This is Ken Carroll's Typepad Profile.
Join Typepad and start following Ken Carroll's activity
Join Now!
Already a member? Sign In
Ken Carroll
Recent Activity
I forgot to add this to my post above on the 13X cost increase story. Not only did the development cost go up by an order of magnitude (i.e. 13X), but the operational and maintenance costs jumped as well. The final version had at least twice as much code, was harder to enhance, required four teams for maintenance and operation rather than two, and introduced more moving parts into production. The team involved was forced to do this by ruthless application of mandatory standards which had been put in place by inadequately experienced people who had no knowledge the the quite important requirements & context of this important application. It is but one of many real examples.
Toggle Commented Mar 18, 2012 on Leading An IT Transformation at Chuck's Blog
Having experienced and fought against this issue on many occassions, I think you have absolutely hit the nail on the head with this post. I believe you have also pinpointed the solution, which is for IT companies to forge relationships with business management. The challenge is that IT people are naturally interested in developing IT things - it is FAR more interesting and self-satisfying to create your own framework for X than it is to simply use one (commercial or open source). However, if your employer is, as cited in your example above, a financial services company then their real & primary interest is in servicing their customers. Customers don't care if you create the circuit boards, write the O/S and extrude the CAT5 cables too - all they want is good service. If IT people in a non-IT company spend a lot of time on infrastructure creation & maintenance then they are wasting time as they are not contributing to producing customer perceived value. Not all IT people are like this but unfortunately there are enough smart & capable IT people in non-IT companies who act as if they are working for IT companies and so they concoct justifications for allowing them to create infrastructure solutions. It is very interesting and challenging work, for sure, but absolutely a complete waste of time to do in an non-IT company. What is often forgotten is that creating something, e.g. software, also creates mistakes/bugs and only extensive usage in many different contexts will rustle out those mistakes/bugs in a reasonable timeframe. Creating your own homegrown solution limits the amount of usage and context variations such that mistakes/errors/bugs live on for a long time and result in many outages & problems popping up. Using, instead, widely used commercial or open source solutions is extremely effective in overcoming this issue. Everyone I've encountered who reinvents the wheel has a tendency to turn a blind eye to this and I've seen it cause SEV-1 outages as a consequence. The challenge is that with very smart IT people leading other smart IT people, they somehow seem to get enticed into continuing to do interesting IT things. This tendency permeates up & down the full hierarchy of IT staff, from the top leadership on down. I have seen it repeated so much that it just seems to be a basic human condition - smart IT people have a tendency to want to do 'smart' IT things such as reinvent this & that. The only solution, I think, is for IT vendors to forge strong relationships with the business management to whom IT is accountable. These should be considered peer customers of your IT management in a company. It is difficult to see any other way to ensure that the checks & balances helping enforce accountability are in effect.
Toggle Commented Mar 10, 2012 on The Dark Side Of Clouds at Chuck's Blog
This is a very interesting, well written and on-the-mark post. I can say this from direct & personal experience, one in which we produced considerable success by identifying & addressing the items you've highlighted here. Most of our IT groups didn't recognize how much things had really changed and were not producing critical results during the 2008 market turmoil. Only by fundamentally changing the model of cooperation between business & IT, taking risks with new processes and technologies, focusing on some related critical areas rather than boiling the ocean, working through organizational boundaries, building the right team with the right people including some new, financing it appropriately and over-communicating, did we clear the FUD and have what turned into foundational & transformative success. You have identified all the major hurdles in your article and I have also seen where these were not done leading to failures.
The power of being able to process large amounts of complex data efficiently is critical. I recall so many times working on many CAE projects, in the past, having to run far fewer simulations, and other processes, than really warranted & desired simply due to the inefficiencies involved. In cases that related to medical devices this was always concerning. Such advances today with Big Data are a true marvel. Thanks for this eye-opening article.
Toggle Commented Feb 28, 2012 on The Other Side Of Big Data at Chuck's Blog
Ken Carroll is now following The Typepad Team
Feb 28, 2012