On Tue, Apr 24, 2012 at 8:15 PM, Olof Kindgren <[email protected]> wrote:
>
>
> 2012/4/24 Julius Baxter <[email protected]>
>>
>> On Tue, Apr 24, 2012 at 5:31 PM, Jeremy Bennett
>> <[email protected]> wrote:
>> > On Tue, 2012-04-24 at 17:18 +0100, Julius Baxter wrote:
>> >
>> >> ChangeLog? Why not put it into the nice, text-based asciidoc we now
>> >> have for or1200, so that it shows up automatically when we build the
>> >> doc, and then we need to keep only 1 place updated.
>> >
>> > Hi Julius,
>> >
>> > That's a different use case. The purpose of a ChangeLog is to see how
>> > the document has developed, to help you when working out the thought
>> > processes and time line for why a feature is there. More obviously
>> > valuable for code than documentation, but still very useful.
>>
>> Hi Jeremy
>>
>> I'm not saying drop all change time and explanation information, I'm
>> just saying that it would be convenient to place it there. Basically,
>> merge the ChangeLog with that document - they're both text-based and
>> it would be handy to have it in the spec document. I think we had a
>> long discussion about this before and I'm not sure we came to an
>> agreement on having a version of the core source and spec
>> documentation which was in sync.
>>
>> Julius
>> _______________________________________________
>> OpenRISC mailing list
>> [email protected]
>> http://lists.openrisc.net/listinfo/openrisc
>
>
> It's this thread that covered our changelog discussions
> http://lists.opencores.org/pipermail/openrisc/2011-August/000063.html
>
> There wasn't any formal agreement, but my interpretation of it is that we
> should be more careful with our commit messages (which we have been since
> then) and make a ChangeLog รก la kernelnewbies.org/LinuxChanges with all the
> highlights of the new release. No one really seemed to like having changes
> in the source files headers, so I guess we could just let them be, and
> remove them eventually.
>
> For the tar ball releases, I would like to have the usual ChangeLog, NEWS,
> README, MAINTAINERS files included, and I'm also all for reducing the number
> of sources for the changes, so why not just include the ChangeLog file in
> the generated asciidoc instead? That would give us a single source, and
> still include the changes when you read the spec

Including the ChangeLog in the asciidoc sounds fine then. I just want
to avoid multiple places for change information to go, and for that to
be missing from the spec document - and this satisfies both!

Cheers

Julius
_______________________________________________
OpenRISC mailing list
[email protected]
http://lists.openrisc.net/listinfo/openrisc

Reply via email to