Re: Move TouchEvents spec tweaking to GitHub?

On Tue, Jan 27, 2015 at 12:11 AM, Sangwhan Moon <smoon@opera.com> wrote:
>
>
> On Tuesday, January 27, 2015, Rick Byers <rbyers@chromium.org> wrote:
>
>> On Mon, Jan 26, 2015 at 11:33 AM, Sangwhan Moon <smoon@opera.com> wrote:
>>
>>> On Tue, Jan 27, 2015 at 12:27 AM, Rick Byers <rbyers@chromium.org>
>>> wrote:
>>>
>>>> On Wed, Jan 21, 2015 at 11:46 AM, Rick Byers <rbyers@chromium.org>
>>>> wrote:
>>>>
>>>>> On Tue, Jan 20, 2015 at 7:55 PM, Sangwhan Moon <smoon@opera.com>
>>>>> wrote:
>>>>>
>>>>>> Art's seem to be fine, so it's just Josh that needs to be fixed.
>>>>>>
>>>>>> Did Doug give us write access to the w3c repo? Don't recall seeing a
>>>>>> update about that.
>>>>>>
>>>>>
>>>>> We've just now gotten write access.  Thanks Art!
>>>>>
>>>>
>>>> Friendly ping.  Sangwhan, anything else you need before pushing the
>>>> repo?  I'm anxious to get to the point where folks can start submitting
>>>> concrete PRs that we can discuss.
>>>>
>>>
>>> Apologies for the delay, I've been drowning in work. Check now.
>>>
>>
>> Totally understood.  Thank you Sangwhan - it looks great!
>>
>> For the record: here are URLs for viewing the current extensions note and
>> v1-errata:
>> http://rawgit.com/w3c/touch-events/master/touchevents.html
>> http://rawgit.com/w3c/touch-events/v1-errata/touchevents.html
>>
>> We may eventually want to use gh-pages instead of pointing people to
>> rawgit, but I think we can ignore that for now.
>>
>> I took the liberty of also rewriting the history for Art while I was at
>>> it; but I don't know how I can get Github to not show those "oh hey you
>>> pushed a branch, of course you want to merge it to master" buttons. (Note:
>>> Don't press them.)
>>>
>>
>> Yeah, the hg-style branches kind of get in the way.  Hopefully we can
>> soon be in a mode where we're working on all documents just in master (or
>> gh-pages), and everything else will be there mainly for history.
>>
>
> Actually, this case is closer to Github being naive. Not entirely sure
> gh-pages is the way to go, since that would cause master rot or lots of
> merging between the two branches. Would have been nice if gh-pages also
> worked as a named tag which we can just move to ToT as a commit hook, but I
> think gh-pages needs to be a branch. Correct me if I am wrong. Haven't
> actually checked myself.
>

>From what I've been able to figure out, gh-pages is triggered either by a
branch name or a special repo name (eg.
https://github.com/RByers/rbyers.github.io has just a master branch).  It
looks like the manifest spec <https://github.com/w3c/manifest> is using
only a gh-pages branch, so presumably that's an acceptable workflow.  As
long as we're iterating on two different branches then gh-pages isn't an
option for us.  Perhaps we should move the TEE to a new document in the
same branch as v1-errata?

If we decided upon gh-pages I would have just pushed master as gh-pages and
> we could have lived without master, might be still possible to do so if
> that's a acceptable workflow. Probably best to ask someone in w3c.
>
>
>> Does anyone have any outstanding concerns over the repo setup?
>>
>
> I forgot to reformat the code, if the PRs aren't too urgent I'd like to do
> that for master today. (If I do this, PRs based on current ToT will most
> likely conflict when merging.)
>

I don't think there's any great urgency, we can wait for the reformat.  Are
you going to reformat v1-errata also?  That's really our "master" version
of the spec IMHO (since 'master' has only the TEE note).


> If not, then I think folks (eg. Patrick) should be free to start landing
>> changes.  I'll start a separate thread to discuss logistics.
>>
>> On Tuesday, January 20, 2015, Rick Byers <rbyers@chromium.org> wrote:
>>>>>>
>>>>>>> Hey Sangwhan,
>>>>>>> Any update on this?  It looks like  josh.soref@nokia.com -> jsoref
>>>>>>> is the only rewrite we need to do.  If you're doing a rewrite step anyway,
>>>>>>> then perhaps we should also rewrite art.barstow@nokia.com ->
>>>>>>> AFBarstow too.
>>>>>>>
>>>>>>> Thanks,
>>>>>>>    Rick
>>>>>>>
>>>>>>>
>>>>>>> On Mon, Jan 5, 2015 at 1:01 PM, Rick Byers <rbyers@chromium.org>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> On Mon, Jan 5, 2015 at 10:59 AM, Sangwhan Moon <smoon@opera.com>
>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Tue, Jan 6, 2015 at 12:37 AM, Rick Byers <rbyers@chromium.org>
>>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>> [Apparently I had this message in my drafts for weeks - sorry!]
>>>>>>>>>>
>>>>>>>>>> On Mon, Dec 15, 2014 at 8:58 AM, Rick Byers <rbyers@chromium.org>
>>>>>>>>>> wrote:
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On Mon, Dec 15, 2014 at 2:56 AM, Sangwhan Moon <smoon@opera.com>
>>>>>>>>>>> wrote:
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> On Sun, Dec 14, 2014 at 3:29 AM, Arthur Barstow <
>>>>>>>>>>>> art.barstow@gmail.com> wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>> [ + Philippe ]
>>>>>>>>>>>>>
>>>>>>>>>>>>> On 12/13/14 11:58 AM, Rick Byers wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> Should we move tweaks to the v1-errata and extensions note to
>>>>>>>>>>>>>> GitHub?  I think that would be easier for collaboration than hg (easier to
>>>>>>>>>>>>>> track changes, review PRs, and avoids people having to learn something new).
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>> Yes and agree (I've been meaning to suggest this).
>>>>>>>>>>>>>
>>>>>>>>>>>>> I'll ask W3C staff to create a new touch-events repo within <
>>>>>>>>>>>>> https://github.com/w3c/> :-).
>>>>>>>>>>>>>
>>>>>>>>>>>>> Philippe - would you please create the <
>>>>>>>>>>>>> https://github.com/w3c/touch-events> rep for the Touch Events
>>>>>>>>>>>>> CG or tell me how to go about getting that directory created? The TE CG
>>>>>>>>>>>>> wants to use this new repo for Touch Events v1 errata (and possibly
>>>>>>>>>>>>> additional work such as features for v.next).
>>>>>>>>>>>>>
>>>>>>>>>>>>>  Art, if we decide to publish an official errata, would the
>>>>>>>>>>>>>> detailed change history being in github make that much more difficult
>>>>>>>>>>>>>> policy-wise?
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>> No (i.e. not a problem).
>>>>>>>>>>>>>
>>>>>>>>>>>>> BTW, my understanding is that it is possible to copy all of
>>>>>>>>>>>>> the hg history to Github so if that is the case, that should be done.
>>>>>>>>>>>>> Philippe might be able to confirm that.
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> Yes, it is possible. I took the liberty and went ahead and did
>>>>>>>>>>>> that so it can be pushed straight to the newly created repository.
>>>>>>>>>>>> Alternatively, any W3C admin can just fork that repository, and I'll nuke
>>>>>>>>>>>> my local afterwards which will make that the authoritative repo.
>>>>>>>>>>>> https://github.com/cynthia/touch-events
>>>>>>>>>>>>
>>>>>>>>>>>> One problem I see is that certain commits do not associate
>>>>>>>>>>>> correctly with the author's Github account - namely these people (in
>>>>>>>>>>>> alphabetical order)
>>>>>>>>>>>> Arthur Barstow
>>>>>>>>>>>> hiro
>>>>>>>>>>>> Josh Soref (a.k.a. timeless)
>>>>>>>>>>>> Rick Byers
>>>>>>>>>>>> Scott Graham
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Thanks Sangwhan.  I've added my google.com address to my GitHub
>>>>>>>>>>> account.
>>>>>>>>>>>
>>>>>>>>>>> I can fix this with the awesome (*cough* evil *cough*)
>>>>>>>>>>>> filter-branch command and correct the authors for those who prefer that,
>>>>>>>>>>>> alternatively the affected people can also add the mail address used for
>>>>>>>>>>>> the commit message to Github. But please let me do that before anyone
>>>>>>>>>>>> starts cloning the repository, since pushing/pulling after a history
>>>>>>>>>>>> rewrite can cause a decent amount of arcane behavior for those who haven't
>>>>>>>>>>>> dealt with a repository that has been touched by the evil forces.
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>> *Sangwhan*, Doug created an empty (uninitialized) repo for us
>>>>>>>>>> here: https://github.com/w3c/touch-events.  Once you're happy
>>>>>>>>>> with them, are you able to push all your migrated commits there?
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Absolutely. I just need to know which parts of the history I
>>>>>>>>> should rewrite to make sure the authors associate correctly. Alternatively
>>>>>>>>> we can just push it forward and hope nobody complains.
>>>>>>>>>
>>>>>>>>
>>>>>>>>  Perhaps we should at least try to fix Art's commits?  *Art*, are
>>>>>>>> you able to add your "art.barstow@nokia.com" address to your
>>>>>>>> GitHub account to fix this?  Or should we look at rewriting history instead?
>>>>>>>>
>>>>>>>> I've sent an e-mail to Scott and Josh to ask about theirs (and cc'd
>>>>>>>> you).  If we don't hear back from them soon then I'd suggest we just go
>>>>>>>> ahead.  hiro's committs appear to be only initial repo test commits so I
>>>>>>>> think it's fine to ignore them.
>>>>>>>>
>>>>>>>> *Doug, *can the CG (or at least specific individuals that ask) get
>>>>>>>>>> write access to that repo?  Or do we need to work on a clone and submit
>>>>>>>>>> PRs?  If you need to list CG folks by hand, then the relevant GitHub
>>>>>>>>>> accounts I know off off the top of my head are:
>>>>>>>>>>
>>>>>>>>>> jacobrossi
>>>>>>>>>> cynthia (Sangwhan)
>>>>>>>>>> mbrubeck
>>>>>>>>>> scottgonzalez
>>>>>>>>>> AFBarstow
>>>>>>>>>> RByers
>>>>>>>>>> mustaqahmed (Mustaq on my team will be helping with the spec)
>>>>>>>>>> tdresser (Tim on my team may also help)
>>>>>>>>>>
>>>>>>>>>>  Thanks,
>>>>>>>>>>    Rick
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>>> --
>>>>>>>>>>>> Sangwhan Moon [Opera Software ASA]
>>>>>>>>>>>> Software Engineer | Tokyo, Japan
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Sangwhan Moon [Opera Software ASA]
>>>>>>>>> Software Engineer | Tokyo, Japan
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>> --
>>>>>> Sangwhan Moon [Opera Software ASA]
>>>>>> Software Engineer | Tokyo, Japan
>>>>>>
>>>>>
>>>>>
>>>>
>>>
>>> --
>>> Sangwhan Moon [Opera Software ASA]
>>> Software Engineer | Tokyo, Japan
>>>
>>
>>
>
> --
> Sangwhan Moon [Opera Software ASA]
> Software Engineer | Tokyo, Japan
>