incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ross Gardler (MS OPEN TECH)" <Ross.Gard...@microsoft.com>
Subject RE: Incubator report sign-off
Date Wed, 31 Dec 2014 03:58:19 GMT
John,

Actually John I disagree with one of your examples (Ripple). This is actually a case where
things have gone as they would expect.

The mail you link to is from me. I had previously made the IPMC aware of the issue prior to
that email on the mailing list. I was asked if I was undertaking to fix it (I replied yes
and requested the podling added me as a mentor in order to do so). The podling report indicated
that getting a release out was a focus "No release made as yet, this will be the first item
to recieve attention." 

The report does not need more detail than that since the IPMC had already been made aware
that there was a problem, that it had been spotted and that the community and mentors indicated
that they were to address it.

Finally, if you review the shepherds notes from that report they acknowledge the concern and
the fact that there was activity to address it.

Ripple still has not addressed the issue raised those emails. Therefore it will not graduate
until it does. The email you link to makes this perfectly clear.

This is, in my opinion, exactly what should be happening. We provide oversight to ensure project
acts as an Apache project. If it does so we graduate it, if it doesn't we retire it.

I do agree with the overall intention of your mail, but it seems I disagree on what adequate
oversight is.

Ross

-----Original Message-----
From: John D. Ament [mailto:johndament@apache.org] 
Sent: Tuesday, December 30, 2014 7:47 PM
To: general@incubator.apache.org
Subject: Re: Incubator report sign-off

On Tue Dec 30 2014 at 1:26:31 PM Ted Dunning <ted.dunning@gmail.com> wrote:

> On Tue, Dec 30, 2014 at 7:26 AM, John D. Ament 
> <john.d.ament@gmail.com>
> wrote:
>
> > > Absolutely not just noise. Take the extra 2 seconds to add your 
> > > sign
> off.
> > >
> >
> > I disagree.  Checking a check box is much different than adding
> meaningful
> > comments, either on mailing lists or on the report itself.
> >
> > For example, which gives you better info that I feel confident in
> Tamaya's
> > board report.
> >
> > My check here: https://wiki.apache.org/incubator/December2014
> >
> > or my comments in this thread:
> >
> > http://mail-archives.apache.org/mod_mbox/incubator-tamaya-
> dev/201411.mbox/%3CCAOqetn8wkYuDNkTwkpKKOGzu%3Ds_cf4VMT5A9_e8mdpM6mOh-
> 6Q%
> 40mail.gmail.com%3E
> >
> > All the check does (from my point of view) is give someone a brief
> summary
> > that things are looking good.  The check mark doesn't imply any due 
> > diligence on the mentor's part.  It's very misleading to see it that way.
> > Take a look for example at the log4cxx2 podling's report.  It has 
> > mentor sign off, but the contents are barely present.  The only 
> > reason it has mentor sign off is because the mentor wrote the 
> > report, after I (as the
> > shepherd) reminded the podling.
> >
>
> John,
>
> Are you seriously suggesting that the board should be delving into all 
> the incubator mailing lists to determine whether you are paying 
> attention to your mentoree groups?
>

No, not in the slightest.  But someone needs to look at it.  Our current notion of a board
report is completely on the honour system.  It doesn't safeguard from the chance (which from
what I can tell is more often the
case) of a mentor writing and signing a report saying it's good to go.

You can see some examples of this effect here:

http://mail-archives.apache.org/mod_mbox/logging-log4cxx-dev/201412.mbox/%3C1418063938.3890690.200338789.1D0EE7B3%40webmail.messagingengine.com%3E

There are also cases where there are clear issues w/ the podling but aren't getting communicated
properly on the report (or maybe just oversight?)
       
http://mail-archives.apache.org/mod_mbox/incubator-ripple-dev/201412.mbox/%3CBY2PR03MB490FFD83B71E97C269A12DF99660%40BY2PR03MB490.namprd03.prod.outlook.com%3E

My point is that just because there's a checkbox checked doesn't mean there's issues.  Maybe
what would help is to have, during shepherd perhaps, some coaxing in to putting more into
the issues for the IPMC/board section.

Maybe it's more of a "don't hesitate to put something in that area" thing that needs to happen.

John


>
> The check-box is the concise way that you indicate that the activity 
> on the mailing lists is happening.  There is a known defect with 
> checkboxes in that they can be ticked without mentoring activity 
> behind them, but that doesn't mean that we should introduce a new 
> failure mechanism where there is good activity but no tick box.
>
> Yes, the tick box is supposed to be an echo.  It is a redundant 
> summarization.  And it is very helpful because all the tick boxes are 
> in one place for easier review.
>
Mime
View raw message