incubator-kato-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stuart Monteith <stuk...@stoo.me.uk>
Subject Re: Happy New Year - and what do we do next?
Date Thu, 07 Jan 2010 17:02:08 GMT
Hello Robert,
     I'm working on documenting the release process on our Wiki. Craig 
Russell posted a link
earlier that is proving useful as a guide to follow, that should give us 
something concrete to discuss.
I'll then get someone to run it through, as best they can, to make sure 
it is reproducible. Shouldn't take
too long, but I'd like to smoothing over some of the rough edges, maven 
seems to be able to do a lot for us.

Thanks,
     Stuart

Robert Burrell Donkin wrote:
> On Wed, Jan 6, 2010 at 7:14 PM, Craig L Russell<Craig.Russell@sun.com>  wrote:
>
> <snip>
>
>    
>>> Quoting Robert Donkin from the 14th December:
>>>
>>> "i've been floating the idea of a multi-stage audit to try to help
>>> podlings pass the technical side of their first release without too
>>> much pain. fancy trying it out?"
>>>
>>> I replied yes, it sounded like a good idea, but we've still to get details
>>> of what that involves.
>>>        
>> I'm not sure either. Robert is busy with some schoolwork so you might
>> inquire on the incubator list to see if he's communicated his ideas more
>> widely. I don't recall any details.
>>      
> given a project that's ready for it's first release (for example, have
> a working, documented and reliable build) there isn't any real need to
> actually produce an artifact using the full release process for most
> of the problems to be ironed out. most of the areas where podlings
> find difficulties could be reviewed long before the release but for
> this to work someone who's done a lot of review would need enough time
> to reach consensus on fully documented guidelines for these prior
> review processed.
>
> the basic idea is that:
>
> * reviewing the source for license type issues can be done from the source
> * if a podling wants to release a binary then the release process
> needs to be replicatable and documented. if it is then anyone should
> be able to build a sample binary release from the source and check the
> basic stuff podlings usually fall down on.
>
> once the podling understands how to do that stuff right, then the
> final bit of the jigsaw is understanding the voting process.
>
> - robert
>    

-- 
Stuart Monteith
http://blog.stoo.me.uk/


Mime
View raw message