WhiteHouse.gov now runs Drupal. What took it so long?
There's a lot of buzz today on the Obama Administration's decision to run WhiteHouse.gov on Drupal, the popular open-source Web publishing system. Given the U.S. federal government's widespread adoption of open source, however, the amazing thing is that it took so long.
It is now, anyway. This is a big shift in the federal attitude toward open source.
Back in 2004, I worked in the Linux Business Office at Novell and met with the CIO for the U.S. Senate. He knew about open source and admitted that other departments were dipping their toes in it, but the Senate was a decidedly Microsoft shop and had no plans to change.
While I don't have an update on the U.S. Senate's adoption of open source, the rest of the government seems to have gone far beyond "toe dipping." Open-source adoption throughout the U.S. federal government is rampant, and started long before President Obama took office.
NASA's Nebula platform is just one example of how the government is actively using open-source technologies (like Apache SOLR, RabbitMQ, MySQL, and Eucalyptus), but also contributing back.
Unlike other geographies, which have relied on government mandates and preferences to accelerate open-source uptake, the U.S. government hasn't been the driving force in open-source adoption, or even the primary force. U.S. public and private sectors have been equally enthusiastic about open source.
For example, it's nice that WhiteHouse.gov runs Drupal, but adoption in the private sector by FedEx, Sony Music, and many others precedes President Obama's choice of Drupal.
The open-source train has left the building. Even companies like Qualcomm, the patent powerhouse that has traditionally disdained open source, are making open source a core business strategy. Qualcomm is setting up a subsidiary to focus on mobile open-source platforms.
Yes, pigs can fly.
What's driving this adoption? It's not necessarily open source's price tag. After all, in the short term, open source isn't necessarily less expensive, once you factor in migration costs, retraining, etc. (Note: you'd hit these same costs even if you moved to a different proprietary system.)
Of course, proprietary software is no bargain-basement cost saver, either. Even Windows 7, that no-brainer IT decision in the wake of Vista's pain, could cost enterprises as much as $1,930 per instance, according to Gartner.
No, in my experience, open source is winning converts because it gives CIOs more control of their destiny.
In part such control stems from the nature of competition itself. As open source proves itself a viable contender for CIO dollars and thereby spark price competition, CIOs save, as Novell CMO John Dragoon notes.
But open source's superior value proposition goes deeper. Open source calls into question the highly profitable maintenance fees that Oracle, SAP, and traditional software vendors use to juice their earnings, but which do little to help customers.
In fact, the traditional software economy can be downright hostile to buyers, as Ingres CEO Roger Burkhardt opines.
Open source is different. Because the code is open, open-source vendors are forced to deliver a constant stream of value to justify subscription renewals. ZDNet's Dana Blankenhorn captures this well:
When you can see the code you have a different relationship with it. You're no longer asking what it can do. You're asking how you can adapt it to your needs.
With code visibility, you and your vendors become partners in trying to make something work. The vendor can't over-promise, but you can't over-assume either. This may be one of main hidden reasons for IT failure, the two sides of the transaction not being on the same page.
It's not surprising, therefore, that Red Hat continues to be the CIO's darling for lowering costs and delivering value. It's also not surprising that the Obama Administration adopted Drupal for WhiteHouse.gov.
No, what is surprising is that it took so long.
Matt Asay brings a decade of in-the-trenches open-source business and legal experience to The Open Road, with an emphasis on emerging open-source business strategies and opportunities. Matt is vice president of business development at Alfresco, a company that develops open-source software for content management. He is a member of the CNET Blog Network and is not an employee of CNET. Disclosure. You can follow Matt on Twitter @mjasay. 
Links to Matt's own argument that focusing on the openness is moronic because CIOs don't care then he follows with:
"Open source is different. Because the code is open, open-source vendors are forced to deliver a constant stream of value to justify subscription renewals."
Personally I've been saying that for years in direct opposition to people like Matt but, who am I? Just a low level grunt to Matt's senior position in an open source company. It does seem to me, though, that he's contradicting himself directly just weeks after claiming the exact opposite to that which he's claiming now.
Matt? Are you feeling OK? You're not having some kind of nervious breakdown are you?
I mean it's great even that so soon after stating that "if Red Hat's pure-play open source strategy was so great everyone would be doing it" you're singing their praises again in what could be taken (or not) as acknowledgement that it's exactly this strategy that has brought them so much success but your audience could be forgiven for wondering if you don't have Multiple Personality Disorder.
I still disagree with one thing though. If OpenOffice.org has demonstrated one thing (I mean, apart from the need for section breaks and individually landscaped pages) in its attempts to work with MS Office documents it's that transition from a proprietary system can not only be costly but impossible.
If you're moving from one system then you may in the future want to move from the system you're currently moving to (you can't know that this will never be the case) so in the scenario of moving from a proprietary system then another proprietary system the cost may be great both times because of the similar level of difficulty but moving from a proprietary system then from an open system will inevitably cost less because the data format for the second move will be entirely transparent and well known.
All of this would be moot if we could all start from open systems but that's a rare luxury if ever and I'd hope a product would have to be damned compelling before someone went from open to proprietary because it's possible the cost could be similar to a move in the inverse direction (a la Microsoft's laughable ODF import plugin) and the cost of moving away will certainly be high.
This is the cost a lot of people leave out in their calculations for system transfer.
My recent fave was when he slammed opengoo for using the same deploy model as Alfresco and saying he couldn't find the launch button and later saying he knew it was a web server app all along.
- by Chris_Maresca October 26, 2009 6:32 PM PDT
- Hhmm, I deployed a ton of open source in the US gov't in 1995. In fact, the first DHHS intranet was built on linux, apache and perl. So were parts of the DoC global email system in 1996 or so (70k employees, 1300 agencies and 130+ countries...).
- Like this Reply to this comment
-
(6 Comments)Besides, government involvement first started with software like GRASS, released in the early 1980's and plenty of other software has been built by the US gov't and released as source. They even released a real-time OS for missile guidance (RTEMS).
So, not new and pigs have been flying for the better part of 30 years....