[OSGeo-Discuss] Re: Whitebox GAT (Chris Puttick)

[Brian Russo]

On Fri, Mar 26, 2010 at 11:54 AM, Chris Puttick
<chris.puttick at thehumanjourney.net> wrote:
> Terribly off-topic now, so feel free to stop reading...

Yes.. if anyone wants to ping me offline about this feel free..

> ...not realising high or often any business value. Business value is where what you expend money and get more in return than you spent. Incredibly easy to measure in small businesses with few employees and a simple business model, harder the larger the business or the more complex the concept of value becomes e.g. in a charity or government organisation. There is good evidence that collectively western economies have spent more on IT than they have realised in value.

I 100% agree that most IT procurement is terrible. People go after
'shiney' technology that solves an immediate perceived requirement but
do not go through the more expensive (in the short term) work of
really assessing how their IT infrastructure is actually
enabling/supporting their business processes.

However this has absolutely nothing whatsoever to do with how the
software is licensed. You can make similarly horrible decisions using
open source software… proprietary... whatever. It doesn’t matter.
Remember all the crappy linux based phones out there? They sucked
until we got Android ones. Companies would have gotten better value
using blackberries or something before that time The true reason
people end up in that situation is because the technology they bought
isn’t supporting their business properly. It’s like buying a gym
membership you never use. Does that mean the gym sucks? It might, but
all it really means is that you're not getting value out of the cost
you expended. It doesn't tell you why.

> The business case is not simple, any more than it is in marketing; but here's my base position in simple terms. I select solutions that maximise our future choices and reduce our costs; a further benefit is derived if I can move any remaining costs from fixed annual overhead to per employee or pure capital; while there may be short term pain as people get used to the changes, any increase in costs for that short period will be more than offset by the long term decrease in costs and increases in flexibility for the organisation.

This is where I disagree with you. If you focus on cost as the thing
to reduce you will more often than not lose. Lowering cost should be
an incidental outcome that happens as a result of increasing value and
efficiency. It's quite possible to end up spending more money on IT
than you were in the first place (more frequently you end up spending
it in the right places instead of the wrong and net overall IT
savings) - but if your overall business value has increased more or
commensurately then spending more is probably the right outcome.

> Luckily for me I don't have to justify to others other than in my long term results. I'm aware that this continues to be a rare privilege for the top of the information systems tree and that many organisations continue to not have technical expertise at the highest level, resulting in many decisions in that area being taken with the wrong information and wrong motivations. I'm working on that too.

I would instead argue that the main problem is a lack of
differentiation between CIOs and CTOs. Most organisations involved in
IT are still primarily technology-driven in terms of their procurement
- rather than remembering that their IT is only a means to an end
(supporting business processes & content).

Running a cheap subversion + trac server on Linux is better than
spending $10k on some proprietary software with more substantial
hardware requirements. However an even better solution may be to
outsource it to someone else and pay them $20/month to manage it all.
As long as they support my business process (managing code) and I have
no other reasons to insist on a specific platform I may not really
care. Shifting host providers down the road isn't really a big deal so
even if they're basing it off proprietary software that's transparent
to me. I have my cake and eat it - I pay someone $3/month to host all
that stuff for me using subversion + other tools.

Look at it this way - do you care if your ethernet switches run
proprietary firmware? There's fundamentally no difference between that
and an operating system if I abstract my processes sufficiently. Means
to an end. I realize the real world is a bit muddier, but the point is
to focus on what's important - your information and process continuity
- NOT the system it 'happens' to be running on now. Assume that will
change and plan accordingly. So when you say that you think .Net is a
bad platform to base systems on - I understand your reasons elicited -
but you're acting like a CTO, not a CIO.


-- 
Brian Russo / (808) 271 4166