release, another bug list being generated, another milestone to hit and a final release that has been thoroughly tested in many different scenarios.
Would this not speed up the development? It's fairly obvious that all scenarios for MP cannot be properly tested so why take the time?
Just release little and often in the test stages rather than the big list approach currently used.
Play nice please and discuss...
Last edited: