A little over a week ago I put out a request to my readers to help me with my meeting with Vivek Kundra.  The response has been awesome!Thank you for your suggestions and recommendations, both public and private. Below, in purely random order, are just some of the comments I've received. If you want to add your voice to the choir, I'll be taking comments through Tuesday evening, May 26th.  During my meeting, I will definitely highlight this dialog and will also leave a listing of all comments for Mr. Kundra's review.

-----------------------------------------------------------------------------------------------

Let me just suggest a little thing I'd like Vivek to keep in mind as a guiding principle for open government: Choice.

What do I mean? Government agencies, as they modernize to approach the agility and performance of great commercial IT shops, will be looking to cloud-solution (and other) providers for "business value and choice in a mixed-source world," in the words of Microsoft's Teresa Carlson. The agency CIOs and CTOs whom Vivek leads need the flexibility to choose the best tools to accomplish their mission regardless of platform. I'd hate to see a "one-size-fits-all" prescribed solution (even if it were based on Microsoft products!), and I think we'll all benefit from a balanced approach with appropriate choice and flexibility reserved for smart CIOs.

I think in order for the govt to adopt such a broad technology there needs to be standards set for interoperability and security first. We all know that standards boards tend to lag behind real world implementations. But in this case the consequences of not going forward with agreed upon security and interop standards is very dangerous.

What has worked with other technologies is lighting a torch underneath the vendors to get together and provide a set of standards before any govt contracts are let. Sometimes providing that carrot (big carrot) is essential to get everyone in the same room. Good luck with your meeting, very cool!

I believe a frank discussion on open source initiatives, with respect to Grid Computing, will be productive. For example, the Open Grid Services Architecture (OGSA), a product of the Grid community at large, represents an evolution towards a Grid system architecture based on Web services concepts and technologies. Furthermore, Data Access and Integration Services (DAIS) is another emerging standard (e.g. WS-DAIR, WS-DAIX) to access and integrate data, within a grid environment, that could drive a unified federated data architecture. All have limited implementations, but are "shovel ready."

• Does he plan to provide additional mandates related to IPv6 - will he take the next step?
  
• Is there a Federal Wide Cloud computing strategy to optimize and outsource government run IT shops.
  
• Does he plan to implement BAA compliance in addition to TAA compliance for IT monies in the stimulus package? (at the fed level)
  
• Any plans to align government with industry best practices to improve service delivery to the public? Agencies should be creating a roadmap for the network infrastructure that will allow them to rapidly deploy new applications to improve business processes and provide quality public services.
  
• Does each agency have a network infrastructure that is capable of handling the applications needed to improve operational efficiency anywhere, anytime, anyplace in support of the mission? This is job number one, every agency must have an agile mission fabric, none of the other things that follow will work well without it.
  
• Who is best to provide oversight, implementation, and ongoing maintenance for a given application? We should always be asking the WHO question, modeling industry, and avoiding business as usual.
  

From the UK there are some open source business processes designed and operating between citizen and government. The main issue with most of our ICT in government is that hitherto procurement has been proprietary ( Oracle ; Microsoft ; EDS ), expensive and completely un-scalable. Each agency has bought its own set.

• A NGO called mysociety did most of the open source, rich content and user-friendly work http://www.mysociety.org/
  
• There is a petitions site to the Prime Minister which has had over 9 million votes
  
• There is a fantastic process to write to your elected representative ( local, state , federal ) http://www.writetothem.com/
  
• There is a method to report any issue to your local government ( poor street lighting ; graffiti etc.) http://www.fixmystreet.com/
  

These are all shovel ready and would be immediately transferable to the USA

The other process that you might wish to replicate, or suggest, involves re-coding government data. It was called the Re-Wired State, and a list of projects is here - http://rewiredstate.org/projects

The projects listed demonstrate the intent. How successful you can be will depend on how much data the government lets go.

Other suggestions - facebook for jobs ( after all it is just a market with bids and offers ) ;

You won't manage to have comparable financials across agencies until you have a standardised General Ledger ( something the health service in UK has battled with since 1948 )

Sun's star office suite never took off ; in the UK not many people use pbwiki in government but it seems really successful and ning we use a lot. We are looking at how mental health service users might create ning groups to exchange content, information and collaborate. Their carers or families could also be involved, as eventually might employers and state agencies. To start with, it is the service users only as they are the most motivated to help one another.

If only the problems here were technology problems. This model has been called a lot of different things over the last decade, but it always seems to fail at SLAs, security, and the unwillingness of leadership to give up their own IT infrastructures. If the PM office tackles just the first two, it will be major progress but still a little short of the change that needs to happen.

Now might be just the right time for a mandate - the business case here is off the charts.

I don't understand why they can't arrange the data to "scatter" across the cloud. If someone hacked into one node, they would only get a "piece" of the information. You would have to be given admin permissions to accurately "piece" the whole picture together. We are doing this with the public health grid at the NCPHI lab at CDC. Seems pretty safe to me.

I'd suggest you take a look at what DISA has done with the RACE (Rapid Access Computing Environment) effort they are bringing online, as well as its companion application development environment (http://disa.mil/forge).

While these are specific examples, I think you'll find interesting information to help you in your meeting with Mr. Kundra. DISA has already been thinking a lot about this already.

[Does Mr. Kundra think] the goal should be to build and adopt new technological standards to gain efficiency/effectiveness or to amend current infrastructure and rewrite policies so such technologies have an easier chance of adoption.

[W]here does he stand on the notion of a "one web" solution for the entire executive branch, i.e. instead of individual departments acting like silos with their own social technologies, shouldn't there be an Administration-wide direction and initiative in one?

[H]ow about working on giving more support to adoption of collaborative and social media tools within the agencies - encouraging the technical, legal and cultural issues that agencies need to get over.

I would second Ms. [Deb] Lavoy, and add what to me is almost embarrassingly basic: an immediate solution to the ban on thumb drives. Perhaps the ban itself can't be resolved that quickly, but surely there is some way to train and trust people who truly need this technology to do their jobs every day. We in Public Affairs are hampered by the inability to move photos from camera to computer, and I have read about med techs with handheld computers in the field whose issue is not just job-critical but life-critical to their patients.

Driving innovation and standardization with a view to improving efficiency is a lofty goal. But change is painful- especially for the folks that have to practice this in their work everyday. Compliance can be enforced, but in order to ensure that folks understand the strategy and are committed to realizing the efficiencies, the human component has to be taken into account. It is challenging to change a culture where every agency/department is used to following their own standards to have to now start adhering to a new centralized standard. Pilot initiatives have to go beyond the notion of "one size fits all". Will there be plans to make sure that the resources that have to use this are on board?

There is more to securing "cloud computing" than a bunch of policies and directives. What is your disaster recovery and continuity plan for when the cloud service suffers a failure? Since the data lives in the "cloud", can you be assured that you know "where" it it? I guess the answers depend on your definition of "cloud".