Hi,

I have cleaned up the branch and the corresponding parts in the trunk to fix
for missing header files.  Here is an updated RAT report.

http://people.apache.org/~svkrish/RAT_0.91/RAT_0.91.txt

Thanks

- Venkat

On 6/25/07, Jean-Sebastien Delfino <[EMAIL PROTECTED]> wrote:

Simon Nash wrote:
> Comments inline.
>
> Simon
>
> ant elder wrote:
>
>> On 6/21/07, Jean-Sebastien Delfino <[EMAIL PROTECTED]> wrote:
>>
>> <snip>
>>
>> as long as something works, I'm not sure why we would exclude it from a
>>
>>> Tuscany release.
>>
>>
>>
>> I think we need a bit more clarity about what it means to include
>> something
>> in the release. Is it just having the code included in the src
>> distro, or
>> included in the src distro as part of the build, or also included in
the
>> binary distro, or included in both distro's along with itests, samples,
>> readme's or web site doc, and mention of the new thing in the release
>> notes?
>>
> My take is that included in the release means included in both distros
> along with itests, samples, readmes or web site doc, and mention of the
> new thing in the release notes.
>

+1 from me.

I'm expecting the itest suite to grow over time as not all extensions
are covered yet.

I just added a sample for the implementation-resource extension, and am
still working on a sample for the implementation-spring extension.

>>
>>> From the 0.90 release things were pulled out for not being quite
>>> there, a
>>
>> lot of the time spent before the final release artifacts got done was
>> because people reviewing the distro's wanted all things up to a certain
>> standard. Getting all this done can take a lot of work. Last minute
>> changes
>> often cause unexpected blocking problems in the distributions
>> resulting in
>> respins and more delay.
>>
>> If we just include everything "that works" is someone reviewing a
>> release RC
>> going to complain that some new sample is missing a readme, that a demo
>> should have an Ant build script, or that some new extension doesn't
even
>> have a sample? If things must be of a certain standard then I don't
>> think
>> its reasonable to expect the release manager to do all the work to get
>> things there.
>>
> I agree that these things are needed and that the burden of doing them
> should not fall on the RM.
>

+1 that's what I had in mind as well when I said "works", as samples and
a readme are the basic requirements for a user to be able to get an
extension to work.

>> How about:
>> - by default everything is kept in the src distro unless there's some
>> reason
>> not to
> >
> I'm not sure about this. What is the value of including modules in the
> source distro but exluding them from the build? The latest levels of
> this code (presumably better) are available in trunk. Since these things
> weren't built or tested as part of the release, the snapshot state they
> were in on release date has no particular significance.
>
> I also recall a discussion some time ago (not specific to Tuscany but on
> [EMAIL PROTECTED] IIRC) in which it was said that the source distro
> is what really defines the content of the release, and the binaries are
> included for convenience. For all of these reasons, it seems better for
> the source and binary distros to have matching contents unless there's
> some reason not to.
>

Matching source and bin distros make more sense to me as well.

>> - only the things mentioned on the release wiki page get included in
the
>> build, binary distribution, and mentioned in the release notes
> >
> This seems OK for new function. I'm not sure what process we use for
> JIRA fixes. They don't seem to be listed on this page. I would expect
> that marking the "Fix version field" accordingly is all that is needed.

Not sure that it's realistic.

>
>> - after the brn is cut we need to ask on the ML before adding new
>> things to
>> the wiki release page
> Agreed.

+1

>
>> - adding something to the wiki release page implies some commitment
>> to help
>> get it to the required standard in line with the release schedule.
> >
> Agreed.

+1

>
> Simon
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>


--
Jean-Sebastien


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


Reply via email to