Tuesday, February 27, 2007

Limits to IT Services - Questions

For any sort of 'new' description to be accepted where it counts - in the Marketplace, it's going to have to fulfill a number of criteria. And before competition takes away the market...

  • Who is going to 'declare', ratify or back the standard?
    It has to be an existing organisation, and with some clout.
    Could the IETF or ICANN be persuaded? If not them, then who?

  • It has to be 'accessible' to the people that will use it.
    Barriers to Entry have to be low - or it will take decades to get used, even a little.
    • Cost of basic materials have to be low. Near zero is possible if they are published on-line like RFC's and W3C standards.
    • The basic manual has to be short and not intellectually difficult. This has to be implemented in 'The Real World' by real, ordinary I.T. professionals - especially those who have families, commitments and outside interests.
    • Training costs have to be reasonable - no more than a few hundred dollars per person, and something freely available over the net.
    • Consulting, training and "Operational Evaluations" have to be available at modest costs.

    • There have to be convincing business arguments for change available for owners and managers.

  • Supporting Software, Templates and proformas
    • 'Reference' implementations need to be available - for free.
      These can be as simple as spreadsheets and formatted documents - or real software.
    • There are enough good Free Open Source packages available to use as a base.

  • Professional Associations. Somewhere to 'network', find good consultants, swap war stories, give those interested a place to learn more.

  • A viable continuous update process. Like the IETF's RFC process.
    Like the Internet, any sort of standard can never be 'finished', only 'complete as of now'. The world keeps on changing - those pesky engineers keep inventing stuff!


Monday, February 26, 2007

"The Question" and Goals and Objectives

Here's The Question from from Neil Gunther, on 'ITILOPIA' that started this thread:

"If ITIL is the equivalent of CMIP (a heavyweight IT Performance data protocol), then what is the equivalent of SNMP (a lightweight protocol)?"

Most of the world gets along "Just Fine" in providing/running their IT environments/operations without ITIL.
Why would they want to spend $10,000-$20,000 per fully trained individual for something that may or may not be worth it? That will take 1-5 years to implement and cost $100,000's in product licences, consultant fees and setup costs?

And something that will have a serious downside - the unpleastantness of 'formalism'.
Something that will give "I.T." the ability to push-back on management, and take away one of their favourite whipping-boys...

And it fails the What's In It for Me test of management self-interest.
What manager, opposed to owner, is going to sacrifice a whole bunch of money and resources on something where the payoff is unknown and the consequences unpleasant (giving to I.T. the power of saying "NO" and making it stick)?

I assert that no politically savvy manager will back that horse... It can only be "A Career Limiting Move", not beneficial to them.

If that sounds cynical, is it really?
What do you want out of a manager?
Bright, aware and capable...
The sort of person who knows what's bad for them, the company, the customers/users and the employees - most likely this is the order they care about things...
Why would you not expect managers to look after their own interests first and foremost??
You'd have to be crazy, "brave" or idealistic to put yourself last... And it's unlikely that's a person you want as a manager.

So The Question stands:
What's a Politically acceptable system for managing I.T. Services/Operations that serves the personal interests I.T. manager and the business managers/owners?

Goals and Objectives

  • Describe the problem.
  • Outline the parameters of a solution.
  • Propose some elements of a solution.
  • Ideas for how to arrive at a published Standard.