Open Stack

On 10/05/18 16:52, Doug Hellmann wrote:
> Excerpts from Zane Bitter's message of 2018-05-10 16:38:37 -0400:
>> On 17/04/18 05:24, Thierry Carrez wrote:
>>> Hi everyone,
>>>
>>> As you know the Technical Committee (the governance body representing
>>> contributors producing OpenStack software) meets with other OpenStack
>>> governance bodies (Board of Directors and User Committee) on the Sunday
>>> before every Summit, and Vancouver will be no exception.
>>>
>>> At the TC retrospective Forum session in Sydney we decided we should
>>> more broadly ask our constituency for topics they would like us to cover
>>> in that discussion.
>>>
>>> Once the current election cycle is over and the new TC chair is picked,
>>> we'll come up with a proposed agenda and submit it to the Chairman of
>>> the Board for consideration.
>>>
>>> So... Is there any specific topic you think we should cover in that
>>> meeting ?
>>
>> There's one topic I've been thinking about that I think would be
>> valuable to discuss with the Board and the UC. I don't know if we still
>> have time to add stuff to the agenda for Vancouver, but if not then
>> consider this my advance submission for Denver.
>>
>> OpenStack was bootstrapped using a very powerful positive feedback loop:
>> in (very) broad-brush terms it started with a minimum viable product;
>> users for whom that was enough to entice them tried it out and offered
>> suggestions; vendors who wanted to sell to those users (as well as the
>> users themselves) implemented the suggestions; both groups joined the
>> Foundation, which marketed OpenStack to folks with similar needs.
>>
>> Obviously that is a good thing, but it also comes with the danger of
>> getting trapped in a local maximum. Users for whom the product has not
>> yet met the threshold of minimum viability are generally not going to
>> show up, and their needs are no match for the feedback loop set up with
>> the users who _have_ shown up. (Specifically, we are arguably only just
>> now approaching the minimum viability point for the types of cloud-aware
>> applications that are routinely written against the APIs of the big 3
>> proprietary clouds.)
>>
>> How can we avoid (or get out of) the local maximum trap and ensure that
>> OpenStack will meet the needs of all the users we want to serve, not
>> just those whose needs are similar to those of the users we already have?
>>
>> Discuss.
>>
>> thanks,
>> Zane.
>>
> 
> This does feel like an excellent topic for one of these strategic
> discussion sessions, but I think the agenda is already full for this
> particular meeting. Maybe we can discuss it within the TC between now
> and Denver so we have a good way to frame the question and discussion at
> that meeting?

+1. As usual I've been struggling to find a good balance between being 
too abstract and too specific, so I'd appreciate help in framing the 
question to avoid sending the discussion directly into the weeds. cdent 
already had a good suggestion on IRC.

cheers,
Zane.