Mailing-List: contact dev-help@openoffice.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@openoffice.apache.org
MIME-Version: 1.0
In-Reply-To: 
 <CAHW+f2JPkr8TSKvaUhU45edYyK2Giphjt6CpO5CVdqWxh9Nqbg@mail.gmail.com>
References: 
 <CAHW+f2JPkr8TSKvaUhU45edYyK2Giphjt6CpO5CVdqWxh9Nqbg@mail.gmail.com>
Date: Wed, 28 Aug 2013 08:54:32 -0400
Message-ID: 
 <CAP-ksoh_W1+280+7TpKHcN4p=fXEVt-JnzEOmdeP=Rq=8SNE_Q@mail.gmail.com>
Subject: Re: Slashdot Article
From: Rob Weir <robweir@apache.org>
To: "dev@openoffice.apache.org" <dev@openoffice.apache.org>
Content-Type: text/plain; charset=UTF-8

On Wed, Aug 28, 2013 at 12:45 AM, Keith Curtis <keithcu@gmail.com> wrote:
> One of the points I would like some clarification on is in the article
> defending ASF by Andrew C. Oliver: http://www.infoworld.com/print/225555
>
> In it, he wrote: "Nearly all of the very active Apache OpenOffice
> developers work for IBM directly or indirectly."
>
> Can someone explain more about this? For example, is there a table showing
> who everyone works for? From that, one could make a chart showing how many
> of the code changes in AOO were made by various companies. LibreOffice
> publishes such charts:
> http://documentfoundation.files.wordpress.com/2013/07/developers6.jpg From
> my research, I'd guess AOO 4 was at least 80% IBM, but I've not been able
> to find out information on several people so it is an estimate.
>

Two basic kinds of metrics:  results-oriented and in-process metrics.
The first tells you how well you did in your outcome.  Things like
numbers of downloads, number of positive reviews, sentiment analysis
of twitter comments, growth in Facebook followers, number of
regression defects reported, etc., are all outcome based metrics that
tell us where we've done well or done poorly.  In the end we're a
software publisher, not a coffeeklatch, so success is based on what we
deliver to the publish in terms of the OpenOffice product.

The downside of outcome metrics is you cannot measure them until the
work is done.  So it is too late to make corrective changes if needed.
 That is why we also have in-process metrics, things that can be
measured during the release cycle, things that correlate to future
outcome metrics.  This includes things like defect find rates, test
coverage, etc.

Both of these are useful kinds of metrics that support our overall
goal of making a great office productivity suite available to millions
of users for free.

Metrics that do not measure results, or are not correlated with
results, are not very interesting, though they may be the focus of
other projects and used primarily for propagandist purposes.

It is natural for those who have excellent results to point to their
results, and those who don't to distract from their results.

In any case I hope this helps explain why we are interested in
particular metrics.

Regards,

-Rob

P.S. If you do want a metric for project diversity, I did propose one
on my blog a while ago, and showed that AOO was more diverse than LO.
You can read it here:

http://www.robweir.com/blog/2012/11/libreoffices-dubious-claims-part-3-developers.html

I thought it was interesting, but since it is not an outcome and is
not correlated with an outcome I do not think it worth the time to
calculate this metric, other than to respond to FUD.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@openoffice.apache.org
For additional commands, e-mail: dev-help@openoffice.apache.org