Month: August 2007
CTOs as Interpreter Translators
I believe Chief Technology Officers (CTOs) should be good at IT (Interpreting and Translating). We should interpret and translate between technical experts and their languages and users and their languages.
We shouldn’t expect every technical person to understand the full elegance of the broad missions we support, but they must grasp the essentials or their tech work will not be appropriately aligned. And non technical leaders and users cannot be expected to grasp the full elegance and beauty of technologies. But it sure helps when they they understand some of the truth about how things work.
So how do you express Service Oriented Architecture (SOA) and Web Service concepts to your users? And what do decision-makers need to know about this area of technology?
To help further the dialog I’ve started building my english language reference to technical concepts, located here. Please take a look and feel free to use it. If you have questions/comments/suggestions for how I express things there please let me know.
Integrating Web 2.0 into existing IT infrastructures
James Connolly just wrote a nice piece on the challenges and rewards of Web 2.0 integration (see "Web 2.0 Integration…" ). James interviewed me for the piece and I found him to not only be very professional but to have a really good knowledge of these new technologies. He was also skilled at asking the right questions to ensure he captured the essence of what was going on.
I think some of what he pulled from me is instructive and I quote what he quotes me as saying below (I guess this is a nested/recursive quote?):
One IT executive who has assessed the impact of Web 2.0 applications on his organization’s technology infrastructure is Robert D. Gourley, chief technology officer for the Defense Intelligence Agency (DIA), an arm of the U.S. Department of Defense.
DIA has utilized blogs, wikis and other Web 2.0 features for several years. Recently, the agency rolled out a mashup application that allows users to meld information from a variety of sources, such as RSS feeds, blogs, mapping tools and traditional databases. The key to the application, Gourley said, is its user interface developed using JackBe Corp.’s Presto, which allows users to select data from any source and interact with it in new ways.
Web 2.0 success
But ongoing monitoring is just one of the final steps in a long process. Gourley shared his suggestions for successfully integrating Web 2.0 into existing IT infrastructures:
- Governance: A strong governance process that puts the leadership in charge of 100% of the IT activities in the organization is essential, he said. "Components of strong governance include a vision that is well-articulated and understood, a strategy for how to execute that vision, and then guidelines that establish the SOA. Then you can start to field Web 2.0 applications."
- Test: And then test again. "We test and test and test, and then we have an independent body test it," Gourley said. Understand that these are new types of applications, and developing test criteria is an iterative process where the feedback from users helps to define the actual test.
- Model and monitor: Proper modeling is needed to ensure that users get adequate bandwidth to use Web 2.0 applications and services. "It’s the same with server loads. Our program management team begins by estimating them. We model that and test it, and even now, as we are fielding this, we are watching very carefully to make sure we’ve judged the server loads correctly."
- User feedback: Develop a system that allows users to report on their activities. This can take any of several forms. "There’s no single answer," Gourley said. "We have a single trouble-ticket system, for example. That is associated with a single FAQ system."
- Recognize the differences: It is important for IT pros to recognize that Web 2.0 applications are of a different breed than what they’re used to. Web 2.0 "is so new and so different, it’s the kind of thing that requires senior management’s attention because not everyone is going to understand it," Gourley said.
I’d like to expand on this first point a bit. The idea of governance applies to all well managed/well led human activities, not just IT and not just Web 2.0 projects. Governance ensures that what must get done gets done and if there is some reason why it isn’t going to get done then senior management and all effected parties get visibility into that right away so they can take action. Goverance translates vision into strategy and translates strategy into actions and ensures that all individuals are goaled to meet the actions required of the organization.
Without good governance, the only things that get done are the things that individuals in the enterprise want to get done. That can get very chaotic and is generally not going to be very productive.
Jonathan Schwartz’s views on CTOs
Of all Jonathan’s posts, all of which I really enjoy, one recently one struck me as particularly insightful:
Link: Jonathan Schwartz’s Weblog.
"I found myself talking to a group of media company CEO’s. I asked a simple question, "do you have a general counsel reporting to you?" The answer was universally, yes.
I do, too. Mike and his team are central to the evolution of Sun (as I’ve said, we are nothing less, or more, than an intellectual property company – it’s hard navigating those waters without a great legal team).
But then I asked a harder question: "Do you have a chief technology officer reporting to you?"
I do, and I talk to Greg at least every day. He plays a central role at Sun. Central as in nervous system. He’s involved in every major strategic decision I make (and a ton of minor ones, too).
But in repsonse to my question, the answers from the group were more dismissive than substantive – most did not. And in my view, if you have a general counsel reporting to you, and not a CTO, you’re saying legal advice is more important to you than technology counsel. Which seems backward for a media company. Why?"
Thanks Jonathan, for spelling out so clearly why the CTO role is becoming more and more critical to the future of the enterprise and to society overall. You also underscore for all of us why Greg must have the greatest CTO gig on earth. He has a great boss.
Requirements and Soda Pop
Two weeks ago a peer from another part of our extended enterprise told me he didn’t understand why enterprise IT folks are always so insistent on having customers express requirements vice just naming the solution they want.
He said it should be like when he goes to Safeway to order some soda pop. He doesn’t go in and express a requirement for a cold diet beverage with bubbles and a fruit flavor. He goes and grabs a six pack of what he wants. He is smart enough to know what to pick and doesn’t need someone else saying he can’t do that just because he didn’t express the requirement right.
So I told him there were some good parts to that analogy but frankly it breaks down pretty quick. For example, soda is relatively inexpensive. The amount he would spend on a six pack of soda is really just a rounding error. IT is very expensive. Additionally, no one cares what he puts in his belly and really few of us would care if he gets a belly ache because of it. But the enterprise must remain healthy and we technologists and our leaders must care what gets put into it. Therefore, I explained, only IT professionals would be picking what technology gets into the enterprise, and others could recommend and state functional requirements, but we would be making all the decisions.
Well, it is hard to convince anyone that they don’t get to make the decisions, but if you want to run a successful enterprise sometimes you have to do that. A key to success in doing this is to always have the customer and their mission needs at the forefront of the decision and to always involve them early on. Another key to success here is to be able to demonstrate a track record of delivering real capabilities that meet the mission needs of the organization. Nothing succeeds like success.
The Future of Technology
I have a theory about the future. Although it is impossible to predict the future of technology with any real precision (every expert has failed greatly), we can apply a focused methodology to extract lessons from the future that are relevant to today’s planning.
My favored methodology is to start with an understanding of how today’s technology works plus an understanding of the mega trends sweeping the IT industry and then add in the key mission drivers pushing the growth of technologies. The result is one view into the future.
I then like to add in linear extrapolations of all key technology categories. This is a very dangerous, flawed way of looking at the future because the future is never linear but it is one of the only tools available so it must be used.
The next step in my methodology is to think through all the above to produce views into the future of technology. Over time I’ll present and update my views on the future of technology in this blog under the category "The Future of Technology."
Top 10 CTO Principles
Consider a principle as a basic generalization that is accepted as true and that can be used as a basis for reasoning or conduct.
I believe there are basic generalizations accepted by all leading CTOs and therefore these CTO principles can be studied and learned. I have my own list that I’ve applied at several organizations and will expand on them in future posts. But my real point here is to underscore that by study of the principles of CTOs we can look for deeper insights into our profession.
For my enterprise, the principles I’ve always used are:
Our systems and architectures will always be mission focused
Our enterprise must be flexible and adaptable
Our technical team will constantly upgrade their skills and knowledge
Our technology plan will be broadly communicated/understood
We unify enterprise architecture, governance and engineering efforts
We embrace the ideal of a single enterprise-wide SOA
We embrace and embody the concept of "mission assurance"
We will divest ourselves of efforts which do not support the mission
We err to the simple over complex
Disciplined collaboration is critical to our success
I’ll expand on each of those over time.
Technology Relevance
No matter who the CTO is, or what style CTO they are, or what organization they are in, there are a few things we all seem to have in common. All of us are relied on to provide assessments on the relevance of technology to business and mission functions.
For me, the most significant inputs I’ve had to decisions made in my organization have generally been on items requiring a very high degree of mission understanding, as well as a deep understanding of the technology (including legacy, current and future technologies).
Which leads to the following opinion. The best highest calling of a CTO is to provide technology relevance assessments.
The Great CTOs
I’ve created a category called "Great CTOs" so I can better categorize some of the hero’s I have in this business. Many of the great CTOs prefer to operate without a lot of fanfair so there are those I would like to write about who I won’t. Instead I’ll write about those hero CTOs who blog or who are in very visible positions. That way I won’t be violating anyone’s privacy desires. Look for more in this category.
CTOvision Blog 