Open Stack

Agreeing with Dean Troyer is generally something I find myself doing.  He's
a smart dude.

That being said, making openstack easier to install / maintain is probably
the first and foremost thing openstack needs to worry about.  Without an
eco-system of smaller / medium users to keep the project grounded
enterprise requirements will destroy the code base.  It has, in my opinion,
already done great harm.

That being said, if OpenStack wants to be a VMWare competitor, as seems to
be the case from enterprise desires.  I am all for that.  I don't care what
direction OpenStack takes.  Whether it's in my personal technology future
depends entirely on whether or not it makes sense to push workloads to it.
Currently, and for the first time in 5+ years it doesn't make sense to.  I
don't even want to run OpenStack at home, because it takes so much of my
time to keep it up in even a small environment.  Whether OpenStack ends up
going after AWS, Docker, or VMWare, and really you need to pick one. It
needs to be easier and less human heavy in cost to deploy and run.  It's
needed it for five years.  We have SIMPLE cosmetic tickets such as
https://bugs.launchpad.net/horizon/+bug/1017606  which have only JUST now,
4 years after having been logged in launchpad, been handled.

You ask an operator ( the guys running the environment ) to rattle off some
hate, and they all have a list they can come up with from the last month
alone.  But, the real issue... is asking someone who failed to install
openstack what hung them up.  Because I assure you, there are far more
people who have failed to deploy OpenStack than have succeeded.  I've met
plenty of them.  Some of them big orgs.  But more importantly... most of
them the little and medium business guys... who whether you believe it or
not are worth a hell of a lot more to this project than any enterprise
partner.

-Matt

On Thu, Feb 11, 2016 at 7:27 PM, Dean Troyer <dtroyer at gmail.com> wrote:

> On Thu, Feb 11, 2016 at 5:46 PM, Jay Pipes <jaypipes at gmail.com> wrote:
>>
>> For those of us who, years ago, warned about the dangers of "enterprise"
>> workloads -- i.e. pets -- not being a good fit for OpenStack, and how the
>> future was cattle workloads and cloud-native apps, this article kinda hit a
>> nerve.
>>
>
> I hear a lot about making OpenStack "Enterprise Ready", and you are right
> Jay, that is not where we started. Spending the amount of resources we are
> today for the last 20-??% of ESX capabilities is starving other things we
> need to do to or should have done years ago.
>
> The desire to be all things to all people is strong, and leads us to a
> place where we are only mediocre at best to anyone.  What market is
> OpenStack the _best_ at today?  After 6 years of trying we should have
> nailed at least one of them by now.  (This is a real question, I'm not on
> the product side of the community and don't know.)
>
>
> So back on-topic... let's not be "ubiquitous", as Roland mentioned.  The
> "works well at all scales" makes me nervous too, but doing a smaller set of
> things in a scalable manner may be achievable.
>
> "To produce an Open Source Cloud Computing platform that is
> easy to use, simple to implement, interoperable between deployments,
> works well at all scales, and meets the needs of users and operators of
> both public and private clouds."
>
> And follow that by beefing up the work on "easy to use" and "simple to
> implement", IMHO even ahead of becoming the enterprise's OSS alternative to
> (or front-ends for) VMware ESX farms.  After all, it's in our mission
> statement, and those are the first two things listed that we want to do!
>
> dt
>
> Also, the thing in that article that hit me the most was Cockcroft's
> comment about not making things easy for developers soon enough.  I wonder
> if we are really doing that yet anyway?
>
> --
>
> Dean Troyer
> dtroyer at gmail.com
>
> _______________________________________________
> Foundation mailing list
> Foundation at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/foundation
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/foundation/attachments/20160211/86b619e2/attachment.html>