incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Craig L Russell <Craig.Russ...@Sun.COM>
Subject Re: Making up policy on the fly
Date Wed, 19 Aug 2009 16:23:20 GMT
Hi Ant,

On Aug 19, 2009, at 12:23 AM, ant elder wrote:

> On Wed, Aug 19, 2009 at 6:11 AM, Ralph Goers< 
> > wrote:
>> On Aug 18, 2009, at 5:13 PM, Joe Schaefer wrote:
>>> ----- Original Message ----
>>>> From: Ralph Goers <>
>>>> To:
>>>> Sent: Tuesday, August 18, 2009 8:00:00 PM
>>>> Subject: Re: Making up policy on the fly
>>>> On Aug 18, 2009, at 4:22 PM, Craig L Russell wrote:
>>>>> So I found it:
>>>>> Please take a look at
>>>> [1]
>>>>> If this document is not normative, please let me know. Granted,  
>>>>> it says
>>>> "should" and not "must", so if there is a discussion presumably  
>>>> it's
>>>> about
>>>> whether "should" means "must".
>>>>> Otherwise, let's shut this discussion down and start following the
>>>>> rules.
>>>> OK then. There is something wrong with how hard that was to find  
>>>> though.
>>>> And
>>>> you're right, instead of "If An Artifact Contains Code Under  
>>>> Several
>>>> Licenses,
>>>> Should It Contain Several License Files?" it would be much better  
>>>> if it
>>>> read "If
>>>> An Artifact Contains Code Under Several Licenses, May It Contain  
>>>> Several
>>>> License
>>>> Files?".  However, I think we should pretend it does.
>>> It's written the way it is because we are aware that not all Apache
>>> projects
>>> comply with the recommendation.
>> Being ambiguous is no way to set policy.
> +1. And i'd go further and say while the ASF does not have a clear
> policy on something that is an ASF issue then the IPMC should not be
> trying to make up our own, and, we should be trying to shield
> poddlings from these ambiguities.

I'd go exactly the other direction.

The fact that policy has "should" instead of "must" tells me that best  
practice is to do the "should" but for historical reasons, it's not a  
requirement. Historically some PMCs have not done the "should" and we  
can't go back and change history, but going forward, projects "should"  

The incubator should be the place where "should" becomes "must" so as  
to align newcomers to Apache with best practice. If there's some very  
good reason for a podling to eschew best practice, then let's hear it.


> If we run into issues of unclear
> policy then take it to legal-discuss@ or board@ or where ever is
> appropriate but in the meantime let poddlings release as long as they
> aren't violating existing policy. We all know that release early
> release often is good but thats really difficult for poddlings as
> release votes so often run into these types of problems.
> One thing I think might help is to change the IPMC poddling release
> voting process in the same way that the committer voting process was
> changed recently - allow release votes to be held on the poddling
> mailing lists and all they need is 3 binding votes from IPMC members
> and they only need to bring it to general@ if they can't get enough
> binding votes.
> Comments?
>   ...ant
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

Craig L Russell
Architect, Sun Java Enterprise System
408 276-5638
P.S. A good JDO? O, Gasp!

View raw message