mina-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Emmanuel LŽcharny <elecha...@gmail.com>
Subject Re: RC1, RC2, etc
Date Sun, 10 Jan 2010 14:27:41 GMT
Alan D. Cabrera a écrit :
> On Jan 10, 2010, at 5:50 AM, Emmanuel LŽcharny wrote:
>> Alan D. Cabrera a écrit :
>>> On Jan 9, 2010, at 11:54 AM, Emmanuel Lecharny wrote:
>>>> Alan D. Cabrera a écrit :
>>>>> Cool.  I think we should rename 2.0.0-RC2 to 2.0.0 to better 
>>>>> communicate that intention.  WDYT?
>>>> Well, until we have a stable version (I mean, no serious bug), I 
>>>> would keep RC*. Then we can switch to 2.0.0-GA ot 2.0.0.
>>> What's the point in releasing a version of MINA that is so badly 
>>> unstable that we must release it under RC2, RC3, RC4, etc.?
>> I never said 'so badly unstable'. I said, we need to have a stable 
>> version which has been proved bug free for a certain time at least 
>> before switching to a GA. So far, we can't anticipate bugs, as we 
>> don't think we intoduce bugs when we write code !
>> We certainly don't want to wait 4 years before going for a GA, but 
>> considering the serious issues we found in RC1, we may want to be 
>> sure that RC2 is used and tested by as many users as possible before 
>> we can consider it as a GA.
>> OTOH, we can also release a 2.0 and switch to 2.0.1 as soon as we fix 
>> bugs.
> That's what I was thinking.  People understand that the first release 
> of any version will tend to be a bit flakey; only the Pope and my 
> mother-in-law are infallible.  :)  We remedy that by quickly releasing 
> patches, 2.0.x.  Releasing an RC will not attract people to start 
> using it.  Slow lumbering RC releases gives the impression that the 
> community has stalled.
Maybe. As a community, it's just a matter of what we decide to do here. 
And I don't think that the pope is infallible : remember Gallileo. Now, 
when it comes to your mother-in-law, that's a different story !
>> IMHO, it's all about the message we send to our users. People tend to 
>> think that RC are unstable by essence, we tend to think that we are 
>> simply cautious. If the GA is buggy, then it's the opposite : users 
>> think we are lazzy, and we think that it does not matter, we just 
>> have to provide a new bug fix release.
>> There is no way we can catch two birds with a single stone ...
> Yep, and that's why I was thinking that Mina should follow the 
> standard convention of tight RC releases that span a few weeks 
> followed by a real release w/ many micro patch releases.  The nice 
> thing about patch releases is that it reveals to outsiders that the 
> community is active and vibrant.
> Just my 2 cents...
What about a vote ? Something like : no RC, just plain release, with as 
many releases as necessary, assuming that we will always have bugs, and  
that plain release is more likely to be used and proven stable than a RC...

View raw message