[teampractices] [Engineering] Feedback requested on proposal for creation of Agile Specialist Group

[Whatamidoing (WMF)/Sherry Snyder]

Matt, I basically agree with you.  The problem is that hundreds of our
"customers" do not.  Erik said the other day, "2) We do have a higher
tolerance for breaking things".  Do you know who "we" is in that sentence?
 Hint:  It's not our "customers".

A noisy segment of our customers either don't believe Agile's promise (that
frequent and early pain results in better products) in the first place, or
they don't care that this is (probably) the best of the available options.
 They're saying, "Stop poking me with a stick twice a week".  Agile
proponents are saying, "Well, the only alternative is that we stab you with
a big knife twice a year".  They are very loudly in favor of "don't poke me
at all".


TLDR:  When those anti-Agile users discover a proposal to spend half a
million dollars a year on making sure that the users keep getting poked
with sticks twice a week, then the people in favor of this proposal should
not be surprised at the results.




Whatamidoing (WMF)/Sherry Snyder
Community Liaison
Wikimedia Foundation, Inc.



On Tue, Mar 4, 2014 at 3:46 PM, Matthew Flaschen <mflaschen at wikimedia.org>wrote:

> Forgot to reply all.
>
> Matt Flaschen
>
>
> On 03/03/2014 11:00 PM, Whatamidoing (WMF)/Sherry Snyder wrote:
>
>> (2)  Did you all know that a significant proportion of our user
>> community is unhappy with the way that Agile affects them?  "Release
>> early, release often" from the user's perspective means "have horribly
>> and/or newly busted stuff screw up your work twice a week."  Among a
>> certain (vocal) subset of the community, this proposal will be
>> understood as "spend lots of money to continue screwing up, instead of
>> hiring someone to fix the bugs before we users ever see them".
>>
>
> The full saying is, "Release early. Release often. And listen to your
> customers." (http://www.catb.org/~esr/writings/cathedral-bazaar/
> cathedral-bazaar/ar01s04.html).  'Customer' isn't exactly the right word
> here, but the 'listen' is still critical.
>
> If we just develop things in isolation, then release to production rarely,
> two things will happen:
>
> 1. Users will not be able to provide feedback until we've gone too far
> down the wrong path.  By releasing early and often, we get prompt feedback
> and can steer more quickly.
>
> 2. Sometimes, more bugs will pile up at those release points (since they
> did not appear/were not noticed in earlier testing, or *could not* appear
> in low-load or non-production environments)
>
> We are working to provide more of a ramp for features (e.g. with feature
> flags and Beta Features), but I don't think we should move to a slow
> release cycle.
>
> Matt Flaschen
>
>
> _______________________________________________
> Engineering mailing list
> Engineering at lists.wikimedia.org
> https://lists.wikimedia.org/mailman/listinfo/engineering
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.wikimedia.org/pipermail/teampractices/attachments/20140304/7ad97002/attachment.html>