incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rich Bowen <>
Subject Re: Concerning Sentry: A disagreement over the Apache Way and graduation
Date Thu, 12 Nov 2015 14:56:40 GMT
Hash: SHA1

On 11/10/2015 10:00 AM, Pierre Smits wrote:
> That is nice! Apache pages drawn up by a member of the Apache
> Software Foundation with the input from many  (both ASF members and
> others) and hosted/communicated through ASF means, and then saying
> that those 'are not Foundation'. And that by/through the fingers of
> a fellow board member....

Pierre, it's as though a document were written by the PMC of the OFBiz
project, and then the people that had been involved in that
conversation were to assert, unilaterally, that it must be adopted by
the entire ASF, without any debate either on board@ or members@   That
is the concern that Greg is citing, and one that I can find no fault wit

When I encouraged, in another thread on this list, that mentors use
the maturity model document as a useful measure to judge projects, I
was careful to phrase it as a useful tool, rather than as new policy.
You'll remember, no doubt, that when it was proposed as new policy,
there was not consensus around doing that. (How's *that* for a tactful

I think that it's a useful tool. I think that mentors should consider
it when they're looking at podlings, because, frankly, it's bloody
hard to judge project readiness to graduate. I've been in the business
(professionally, across several jobs) for some time, of attempting to
measure project health, and it's really, really hard, and most (all)
programmatic ways of doing so get it wrong at least half of the time.
As such, they should be used only as a early warning system, and not,
as Greg says, as a yardstick. Or as a checklist.

Consider, for example, the numeric health score given by the tool. If it gives your project a -6, it means that
you should look for warning signs and consider whether something is
broke, but it does NOT NECESSARILY mean that your project is trouble.
Likewise, if it gives you a 9.6, it doesn't NECESSARILY mean that
everything is peachy, because there are so many unmeasurables.

So, yeah, Greg's concerns here are valid, and I agree. I want as many
tools as possible, and, I will always mistrust all of them to a
certain degree. The 'maturity model' is not a checklist, and if we
start seeing people using it as such, we should be very worried.

Always be very, very careful what you measure, because people will
naturally optimize for those things, to the detriment of some of the
less measurable things.

- -- 
Rich Bowen - - @rbowen - @apachecon
Version: GnuPG v2


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message