Upcoming VeloSLC meeting

[Mike Place]

We had a great VeloSLC last month, which lots of great conversation about dev vs. ops and some Agile methodology thrown in for good measure.

This month, Andrew has volunteered to give a preview of his upcoming presentation he's preparing for Agile 2009. He told me it's going to be "content heavy and slide lite", so have no fear of Powerpoint and come by and check it out.

We're on for this Friday at 10AM at the downtown city library.

We have a new room at the library this time around. It's on 3rd floor on the west side. I'll be there early so you can just give me a call or send a text message once you arrive for directions. It's 801.349.8421.

Send me questions or post to the list. Hope we can see you all there.

-mp

[Jen Browne]

I'm looking forward to the meeting -- can't believe it's already been
a month. I got some very good information out of our last meeting.

I don't know what you're going to cover, Andrew, but one of the
concepts I've been struggling with this month is how implementing
structure and process fits in an agile organization. I have been
asked to assume full ownership of IT Infrastructure process here at
Backcountry and increase the maturity of the IT Operations group while
avoiding the dogmatic boat of ITIL and TOGAF. Any ideas?

[littleidea]

I have a few ideas...

I'm doing a mini series with Isreal Gat and Micheal Cote on the
subject. Here's the first podcast.
http://www.redmonk.com/cote/2009/07/17/agileexec004/

Generally, I think the principles of systems management are the same
as web ops, it's just the time scales and SLAs that change.

You have the same problems that need to be solved. How do I get new
systems? What services need to be running? What services are running
now?

What you want to avoid is the situations where the are systems
running, but no one really knows what or why they are running, until
there is a fire, and people are freaking out. Sanity lies somewhere
between total chaos and ITIL, and the perfect process is probably
different from organization to organization and will likely change as
your services and their criticality do.

I'm going to blame Alistair Cockburn for this, but my personal
preference for process is start with the sloppiest thing that could
work, then improve with constant reflection and refinement.

I'd be happy to throw around ideas in the context of specific problems
you are hoping to solve.