Visualizing Progress

Data visualization gets (from me) a bad rap, because so often it is used as a way to make fundamentally unsexy things just sexy enough for management to buy, without reference to whether the item in question is, at core, useful or effective.

Caveat aside, this graph from the GeoTools JIRA side made my hair stand on end.

Yikes, do you close more tickets than get opened? There’s a visualization that lights a fire under my butt!

Even when I’m in full ticket-hunting-mode I feel like I sometimes am not making forward progress, since Regina gets into bug-finding-mode and starts opening them up on me faster. And the last year of adding features to PostGIS (raster, topology) while altering core structures (serialization, indexes) has created an avalanche of tickets. More than fits on a page. Time to get closing!

To the Cloud!

David Pogue has a little mini-rant today on the use of “cloud” as a generic term to mean “internet” which I found fun, if only because it reminded me of the odd origins of the term.

As far as I can guess, the term leaked out of white-board shorthand used by folks drawing up high-level architecture diagrams. Who it was who popularized that diagram convention is a mystery, probably it is an elaboration on some even earlier trope. Now, thanks to Microsoft marketing, the term means… nothing? everything? kung fu music? computers-can-magically-solve-anything?

IT as Obstacle

Getting the re-tweeting rounds today, an article on why business are moving to the cloud. The answer? To avoid their own IT departments.

Their top reasons for going around IT? The need to respond quickly to changes in the market, self-sufficiency of their IT-savvy workforce, and the easy availability of top-quality it services that can be bought without long implementation or testing (cloud and SAAs apps, primarily).

Watch out, the natives are restless.

Update: Check out the comment thread in the posting of this article for the good, bad and ugly of IT reaction to being sidelined. Ranges from “stupid users” to some thoughtful comments on the nature of organizations.

Outsourcing in BC

Every year, in a fit of morbid curiousity, I cruise the detailed schedule of payments in the BC Public Accounts as a way of keeping tabs on the local IT industry. Who is up? Who is down? Who is gone?

One of the certainties, year after year, is that IBM’s take just keeps getting bigger. I thought it was pretty impressive when they were doing $30M / year, but last year they cracked $100M.

Another trend has been the slow squeeze of the market place from one with a mix of companies, ranging from 10-person consultancies to 100-person regional integrators to the big international firms, to a market dominated almost completely by the internationals. The mid-size companies were largely bought up, and the small ones squeezed out. Deal sizes have kept growing, making it hard for smaller companies to compete, while the migration of infrastructure roles to outsourcing companies has given those same companies an built-in edge in competing for systems work. Even long-time regional heavy-weight Sierra Systems is getting beaten down.

There are many big mistakes being made by government as this process goes on, but perhaps the biggest is the long-term sole-source outsourcing agreement. The fantasy that costs can be managed in a single vendor environment, that the contract just needs the right review clauses and penalties, continues to suck in government outsourcers. There’s no true competition in a system where a vendor only needs to compete every 5-10 years (and even then, from the position of the incumbent). Expecting market efficiencies to work in your favour in such as situation is folly.

Postscript: In 2009/2010, HP and IBM combined to bill the BC government over $200M, mostly in services. IBM’s 2010 annual report indicates 34.7% gross margin for IBM Global Services. The HP annual report shows a similar margin for services. The implication is, of the $200M we are spending with HP and IBM, $70M is going to profit, not to providing service. Put another way, if the government chose to provide those services internally, it could pay all the staff at HP/IBM salary rates and still afford to employ 50% more people to do the work.

Superstar CIOs

The superstar CIO is necessarily going to be a bit of a flimflam artist. He has to be, because in order to get into the position, he’s had to dazzle his non-technical superiors. That means eye-candy, first and foremost, delivered fast. Delivered fast, you say, how can he do that with his sclerotic organization? Easy, he contracts off to the side or builds a private skunk-works.

So, from Vivek Kundra (US Federal CIO) on the one hand we have the flimflam of, and various high level dashboard trackers ( anyone?), tossed up quickly by a tiger team. Not indicative of transformation, but rather the opposite: in order to get anything completed quickly the CIO has to actually route around his existing organization. But there is also some substance.

Some of the substance is pure management, like the review of failing projects. Some is technological silver-bullet-worship (of a sort I can appreciate, being a long-time silver bullet lover) like the move to government use of cloud infrastructure (both external and self-managed). (Because the answer is always another layer of abstraction!)

But one thing comes through clearly watching Mr. Kundra: it’s a really big organization and he’s just one guy. The culture will out-live him. The men in grey suits will win. In some ways a full-on flimflam artist would be better, because at least then I wouldn’t be disappointed when nothing changes.

Postscript: Kundra did manage to change some things but many of his marquee initiatives were separately funded initiatives that Congress is now de-funding for political reasons. And the news today is that he’s leaving his post for academe to take a well-earned break. Did he push the culture in the right direction? Probably so!