Executive Summary

Columbus Ohio is rated by Neilson Online as the 8^th highest user of social media applications and has been rated as a top technology city in the US. As the center of State government we have an opportunity to broaden the capability of our current human capital while becoming more attractive to a younger generation of public servants. This general familiarity with social media applications and technology means that the time is right for the State of Ohio to seize the moment.

Governor Strickland has appropriately suggested that our economic recover and the State of Ohio’s future must include investment in developing new technologies and better utilizing the technologies available today.

I believe we can begin implementing a 5 point program now which will save the State money, empower both State employees and the general public and create a technological culture that will stimulate growth in the State of Ohio.

• Utilize social applications and multi-media in a professional manor to maximize knowledge transfer, collaborative effort and transparency.
• Make use of linked data as a way to maximize enterprise information sharing.
• Create an Open Data Store for the State of Ohio that allows employees, citizens and organizations direct digital access to public information.
• Design future applications using Open Source standards and include Application Programming Interfaces (APIs) that allow employees, citizens and organizations the ability to create useful mashups.
• Centralize all agency applications and data to a State of Ohio “Cloud” and develop desktop standards that reflect the lesser local system demand.

The key to success in this endeavor is to utilize resources that already exist such as GPL Open Source Software that many State employees already utilize in their personal lives. Another important component of success will be to clearly define the parameters of public data for the sake of protecting citizen and employee identities. Security should not be a barrier to moving in this direction; the data already exists on the network and we are already taxed with due diligence to protect it.

This set of ideas is not bleeding edge and is thus appropriate for consideration by a public entity such as the State of Ohio. In fact, some of these ideas may already be partially implemented and just need to be brought full circle.

A quick review of some of the most useful applications on the internet reveals the success of opening up your data and providing APIs to the general public. Without this we wouldn’t have ended up with resources such as Govloop.com, EveryBlock.com, The Commons, Google Transit Maps, Recovery.gov, etc. Clearly this is the time for the State of Ohio to embrace these ideas.

Point-by-Point

Utilize social applications and multi-media in a professional manor to maximize knowledge transfer, collaborative effort and transparency.

Social applications are loosely defined as applications that allow interaction and two way communications of ideas and thoughts. Some of the most popular social applications are iGoogle, Wikipedia, WordPress, Facebook, Twitter, YouTube and Flickr.

What makes these sites so popular and how can we leverage those ideas for doing business at the State of Ohio?

If we break down what each application actually does for us we can start to understand how the principles can apply to our various business models.

iGoogle: This juggernaut of a company provides extensive social applications but is primarily known for its search, email and news reader functions.

The search function allows for users to promote pages that more correctly match what they were searching for as well as remove results that were erroneous.

Often when searching for documentation of policy, procedure, law, etc. on the State of Ohio website the search results that are returned have little to do with what you hope to locate. This is primarily related to the searcher’s lack of knowledge about State terminology and acronyms. By giving the searcher the tools to teach the search engine their understanding, our search results can improve exponentially.

We currently utilize Google as the provider of our search services so implementation of this would be relatively easy.

I don’t believe that the case could, or should, be made to utilize Gmail however I believe that the iGoogle RSS news reader function is, at least functionally, something important to consider. Utilizing an RSS news reader allows for delivery of important information whether it is keeping up with the latest industry trends or using internal feeds for important stories about a department or agency. In our current environment there is a desire to standardize our email system with MS Outlook. If that direction is followed there are free, powerful add-ons such as Attensa which can provide this functionality inside our current email application.

RSS news readers are only as good as the feeds they deliver. This means that our networks will have to be open enough to navigate blogs and other social media sites that provide RSS subscription. Policy will need to be adjusted across the agencies to create the appropriate balance. Access and usage of the internet is related more to management of employees than it is to locking down network access.

Besides external RSS news feeds agencies should embrace the use of internal RSS feeds. Almost any finalized document can be setup to syndicate via RSS. This could allow management and clients (when appropriate) instant notification of published documents or other media. Digital notifications could go out this way instead of sending large emails, which reduces system resource strain.

Wikipedia: This application is best known for its role as a user generated encyclopedia but it is the foundation of the application that is most useful to the State. A Wiki is free software that allows multiple users to collaborate on document creation without the same sort of ownership issues caused by applications such as MS SharePoint. Wiki’s are developed around standards such as XML/RSS which allows for automatic notification of changes and completion of projects.

Wiki software can allow various levels of collaboration including allowing the general public to provide comments during project development in the cases that this is appropriate. Not only would that provide the State with a diversity of input but it could also provide the State expertise that it otherwise wouldn’t have had access to. This is similar to President Obama’s current plan of allowing for public feedback on bills which are not considered emergency legislation.

Wiki software could also act as a repository for best practice information that currently exists only in the minds of employees or perhaps in emails or standalone documents. The linkable data within a Wiki would make this information significantly more accessible than in those other mediums.

WordPress: WordPress is one of the most popular blog platforms; it is Open Source and free to download. The blog application however is more important an idea than the specific WordPress product. Blogs offer a very simple interface for sharing information on any range of topics. The blog posts can instantly feed via RSS making them another good option for immediate reporting. Blogs are also a great way of creating linkable data.

Currently information which may be important to multiple individuals, such as lessons learned, project status or even weekly status reports are trapped in standalone systems or provided in a closed format.

Providing narrative type information via a blog post can create searchable, linkable information and history which becomes a resource for the whole organization. If an individual has downtime during a week they can spend that time learning more about best practices, etc. which they can then share on their blog status report. This offers the opportunity to provide knowledge to everyone subscribed to their RSS news feed.

Facebook: The actual Facebook application may be a source of outreach for certain State agencies or departments and that may include developing applications to further our understanding of citizens needs. More important than the Facebook brand however is the community and networking that comes from such an application.

Ning is a social networking platform similar to Facebook which can be built for any community of interest desired. Similar to Facebook, a Ning allows for RSS feeds to come into it so that you do not have to replicate work you have already done in say a Wiki or a blog. The power of these communities is the ability to see what a colleague is working on or questioning and then contributing to that effort in either a small or large way.

The nature of these community applications is growing your network by becoming a “Friend of a Friend”. In the business environment this would be an expansion of your professional network based on similar work roles, shared projects and other collaboration opportunities.

As your network grows you will find greater opportunity to serve the State in ways that would otherwise not have been considered due to the current status of silo type work environments.

Twitter: Twitter itself is a powerful micro-blogging community which allows for a great deal of knowledge transfer in short, easy to digest bursts. This micro-blogging type of application is available for internal only communications as well and it works with RSS to provide efficient communication across multiple social networks.

Consider using micro-blogging as a way of updating internal workgroups and clients about immediately useful information or as a way to poll others internally for ideas without needing to formally meet.

YouTube and Flickr: There are several other social media applications along these lines; not the least of which are Utterli (primarily an audio based application) and Slideshare (a slide presentation application) which can all be defined as multi-media based social networks. The opportunity for the State of Ohio in these types of applications are many fold but mostly center around training issues or delivery of ancillary data to the social media applications listed above.

Beyond the ease of outreach of social media the single biggest advantage is the ability to communicate the same message in any number of different mediums. Looking at how news media is adopting to the online environment you see traditional newspapers such as the NY Times providing an enormous amount of video, audio and slide-show content while television and radio broadcasters are providing transcripts along side their recorded media.

The State now has the opportunity to provide low-budget multi-media laced information to internal and external users who internalize information in different ways (i.e. Visual, Aural, and Spatial learners).

Make use of linked data as a way to maximize enterprise information sharing.

The idea of linked data is to use semantic tools to create a more useful set of search options for your data. Linked data includes the actual hyper-links that a user determines are appropriate to clarify or support a document but also information such as tags and other meta-data which makes a document, picture, video or other form of data more understandable to search engines and machines.

As I pointed out in my examples of social media applications above they can all link between each other, feed information via RSS, and most utilize a standard format such as XML. The power of providing linked data is expected to grow over the coming years as users become more adept at providing appropriate meta-data and machines become better at understanding users.

To maintain some sense of order to an otherwise chaotic world of data the adoption of policies which encourage the creation of meta-data are essential. The benefits of easily retrievable information makes this well worth the time investment on the front end.

Create an Open Data Store for the State of Ohio that allows employees, citizens and organizations direct access to public information.

Organizations that create massive amounts of data are creating data stores that allow others access to the data in a scrubbed, but otherwise raw format. The theory behind this is that if someone can have access to your data they can quickly and easily reuse that data to create a remix or mashup that is useful to them.

Most individuals and small organizations do not have the resources to build the kind of databases that large organizations have built. Similarly, large organizations often miss the opportunity to build small, niche query applications because it is not a part of their mission.

The social media application EveryBlock.com exploits data from thousands of resources to build hyper-local profiles useful to citizens. Without that data being made freely available this extremely useful remix and mashup would never exist.

Internal State staff as well as external individuals and organizations would be able to utilize data across all State resources to create new, meaningful applications and reports that otherwise would be impossible to deliver.

An Open Data Store would include access to finalized documents coming through the social media applications discussed above as well as data from legacy agency applications. This availability would significantly cut down on Freedom of Information Act requests as the requester would be able to simply search and link to the data themselves.

Design future applications using Open Source Software standards and include Application Programming Interfaces (APIs) that allow employees, citizens and organizations the ability to create useful mashups.

Due to growing concerns about transparency organizations such as the State of Ohio have an opportunity to be proactive by creating software based on Open Source standards. The adoption of such standards under the broader goals of the Free Culture movement would then show a proactive stance in regards to how transparently government functions over all.

In general the primary cost benefit of developing applications under Open Source standards would be to provide greater competition between vendors for development, maintenance and upgrade work. When combined with providing APIs this greatly increases a user’s ability to enhance an application without directly impacting the core system.

Similar to opening up data stores, by providing APIs many users will be able to provide functionality useful to them that otherwise wouldn’t be affordable or may have not even been imagined. The additional functionality would then become useable by the State or any other entity.

As discussed above, by opening up data across the board, including access to process and procedure documentation, both internal and external clients can become more effective in how they utilize the resources and services the State provides.

Centralize all agency applications and data to a State of Ohio “Cloud” and develop desktop standards that reflect the lesser local system demand.

While a significant number of applications already exist at the SOCC most State agencies still host large client/server environments, silo email systems, expensive desktop machines and software, support staff across all layers of the agencies and the resulting diversity of procurements necessary to maintain everything.

By centralizing the applications and data significant cost savings could be realized through the support of fewer computing centers. Reducing the diversity of hardware and software on the server side would also provide greater uniformity to applications; making the creation of an Open Data Store and APIs easier.

Moving all applications and data to the cloud will provide the State with greater control over enterprise standards and procurements. This will allow for greater cost savings, particularly when purchasing IT commodities.

Following the lead of several foreign governments and international corporations the State of Ohio could shift desktop computing away from proprietary software and towards OSS. Linux and Open Office are both extremely user friendly, powerful and interoperate well with Microsoft products until a complete transition can occur. In the case of most State employees very inexpensive desktop computers could be utilized since much of the processing and all of the data would be dealt with by the cloud.

On lower-end hardware Linux based systems perform much better than Microsoft based systems due to lower processor and memory overhead.

There is immense savings to be found in centralization albeit at great political expense. Like a majority of other states, governments and private entities, the State of Ohio can and should elevate the Office of Information Technology to a prominent position giving them the political capital necessary to best serve the needs of State government and the taxpayers.

Conclusion

There will always be a pull over what are appropriate uses of technology in the work place. The argument is greatly rooted in the idea that managers must maintain rigid controls or else they will lose productivity. While this may be the case with certain individuals the reality is that efficiencies and cost savings never previously imaginable are attainable by properly utilizing the collaborative technology available and managing as is appropriate to the current technological environment. The State of Ohio has an enormous opportunity that should be taken advantage of.

Basic Principles

• Social Applications to interconnect groups across disciplines and agencies.
• Multi-media learning environments.
• Encourage use of social networking for professional growth.
• Make office perform more collaboratively.
• Flatten organizational hierarchy.
• Provide data in XML and other standard formats internally and externally.
• Common data repository for scrubbed (no address/SS#) data.
• Connect voicemail to text-mail (i.e. iPhone or Google vmail to text apps) which is more portable.
• Tag all documents as part of process to create “linked data”.
• Allow entry point for external users.
• Create internal cloud as shared service with all agency applications being web-available.
• Create all future applications as Open Source Software (OSS) for higher vendor competition.
• Follow GPL/OS standards to allow for better transparency.
• Utilize OSS desktop for cost savings.
• Provide API to allow for greater customization by users.

Introduction

Recently I was asked to weigh in about setting policy as it is related to the procurement and usage of Open Source based software. Specifically I was asked to research the issues, concerns and best practices surrounding the use of Open Source Software by State of Ohio agencies. Open Source Software represents a diverse offering of software in regard to functionality, quality, innovation, support and licensing. This diversity brings with it several issues which may be addressed through policy guidelines to mitigate some of the inherent risk.

Background

Open Source Software (OSS) is generally considered software which the source code is made publicly available through licensing vehicles that follow the Open Source Initiative (OSI) definition of Open Source[1]. Licensing generally allows for free redistribution, inclusion of source code, derivative works, non discrimination (human, endeavor, or technology) and expects continuation of equivalent licensing. According to Gartner, Inc. 80% of commercial software packages will include some element of OSS technology by 2012[2]. Currently the State has addressed some aspects of licensing of OSS for use by the State under ITP-A.26 which should be considered for inclusion or modification when moving forward with a standalone OSS Policy. ITP-A.26 section 9.7 deals specifically with acknowledgement of the unique licensing under OSS.

Currently very few states and only a few federal departments have attempted to enact policy and legislation directly related to the procurement, development and usage of OSS. Oregon and Massachusetts appear to be the most active in pursuing prominent OSS IT platforms. Self reporting shows that only 6% of US government agencies are using OSS across the enterprise while 39% report pilot and evaluation projects[3]. The largest endeavor so far appears to be the cross state agreement between the KS, MA, MO, PA, RI, UT, VA and WV who have stated, “The Government Open Code Collaborative is a voluntary collaboration between public sector entities and non-profit academic institutions created for the purpose of encouraging the sharing, at no cost, of computer code developed for and by government entities where the redistribution of this code is allowed.”[4]

Policy Considerations

• Lower burden on procurement process. Low or No Cost license agreements create a situation where OSS components and systems could be brought in under minimum procurement guidelines. This lower burden on the procurement process would also remove an opportunity for enterprise level review of agency or department IT initiatives. Procurement of OSS should be considered within the more defined components of procurement governance to reduce the risk of unsupervised implementation.

 

• Reduced Total Cost of Ownership (TCO). Low or No Cost license agreements along with some level of community development (could include internal agency interactivity or external state consortiums) and support greatly reduces the TCO. However mission-critical systems may still rely on strong vendor relationships which could negate some of the advantage to TCO. Data supports that TCO is between 15-35% less for OSS over closed source software.

 

• Increased competition. While OSS does not provide true vendor independence it does provide for greater competition between software products. Software development can have a very high cost of startup in both capital and intellectual resources making the use of open source development much less burdensome. Consideration should still be given to vendor supported OSS however. Mission-critical enterprise quality software means that the vendor relationship may need to be maintained to mitigate risk because reliance on community support may not be reliable enough to complete vendor independence.

 

• Development and training. Internal and combination internal/external development around open source standards creates questions about worker skill sets as well as administration and user understanding. By maintaining a policy of using open source standards the interoperability of systems and the repurposing of developers across the enterprise should be possible. According to the International Open Source Network (United Nations Development Program) there are currently training standards in place for the most widely used OSS and this should be a consideration during development and implementation of OSS systems5.

 

• Maintenance, support, security and liability. Use of OSS can potentially increase the State’s liability risk as it pertains to maintenance, support and security. With no vendor under direct responsibility for these components the State’s liability rises. OSS development may include multiple flavors of a very similar base system. Version controls from original procurement through the lifecycle should be in place. As a component of reducing liability, documents tracking all changes made by the State or other related entities should become a permanent component of the software package throughout its lifecycle. Ohio IT Policy ITP-B.1 “Information Security Framework” should be considered in the development of OSS Policy; particularly section 5.1 which details Risk Management.

 

• Licensing and Legal review. As an extension of the Risk Management component OSS usage may include multiple licensing requirements through the community nature of development which would include intellectual property (IP) and other rights that may have no sunset. Legal review should occur for all licenses attached to the OSS being introduced. States generally have greater protection over IP lawsuits than private individuals and organizations due to supreme court case 527 U.S. 666 (1999) aka Florida Prepaid. A document tracking the license requirements must become a permanent component to the software package throughout its lifecycle to protect against any possible infringement.

 

• Shared resources. Use of OSS and related standards allows for ease of collaboration between independent organizations or even private individuals. Consideration should be made in respect to how government systems built under OSS licensing might be utilized by other entities and the associated risks that may bring to both the new user and the security of the developer.

 

• Mandating, preferring or normalizing OSS. In an effort to provide greater government transparency, TCO savings and vendor diversity some states and national government bodies are mandating OSS standards or showing preference for OSS. The least controversial path is providing very limited policy to normalize the use of OSS instead of outright preferring closed source software and standards.

References

1. Coar, Ken 2006 “The Open Source Definition” The Open Source Initiative http://www.opensource.org/docs/osd (accessed 12.02.2008)

2. Driver, Mark 2008 “Key Issues for Open-Source Software, 2008” Gartner, Inc. http://www.gartner.com ID Number: G00155800 (accessed 12.02.2008)

3. Di Maio, Andrea 2008 “Government Survey Dispels Five Myths About Open-Source Software” Gartner, Inc. http://www.gartner.comID Number: G00154776 (accessed 12.02.2008)

4. Lewis, James A. 2007 “Government Open Source Policies” Center for Strategic and International Studies [CSIS] http://www.csis.org/component/option,com_csis_pubs/task,view/id,4663/ (accessed 12.02.2008)

5. Wong, Kenneth 2004 “Free/Open Source Software: Government Policy” International Open Source Network (United Nations Development Program) http://www.iosn.net/government/foss-government-primer/foss-govt-policy.pdf(accessed 12.02.2008)

Overview

In an effort to maximize the collective knowledge and experience of individuals, organizations have utilized technology to collect and distribute this information. Storytellers were the first repositories of this collective knowledge. Eventually writing allowed for the building of libraries which was the first great expansion of knowledge repositories. Johann Gutenberg created the next piece of the puzzle with the printing press, making it easier for all levels of knowledge to be captured and distributed. Finally in modern time we have learned to create databases which store exponentially larger amounts of common knowledge and experiences in a searchable, electronic format.

Presently we find ourselves on the cusp of another leap forward for the distribution of knowledge. My proposal is to build a knowledge mapping database that provides all participants with the information they need in any number of formats in real-time. This system will fundamentally change the way the State of Ohio does business; in doing so it will drive business and residents to the State of Ohio.

I call this the Human Knowledge Mapping system. The mechanics of the system can be implemented using primarily off-the-shelf products while the soul of the system will be the employees of the State, then students, then business professionals and finally individual citizens. Through the growth of this system it is possible to dream of a day when all Ohioans, and perhaps all the people of the world, can interact and grow the collective knowledge exponentially.

Summary of Benefits and Initiative Intersections

The scope of this system will leverage current resources in such a way that several current initiatives will benefit. The intersections with these initiatives will not only help this system grow but will help these initiatives to exceed their initial expectations. Below is a short list of the collaboration between this system, the initiatives and realizable benefits.

• A primary benefit is that a rapid, real-time response to inquiries will be available to all State of Ohio employees across agencies.
• Human Resources will have access to a cross agency listing of employee knowledge backgrounds, based on current responsibilities as well as previous positions and extracurricular based knowledge.
• OAKS will benefit directly through this exponential growth of personnel information which will expand agency participation and use of the OAKS system.
• Internal Customer Service will improve immediately through the sharing of experience and knowledge when working through standard procedures or previously undocumented and unexpected scenarios.
• External Customer Service will improve as the system expands. By allowing the public at large to utilize this real-time system, many issues will be resolved more efficiently. Industry engaged in business with the State will be able to resolve issues more efficiently which will increase participation and investment.
• Disaster Recovery efforts will benefit from a wider pool of knowledge to fill positions in the event of a significant loss of personnel or in the event that certain activities require greater resources. Using physical mapping mash-ups with this system will provide information about where local and regional resources can be most efficiently relocated.
• The Healthy Ohio Initiative identifies issues with healthcare worker shortages. Upon wider implementation of this system specific shortages and resources can be assessed and resources can be reapplied appropriately.
• The Broadband Ohio Network Initiative has a focus on innovation, research, education and economic competitiveness via OSCnet. This system will utilize Broadband Ohio and OSCnet to create greater collaboration between organizations and individuals, substantially affecting all of the listed focuses.
• The Knowledgeworks Initiative looks to use human resources and technological resources in a way to improve High School, College and Adult Education. This system would provide greater collaboration between educator and learner. The system will open up greater channels for the pursuit of knowledge in education.
• The Advantage Ohio Initiative can be expanded by providing detailed reports about concentrations of knowledge, locations of these bases and specific needs to businesses interested in investing in Ohio. New markets will be exposed as greater reporting about resources and needs occurs.
• Via the development of this system the State of Ohio will establish itself as the most technologically innovative government body in the world.

System Mechanics and Function

The Human Knowledge Mapping system begins with a relatively simple procedure. A sample group will develop lists of what they feel they have to offer as knowledge. This knowledge will include current job analysis, previous work experiences, personal hobbies and other extracurricular based knowledge. It is not necessary that all of the knowledge be professionally related however, as part of the documentation, lists should be separated into professional, non-professional, current and previous experiences. This will give users a more comprehensive understanding of the support they are receiving. Through compilation of these lists common terms will be grouped together under headings which will then develop into a matrix of modules; or, for a more visual context these could be considered islands of knowledge.

The primary components that make up the system will include a chat screen live on every participant’s system, a database, a chat window with active text on agency websites along with a data collection tool. This knowledge information will be added to the system database which could be a part of the OAKS database. A user will enter a question into the agency website chat window. The question will be read by the system which will then use keywords to search the database and determine which islands of knowledge are related. The islands will be identified to the user so that they can see which connections their question has made; this gives the user a chance to refine their question further. All of this will be done using a graphical interface which cues the user to choose narrower keywords (tags) so as to make the system as efficient as possible.

Once the question has been refined the user submits it which broadcasts it to the chat window of all participants who reside on the related islands of knowledge. Assuming the question broadcasts to several people any number of them may participate in the chat. During this chat period the system will record the full session. Links may be shared which the system will follow and record a snapshot of for future reference. File sharing will be limited to include virus protection as well as document protection scans. The recorded chat session is then stored in the database with a record of all participants, the initial question and searchable keyword text. For the initial installation of the system this chat function will use keyboarding but future releases could have live voice or video capability added.

At the end of the session the user who asked the initial question will be emailed a transcript of the chat session and will be invited to participate in the system. If they choose to participate then they will enter their personal knowledge information and, once accepted, they will be given access to the desktop chat program. As the system grows the recorded chat sessions will be included in the initial keyword search so that in some cases a new chat session will not be required to get the user the information they need. These searches will also include access to related links, screenshots and files found in the text searchable chat sessions.

The chat sessions will be secure and the fact that they will be recorded 100% of the time will dissuade system abuse. Initially the system would be utilized for internal state functions but could quickly be implemented with local government co-ops as well as educational institutions. Eventually the chat window would be active on the State of Ohio website allowing for citizens and industry professionals to quickly resolve issues they have in the course of doing business with the state.

As mentioned earlier, all of the key software components already exist. Current examples of similar pieces of software can be found in wide use in private industry and on personal websites. The key difference is the way that this system will bring these software components together. For example, Hewlett Packard provides support via chat sessions which are recorded and then emailed to the user along with a follow up questionnaire. Blog sites running WordPress regularly utilize tagging tools which provide a visual search experience by assigning larger font sizes to tags that relate more directly to the current topic and slowly taper off as the tags lose relevance. This software also displays the tags in a visually stimulating way such as a layered circular design. Web search engines such as Ask.com provide instant links to multiple media types during search queries. This would be beneficial as other resources are added such as video, audio and graphical or photographic information. Wikipedia currently uses this collaborative style content building and provides yet another dimension of this system. You can look at any number of thousands of sites that use Google maps to create mash-ups of information with physical locations. As you can see the mechanisms necessary to deliver such a system are widely available.

By using Service Oriented Architecture the system will allow for quick and easy enhancements as participation increases and needs are realized. The key element to making this a successful system is in its simplicity which will draw in participants. Since this is a collaborative support system, support costs will be low because the participants in the system will provide the user support.