Numbers can tell the whole story, or miss it entirely

June 2, 2015 Leave a comment

Several of the Defense Department’s top brass have said it in recent speeches. With the cash it has on hand, Apple could acquire the stock in Lockheed-Martin, Raytheon, Northrop Grumman and General Dynamics outright. I checked the numbers myself. It’s true. This week, the market cap of the four defense contractors was roughly $170 billion. Apple’s cash hoard was just shy of $200 billion. Apple’s own market capitalization is somewhere in the $772 billion range.

Almost in the same  breath, military leaders note that with a market capitalization of around $225 billion, Facebook is more valuable than the same four companies combined.

Air Force chief of staff Gen. Mark Welsh III and former Defense Deputy Secretary Bill Lynn both cited them within the last few weeks because of the scale of research and development going on in consumer and commercial electronics and computer science, versus R&D investments benefitting the defense sector. Unsaid is how surprising hard-headed military leaders must find it, that something as seemingly trivial as Facebook could command such value in comparison to companies that make sophisticated hardware like Patriot Missiles or F-35 fighters or nuclear submarines. The comparisons come in the context of the DOD wanting to re-establish the technology offset that produced so much U.S. military superiority in recent decades.

Numbers startle. They provoke thought. But they only tell part of the story. Apple may be worth three quarters of a trillion dollars, but it can’t build bombers or ships or submarines. It’s entirely capable of building sensor networks and software underpinnings for weapons, but it won’t.  Numbers don’t tell what’s in a company’s DNA.  SpaceX, essentially a start-up with 3,000 employees but no public shares or even statement of revenue, snared $1 billion in private financing in January. And now it’s on the very short list of exactly two companies certified by the Air Force to launch rockets putting military satellites into space. It competes with a partnership of giants Boeing and Lockheed. Boeing built the first modern airliner — in 1933.

Few remember the mini-computer wars of the 1970s and 1980s. But like the PC business in the 90s, the minicomputer business had a large number of fierce competitors — Apollo, Digital Equipment Corp, Data General, Hewlett-Packard, Honeywell, IBM, Nixdorf, Prime, Wang. So fierce was the competition, it spawned the 1981 Pulitzer Prize-winning non-fiction account by Tracy Kidder. Soul of a New Machine chronicled the late Tom West, an engineer at Data General, as he coaxed out an inexperienced team a new 32-bit processor to compete with the mighty DEC.

Nearly all of the companies and people involved in the minicomputer wars are gone, but the title of that book persists. One reason is that so often, objects defined by numbers do take on a sort of soul or presence, because people made them. “Soul” might be a troubling word to some people, connected as it may be to something the deity conferred only on human beings. But the idea that numbers add up to more than numbers — that’s universally valid.

Over the recent Memorial Day weekend the Rolling Thunder rumbled through Washington for its annual Memorial Day observance. My wife was bicycling right by where the motorcyclists were rolling onto the course. She stopped and took a 90-second video of the endlessly varied bikes and riders chugging past. Most rode Harley-Davidson machines. We watched the video over and over. What is a motorcycle but a chassis, a pair of wheels, a V-twin engine, and tear-drop shaped gas tank? (Gold Wing riders, spare me.) Ah, but of course a Harley-Davidson or any motorcycle is more, much more, than the sum of its parts. Gear heads know the meaning of the stoke-bore ratio, horsepower, torque, and the myriad other numerically-expressed specifications that describe motors. But they don’t explain that sound.

Computers, motorcycles, musical instruments, whatever your passion, are more than the sum of their parts and specifications. That’s why automobile junk yards and airplane boneyards look so sad. Or reading about a theatre or defunct church junking its pipe organ.

But what about human endeavors? They, too, are more than the sum of their parts. This came to mind recently when reading an analysis of federal inspectors general based on research at the Brookings Center for Effective Public Management. I also interviewed John Hudak, one of the study’s co-authors along with Grace Wallack. Their research quantifies the work of IGs using a return-on-investment metric. It proceeds from the postulate, correct I believe, that in general IGs return many dollars to the government for every dollar spent operating the IG office, and that this is quantifiable. The authors acknowledge that the pure dollar ROI metric is less useful in agencies where the mission isn’t primarily disbursement of money. So, for example, the IG with the highest ROI is that of the Social Security Administration, where the return is $43.60 for every dollar the office costs. The lowest financial return belongs to the Justice Department, where the IG produces a net cost. It’s ROI is about 43 cents for every dollar spent on the IG operation.

As Hudak points out in the interview, that the ROI looks weak is not a reflection on the Justice IG operation, just that the particular ROI number doesn’t really capture the essence of the office and how it goes about its work. The Justice IG shop is looking at programs mostly. The department doesn’t exist to disburse hundreds of billions of dollars, as Social Security expressly does.

No metric can really capture the essence of any object or program, or the people’s dedication to it.

Government overlooks diamonds in its own back yard

May 26, 2015 Leave a comment

Nearly a century ago, a speech called “Acres of Diamonds” ranked high on the entertainment charts. Russell Conwell, the founder of Temple University and the author of the essay on which it was based, delivered it 5,000 times between 1900 and 1925. Although part of the American canon of rhetoric, few nowadays know much about it. The main theme of the essay and speech is simple: Often people overlook their own back yards when seeking fortune and success. It’s an old-fashioned sounding piece, but it came to mind earlier this week during a panel discussion I moderated along with Jonathan Aberman of TandemNSI and Amplifier Ventures.

Our ostensible topic was how government program managers, R&D people and executives can more readily take advantage of the entrepreneurial and start-up technology community in the D.C. region, the federal government’s back yard. But a good deal of the discussion centered on the government’s seeming fascination with Silicon Valley. Homeland Security Secretary Jeh Johnson says he’ll establish a satellite office there. Defense Secretary Ash Carter also has the Silicon Valley bug. The Obama administration has recruited several appointees from the Valley, including Chief Technology Officer Megan Smith, lately of Google. The White House Digital Services team is headed by another ex-Googler, Mickey Dickerson. The rest of it is heavy with people from Twitter, Facebook, Amazon and the like.

With respect to DOD, maybe a little perspective is in order. Mike Daniels has been founder, executive, seller and acquirer of federal contractors for decades. He says Valley fever has swept the government before. He counts three times over the last 35 years or so. And every time federal leaders traipse out there looking for start-up diamonds, they find out how culturally different things are out there. Without judging which is superior, Daniels says he’s never heard a Valley venture capitalist or entrepreneurial say a primary reason for a new company was to help the government solve a problem. The systems integrators and technology product companies locally have often been started by former feds, including military officers. Daniels says their orientation is, of course to make money, but their frequently-stated mission is to help the government.

Tech veteran Anup Ghosh — he founded cybersecurity outfit Invincea — comes from DARPA and so has something of a federal orientation. He says federally-focused startups proceed from a different model than those in Silicon Valley. In the D.C. region, the typical model starts with government services. This lets them win some contracts, establish a revenue stream and make a profit fairly soon. It’s a bootstrap model that can lead to second-round financing. By contract, in the Valley startups tend to lead with groundbreaking products, little thought of revenue, and the agility to keep morphing until they hit on something someone wants to buy.

Ghosh argues, the services or body-shop model doesn’t scale easily because it requires a constant linear increase in the number of people to be able to do more professional services. By contrast, products, especially software, can scale exponentially without a matching overhead increase.

The bootstrap-contract model, according to former Naval R&D guy Bob Morgan, a founder of MorganFranklin Consulting, gives the D.C. startup community a boring reputation. But often, he and others note, the products they are working on are secret because they have a national security application. Or because the government is reticent, and doesn’t want contractors talking about what they are working on.

Ghosh, Morgan, Daniels all agree, the D.C. startup environment is smaller and more specialized than that of Silicon Valley, and it lacks a catchy and enduring moniker. Daniels noted the D.C. tech community tried to brand itself back in the 1990s, but whatever name it came up with quickly faded. Silicon Valley acquired its nickname around 1980, when silicon — semiconductor design and manufacturing, as opposed to software — powered the torrid growth of the place. Plus, its name came externally and spontaneously, not from an ad agency. Throughout the world, D.C. is inseparable from the federal government. Those of us who live and work in and around D.C. know it’s a real place with real people and culture, but no ad agency could alter the external view of it.

The D.C. region is rich in startups and entrepreneurialism in cybersecurity, biology and genetics, robotics, and data analytics. From the discussion I heard, the consensus is that the federal government’s requirements remain a valuable source of money to fuel this growth activity. TandemNSI founder Jonathan Aberman correctly points out that Silicon Valley itself was fueled early on by military and NASA requirements for products pioneered by Fairchild Semiconductor and its many offshoots. (The original company in San Jose, not the one in Portland, Maine is typically put at the top of Silicon Valley genealogy charts.)

Silicon Valley today isn’t inclined towards selling to the federal government as a business goal. And, as local venturist Mark Walsh points out, big bets on tiny companies without revenue and serial business failure are common in the Valley .In the D.C. region, business failure and loss of investors’ money is less forgivable. Plus people who bootstrapped a company, got government contracts and have some sort of cash flow aren’t accustomed to the idea of ceding control to a V.C. More evidence of the big cultural divide.

All new companies, regardless of origin, face the same hurdles in selling to the government, the panelists agree. Ghosh says there’s nothing in the Federal Acquisition Regulation to prevent a company from getting a contract within five days should an urgent need arise, but typically the bureaucracy plays it safe. Often for good reason. By the same token, the purportedly 18-month minimum sales cycle for a federal agency isn’t so far off what those same companies encounter when selling to the Fortune 500.

Not fully answered the other morning were these questions:

  • Does Silicon Valley really hold the scratch to the government’s technology itch? My take, yes and no. Of course there are companies originating in the Valley that could help solve federal mission challenges. It requires a great deal of research the selectiveness. Whether the Pentagon needs a mini-Pentagon off U.S. 101 is debatable.
  • Does the government overlook the acres of diamonds in its own back yard? I don’t think so. Invincea itself is a case in point. Lots of product and services companies got their start with federal requirements, and not just items unique to the military.

In Silicon Valley, people start companies with an idea, and expect the world to join in. No one realized there was a better way to hail a taxi cab. But Uber — we’ll stretch the Valley to include San Francisco — became a valuable company,  billion dollar unicorn, in Valley parlance, without owning a single car or employing a single driver. But a mobile app will never get a human to Mars, or destroy an cutthroat enemy, or analyze a half-trilliion-dollar federal program to ferret out waste, fraud and abuse. Given the size and diversity of its missions, the government needs to talk to entrepreneurs everywhere.

Fact: Federal Employees Have Property Rights To Their Jobs. So There

May 14, 2015 Leave a comment

In my line of work, I learn something new every day. I’ve been at this a long time, so it’s particularly gratifying to get some insight into something I thought I thoroughly understood.

The most recent example came in my latest, more-or-less monthly interview of Merit Systems Protection Board Chairman Susan Tsui Grundmann. Susan is one of those rare people with an exquisite balance of qualities. She seems idealistic and realistic at the same time. She hews to the mission, but retains just a tincture of skepticism. She knows more than she says, a refreshing characteristic. She’s open but discrete. And she has a sense of humor.

Grundmann pointed out some things that everyone should know, but need pointing out nonetheless.

This fact startled me: Career federal (and other public) employees have a property interest in their jobs. It’s a major distinction with the private sector. It’s a long established fact. But when’s the last time you heard it stated? Grundmann points out, this property interest has been affirmed by the Supreme Court through its insistence on due process for removing federal employees or other adverse actions. It requires 30 days notice, followed by the right to an appeal hearing. If an employee does something seemingly criminal for which imprisonment could result, removal requires 7 days advanced written notice. This apparatus of due process grew in direct response to the spoils system of the 19th century.

However frustrated members of Congress or editorial writers might be by the performance of managers at the Veteran Affairs Department, Drug Enforcement Administration, or Secret Service, those employees have a constitutional property interest in their jobs and are entitled to due process. All of this apparatus doesn’t absolve supervisors from good management, because poor or malfeasant performers can and often are removed from federal employment. The property interest doesn’t translate into immutable tenure.

In fact, that was another standout fact in the Grundmann interview. Between 2000 and 2014, 77,000 feds were fired for performance. We don’t know how many else left ahead of being fired. But this fact belies the popular notion that a federal job is a job for life.

Grundmann was annoyed that media outlets often mischaracterize how federal civil service due process works. One national newspaper stated, but later corrected, that federal employees removed for cause stay on payroll until MSPB completes its adjudication. That’s flat wrong, Grundmann says. They’re not being paid.

Last week I wrote about fed-bashing, and how I’m skeptical that it really is a thing, to use the Buzzfeed-age vernacular. What is a thing is widespread misunderstanding of the rules for federal civil service. That’s why anyone who has an interest in federal civil service, or chooses to carry on that it should be exactly like the public sector should read the MSPB’s new report to Congress. It details the due process and, equally important, the history and court cases behind it.

Few federal employees, or anyone, probably remembers James Loudermill. His 1979 dismissal as a security guard from the Cleveland Board of Education ended up at the Supreme Court, which upheld the due-process-from-property-right principle. He might not have been the finest test case. He’d been convicted of grand larceny 10 years earlier, but lied about it on his job application. Justified as the summary firing may have seemed, ultimately he was found to have been denied due process and his property right to his job was upheld by the Burger court.

Do Feds Really Receive Regular Evil-Tongue Lashings?

May 7, 2015 1 comment

Fed-bashing might headline a lot of articles and interviews, but I question whether fed-bashing exists as a palpable phenomenon. Or whether it’s loud or prominent enough to justify hang-dog morale on the part of federal employees.

Two thousand years ago the Talmudic scholars cautioned against the “evil tongue” — disparaging of people by the spoken word. Many cultures share this orientation. When I used to teach the Dale Carnegie course, the very first human relations principle instructors were to convey to class members was, “Do not criticize, condemn or complain.” Given the level of discourse nationally, you’d think hardly anyone thinks that way. The “sticks and stones…” ditty lacks truth — words and name-calling can and do cause damage.

Like many readers, I’m addicted to the comments sections in big publications. Political stories, harshly polemic columns, and sensational reports with heavy cultural overtones often generate thousands of comments. Some of them I read with my hand close to covering my eyes, as if watching a slash-and-guts movie, so vitriolic the posts often get.

To qualify as fed bashing, I think statements or published material must meet certain criteria:

  • It originates in the mainstream, not from the fringe. Taken as a whole, fringe groups hate everything. So anti-government raves coming from Timothy McVeigh groupies or The Westboro Baptist Church don’t count. Or…
  • …it lumps every federal employee in when using particular ones, like some of the failed managers at Veterans Affairs, as if everyone at VA was bad. And…
  • … its source carries sufficient standing or prestige with ordinary people such that they might take away from the disparagement the distinct possibility that federal employees really are all bad.

Many statements that might strike the ear as fed-bashing really aren’t. A story in the Wall Street Journal the other day detailed how administrative judges at the Securities and Exchange Commission prevail in 90 percent of the cases. That leads to the legitimate concern that the rulings don’t come from an impartial judiciary but rather from agency employees judging the charges brought by friends and colleagues. I’m not issuing my own opinion here, just pointing out the arguments. So if someone were to say, as someone inevitably will, “I don’t want some bureaucrat deciding a case brought by another bureaucrat, I want a honest-to-goodness judge!” I judge that not to be fed-bashing but rather an opinion, however clumsily stated, that a powerful administration lacks sufficient oversight and is prone to conflict of interest.

Congress remains a rich source of what some construe as fed-bashing. Regular debates over federal pay and benefits can produce the feeling of being bashed. But think about this: Compensation of employees in every industry, public or private, ultimately respond to markets. In my career of close to 40 years, I’ve twice had pay cuts imposed by employers citing financial or economic conditions. Like millions of worker bees, I’ve seen changes in pension plans, 401K contribution formulae and health care benefits — rarely for the better. Lord knows the media, whatever that is, gets regularly bashed (often in the media) — and often deservedly so. I bash it myself sometimes!

My advice for federal employees who think they hear fed-bashing: Shrug if off. Out of 319 million Americans, probably 300 million only occasionally even think about the federal government. The rest have opinions falling somewhere on the scale from the John Birch Society to the  Socialist Party of America. If you work in the public sector, you operate in close quarters and at the behest of politicians and their appointees who, in many cases, signed up precisely because they thrive on conflict. So sometimes you’ll get splashed with what hits the fan. Most of the time, it’s not aimed at you. Sometimes you gotta duck.


Digital Services Shouldn’t Feel Like Thumbscrews

April 30, 2015 Leave a comment

I’ve been a blood donor for 40 years. I like the cookies they give out afterwards. Over the years, the Red Cross, through which I donate most frequently, has improved its pre-donation process from a paper-intensive pain in the butt to a mildly annoying, semi-automated deal. Unlike the Transportation Security Administration, there’s no “Pre”-like arrangement under which regular donors can bypass the tedious questions about whether they’ve lived in certain places or engaged in particular behaviors associated with no-good blood.

Sometimes I donate via Inova. Once a year they team up with the Washington Nationals. You go to the ball park at a designated time and after making your blood donation, you receive two game tickets, a T-shirt and a park tour. I don’t mind seeing the massive underground beer distribution system one more time.

Inova-Nationals’ process for setting up your designated time provides a lesson in how digital services can, if badly designed, make a process worse than the old way of phoning and singing up. From a Nationals e-mail you follow a link to a page listing the entire donation day in 15-minute intervals. Pick your time, click on the “schedule” button next to it, and end up at an Inova logon page.

One of the most irritating features of the online life is the multiplicity of “accounts” you have to have. I knew I must’ve had an Inova account. Its logon page seemed to recognize my e-mail and donor ID, but I simply could not get past the multiple security questions, ending up stuck in a frustrating closed loop of dialog pages. So, I dialed the 866 number. After being on hold for about 5 minutes, a young operator came on and said, “Hello, Mr. Temin.” Caller ID. She made the appointment in literally 15 seconds. No passwords, IDs, security questions. They’ll check my driver’s license when I get there. I confirming e-mail came seconds later.

This is the sort of thing Mikey, Tony, Megan and the other stars in the government’s digital constellation need to keep in mind. People don’t want digital services per se so much as they want easy access to whatever services they seek.

One of the news sites published a report on a bogus-sounding survey of Americans, concluding we are not particularly interested in digital services offered by the federal government. The implication of the story is that the government is chasing something people don’t want. The particular research firm has a reputation for iconoclastic studies. It gets headlines. But the research and the story dutifully transmitting it are beside the point.

For the average American, the federal government is typically no more central to his or her daily thinking than the question of whether Queen Victoria should have annexed the Transvaal. So what? That’s what you should expect from a healthy, free people. Yet everyone at some point in their lives will need to interact with a federal agency. So it’s incumbent on agencies to have good online services. As Veterans Affairs CIO Steph Warren pointed out the other day, if the surveyors had asked whether people want fast, convenient and accurate results when they do need something from the government, 99 percent would have answered, “Of course!”

Open data, or the availability of federal agency data sets, also fail to catch fire with the American imagination. A Pew Research Center study confirms that two thirds of Americans in the past year have visited a federal web site. Most people were looking for some piece of information, not specifically wanting to download a data set. In fact, most people think government doesn’t do a particularly good job of sharing information. I suspect that’s related to what they hear in general media about government and the fairly low responsiveness to Freedom of Information Act requests so many agencies exhibit. Here again, that doesn’t absolve agencies from being thorough in how they release data. Data sets are’t suitable for mass consumption. But each one is highly relevant for some business, trade group or individual. And so agencies spend their time well when they put some effort into making their data sets accessible.

Blood donors interact with Inova or the Red Cross at most a few times a year. But when they do, they should have a good digital experience. By the same token, most people ping the government way less frequently than they do Zappos or Amazon. But when they do click on a dot-gov, they expect the same grade of experience.

I hate to break it to you, but FITARA does NOT give federal CIOs total IT budget authority

April 19, 2015 Leave a comment

Sorry, but CIOs will not get the budget authority everyone has been reporting they will under the Federal Information Technology Acquisition Reform Act.

Now that I have not buried the lead, a little context.

One of the more intriguing recent Government Accountability Office look-sees didn’t get enough attention. I featured the report on The Federal Drive when I interviewed GAO’s Dave Powner, but I hope more agency executives take a look. The gist of the report is this: Federal chief information officers have to file a lot of reports every year, and  many of them are at least a partial waste of time. CIOs must file 36 mandatory reports, and a survey GAO conducted shows CIOs think 24 of the reports don’t help them much in managing their responsibilities as they relate to their departments’ priorities.

One big reason is that many of the reports concern progress on generic, government-wide requirements — IT strategic planning, capital and investment management, cybersecurity, IT acquisition, and e-government. GAO points out, the Office of Management and Budget directs CIO’s to concentrate on things like proper governance, meaning budget, acquisition and portfolio analysis. Or commodity IT, meaning data center consolidation, cutting redundant systems and using more shared services. That’s only natural, given OMB’s governmentwide view.

It’s a prime example of how large bureaucracies, both private and public, default to detailed process to achieve goals. The recipe approach indeed works in many domains where you want absolutely repeatable outcomes. If you are producing 1,000 wedding cakes a week or 1,000 airliners a year, process is paramount.

For CIOs the problem with all this process and reporting, aside from the cost and questionable relevance of much of it, stems from how little use it can be in resolving department-specific challenges. Filing 36 reports a year didn’t get the Veterans Affairs Department to Nirvana in scheduling and treating veterans, or Health and Human Services to stop wasting 15 percent of Medicare and Medicaid dollars. It won’t get NASA to Mars any time sooner, nor will it save State Department systems from being turned into a botnet for Russian hackers.

The reporting questions came just as it dawned on CIOs that the Federal Information Technology Acquisition Reform Act (FITARA) I mentioned at the beginning, is law. Now CIOs are awaiting OMB guidance on how to implement what passed as a large amendment to the 2015 National Defense Acquisition Act.

Some CIOs are asking what FITARA will really change. Its most widely-quoted visible provision gives CIOs authority over departments’ IT budgets. That authority was not given by the 1996 Clinger-Cohen Act, which made CIOs the primary advisors to agency heads on IT architectures and investments. But it left primary budget authority with chief financial officers, consistent with Title 35 of Chapter 44 of the U.S. Code. CFOs understandably like and defend their budget authority, as many a CIO has learned.

In my reading of the law (scroll WAY down to Section 5101), CIOs still don’t get total budget authority. FITARA elevates CIOs such that “amounts appropriated … for information technology shall be allocated … in such a manner as specified by, or approved by, the Chief Information Officer of the agency in consultation with the Chief Financial Office of the agency and budget officials” (emphasis mine). So the complex interplay among OMB, CFOs and CIOs is still there, but now CIOs have more weight. What actually changes will depend on CIOs making the case for how they allocate IT spending. To the extent everyone else involved in the decision agrees on specific agency priorities, that shouldn’t be a problem.

A couple of CIOs I’ve spoken to note that because of the highly federated nature of their departments, there’s a long tradition of autonomy by large agencies over their own IT. That won’t be easy to change. Down-a-notch CIOs won’t give up easily. Department CIOs don’t want to become clerks, signing off on every little thing an agency level CIO decides to buy. If title defines the job, agency and bureau CIOs will have “deputy,” “assistant” or “associate” added to their titles. That might help underscore the superior authority of the departmental CIO.

FITARA might bring the beginning of the end for so many CIO reports. Section 5102 gives the CIO Council a bigger job. It’s supposed to “develop cross-agency portfolio management practices” and “issue guidelines and practices for infrastructure and common information technology applications.” All of this is to be “used as the basis for comparing performance across diverse missions and operations in various agencies.” I’m thinking maybe the CIO Council, which of course is made up of CIOs, could take over some of the common reports departmental CIOs are obligated with. Then, CIOs would have more time to think about specific departmental missions and playing nice with the CFOs.

There’s much more to FITARA. The law pushes data center consolidation, rationality in acquisition of software licenses, cloud computing, and reduction of duplicative contracts. Nearly every item in FITARA has been proposed either in law or by executive branch initiative going back to something called Reform ’88, launched in 1982. It made the cover of the first issue of Government Computer News, but rated only Page 7 in Computerworld.

Categories: Uncategorized

Five Steps For The Government To Regain Trust

March 30, 2015 Leave a comment

Last month the Obama administration rolled out something called the federal feedback button. Officials describe it as a Yelp-like way for people to give feedback on the online service they get. That is all well and good. People visiting federal websites should have a good experience, easy to navigate and returning the results they seek. I think for the most part they do. Still, you can never have too much feedback. Sites vary. Some are still tough to navigate, others are right up there with the best of them. Some adapt perfectly to mobile devices, others have yet to be redone with responsive, mobile-aware coding. But on the whole, people responsible for federal web sites care a lot about their work.

One goal of the federal feedback button puts a little too much on the shoulders of web managers. Specifically, the notion that better digital service and gimmicks like a website button can help restore faith in government. A lousy web experience might reinforce the notion that government is incompetent if a visitor is inclined to think that way. Most people take a poor web experience for what it is — a poor web experience. To make an analogy, I’m highly loyal to the brand of car I drive. The company’s website is over-engineered and precious to the point of being annoying and hard to figure out. But that shakes my faith in its web people, not in the car.

Distrust of government stems from problems way deeper than digital service. All you have to do is scan the last few weeks’ headlines to see examples of what makes government sink in citizens’ estimation. None of these sources of mistrust will be remedied with the federal feedback button.

Nor will they be fixed with simple-minded assertions about the efficiency or motivation of the federal workforce. Good people working in bad systems will produce bad results. The way to better, more trustworthy government tuns through fixing the systems and processes, and funding them adequately. Then you’ve got the tools necessary to hold people accountable.

Here are my five picks for systems that need fixed to restore faith in government.

1. Fulfill FOIA requests. How many more decades must pass before federal agencies figure out a way to answer Freedom of Information Act requests within days or hours, and then fulfill most of them? A default to secrecy and withholding clings stubbornly. Just a month ago the Center for Effective Government came out with another dreary accounting of agency FOIA performance. The open data movement, exemplified by and the hiring of a chief data officer at the Commerce Department are fine moves for helping untrap the government’s vast stores of data. But FOIA performance is a powerful indicator of how open the government is with respect to information people demonstrably want.

2. Get serious about not wasting money. $124 billion in improper payments for fiscal 2014. That’s two years worth of Overseas Contingency Operations budgets. Three years of operating the Homeland Security Department. Four years of the Energy Department. It’s around $350 for every American. The administration deserves credit for diligent efforts over the last few years to push improper payments down. But it’s like trying to suppress in your hands a balloon that’s connected to an air source.

3. Remind high (and low) officials to think before they act. A secretary of state used a rigged-up server to do four years of federal business then erases the whole thing. The deputy DHS secretary is found by the inspector general to have improperly intervened in staff work regarding visa clearances, on behalf of politically connected-individuals. A member of Congress spends $40,000 of somebody else’s money decorating his office. The Justice IG can hardly keep up with all of the misbehavior at law enforcement agencies. Not all the people in these episodes are bad or evil. Alejandro Mayorkas contends that, in the case of the visas, he was expediting stalled applications. He has a distinguished record of public service, but golly, I wish he’d stopped for just a sec and looked at the expediting from a poor taxpaying schlub’s point of view.

4. Stop writing badly-worded laws. Like the VA overhaul bill that gives veterans living more than 40 miles from a VA facility the option of using private health care. Congress wrote in a provision telling VA to use geodesic measurement, meaning a 40-mile radius drawn by protractor around each VA facility. But people don’t drive like a crow flies, as Deputy Secretary Sloan Gibson pointed out at a hearing. The whole thing made VA look goofy. It bewildered veterans. And it limited the utility of an expensive program. Now they’ll use online maps to calculate 40 miles even though that’s not really what the law says. Sloppy.

5. End backlogs. Good service means speedy service. Veterans Affairs has a first-time-claims claims backlog of about 245,000. That’s a sharp reduction from its peak, but it’s not likely to disappear, even though the department has promised a zero backlog by the end of the year. Social Security’s disability claims backlog runs close to 1 million. The Patent and Trademark Office, the backlog runs to more than 600,000. The people handling all of these claims aren’t lazy or incompetent. But they’re working in a system that makes them look that way.

The administration favors challenges and crowd-sourcing of ideas. Here are five persistent problems that, if rectified, would significantly increase faith in the competence of the government, and by extension, the people who work for it. These conditions persist not because government employees are bad or don’t care. It’s because they work in a culture that avoids risk an makes easier to say no to an idea than it is to push it through to completion.

Federal open source software activities are growing

March 21, 2015 Leave a comment

Patricia M. Loui-Smicker of Hawaii was confirmed by the Senate, just the other day, as a director of the Export-Import bank. Not the kind of routine confirmation that makes the news. Gilberto de Jesus of Maryland withdrew his nomination to be chief counsel for advocacy at the Small Business Administration. The Senate Committee on on Homeland Security and Governmental Affairs reported favorably on a bill “to reduce the operation and maintenance costs associated with the Federal fleet by encouraging use of remanufactured parts.”

If, like me, you sometime pause in wonder at the majesty and trivia of the federal government, a universe within the universe, then you may from time to time like browsing the daily digest of the Congressional Record. Thanks to the simple and fast web site, operated by the Government Publishing Office, you can zero in on a particular day between yesterday and 1994 in a few clicks. GPO says that since FDsys became its distribution system of record in 2009, some 1 billion documents have been downloaded.

Now, the GPO has made portions of the code that controls FDsys available on GitHub, one of the more active sites for open source software communities. Among the initial chunks is FDsys Collections, which GPO describes as a specialized parser using regular expressions to extract relevant information from source documents. The collection provided covers congressional hearings. More will follow.

GPO is not alone among federal agencies contributing to GitHub. The General Services Administration has put in lots of code it says is useful for building mobile apps.

It appears the adoption of open source by the federal government is growing.  In its 2012 digital government strategy, the Obama administration stressed open data and public application programming interfaces leading to better online services, particularly mobile apps. The strategy doesn’t explicitly ask agencies to make their code public, nor to join open source communities. It does, however, call for a “new default” of exposing federal data and web APIs publicly. Although some of the strategy has been fulfilled and some not, maybe open source application code is the next logical step the strategy might have taken.

Some of the vendors sense this.  I spoke recently with Jarid Cottrell, an open source expert at Booz Allen Hamilton. BAH is planning several moves in federal open source, in the manner in which the company pushed cloud computing early on.

Contrell says the main advantage of open source code is not so much that it’s distributed free initially, but that that it fosters communities of developers. They continuously develop and improve it. And by having so many eyes on it, open source code tends to be freer of cybersecurity weaknesses, its proponents say. Open source can potentially change the economics of whole classes of software when communities create free or low cost versions of functionality otherwise dominated by expensive, proprietary software.

Open source is not free of effort. Once a team develops code for whatever function the organization needs, the code must be compiled into a runtime package and thereafter maintained. That’s why for-profit companies have formed around open source, among the oldest being Red Hat. It and others sell value added support and training services around code that is open sourced. Or they acquired proprietary products, release the code under various non-revenue license agreements, and sell value added services. That the code is available for anyone to contribute to makes the open source subscription model fundamentally different from the licensing models that apply to proprietary software.

Both models have a place in computing.  But I sometimes wonder whether super-popular proprietary products like Microsoft Word or Excel might have spawned more secure and less annoying versions had the company chosen an open source approach.

Booz Allen has created its own open source community called Project Jellyfish. It’s devoted to software for cloud connectivity and cloud brokering, which may be the next hurdle for federal agencies looking to put more data and workloads into clouds. Cotrell said Jellyfish is just part of a larger open tech initiative BAH will launch over the next few months. It will do its consulting gig, helping federal agencies understand open source. It will incubate open source communities around cloud and mobile. And will build more apps and share the code, as it has done with Jellyfish.

The White House may now be a source of acceleration for use of open source tools in federal systems. The newly appointed White House director of Technology, seems to have solid open source creds. He’s been a proponent of the OpenID Connect authentication technology. CrunchBase lists him as once being the president and director of the Open Web Foundation, while the White House blog about his appointment mentions open source work while at Facebook.

New contracting rules published last week by the General Services Administration make it harder for software vendors to tie up agencies with self-renewing licenses and support/upgrade packages. That plus the need to squeeze cost out of maintenance and operations will also give agencies impetus to at least look at open source. You can find open source software for nearly every function, from hypervisors to text editors.

Will Somebody Tell Me Why We Need Net Neutrality?

March 6, 2015 1 comment

I join the crowd of those totally mystified by the FCC’s plunging headlong into Internet regulation. Sixty percent of the FCC commissioners, that is. I have two objections, one procedural the other substantive.

Even if you think net neutrality is the best thing since the Princess Phone, the way the commission did its work hardly seems open and transparent. There’s a five-page summary at, but the estimated 300 pages order hasn’t been posted. Chairman Tom Wheeler says the commissioners got millions of comments “overwhelmingly…in favor of preserving a free and open Internet.” I doubt they were in favor of regulating the Internet as if it were the switched-circuit, monopoly telephone service, 1934-style.

Then there is the idea of net neutrality itself. One young person who free-lances at Federal News Radio simply assumed: Well, the Internet is important, so shouldn’t the government protect it by regulating it? If that’s what the so-called digital natives think, we’re all in trouble. I explained that I remember when a thrilling innovation in telecommunications was installation of an “extension” phone upstairs. Before that, someone would have to run downstairs to answer the phone in the kitchen or front hall. Eventually we got Touch Tone (on which my friends and I would play melodies until the “tilt” signal emanated from the earpiece). Then came modular jacks so you could replace the tangled cord on a telephone yourself.

I also made the analogy to airline service. When it was regulated in the so-called Golden Age you could fly first class, tourist, or student fare. Relative to before airline DE-regulation occurred — with the blessing of a Democratic administration — 10 times more people fly at about half to a 10th the cost, inflation and population adjusted.

Yes, I know. Planes are crowded and in-flight service stinks. Well, not if you can afford business or first class, especially overseas. You can’t? Neither can I. Get over it. A lot of those West Coast Hollywood and Silicon Valley campaign donors and net-neutrality pushers don’t even fly first class — they fly in private jets. So would you rather put up with 8 hours of discomfort on a $1,000 round trip fare to Paris or not go? Or go, but pay the pre-deregulation, inflation-adjusted price of $10,000? This 2013 article from The Atlantic shows just how far those prices have fallen. In 1974, it was illegal  for an airline to charge less than $1,442 (inflation adjusted) one way from New York to Los Angeles.

You still can talk to elderly people who rush off of long distance phone calls because they have vestigial instincts of when those minutes added up. A 1960s Bell System ad touted only 25 cents a minute on weekends. I recall one GSA official in the 1990s promising that the FTS 2000 contract would get long distance rates to under a nickel at some point in the future. That was when deregulation was still just winding up and before IP telephony, wireless and all the rest. Now a $10 a month landline gives you unlimited long-distance.

Ironically, people feared the deregulation of both phones and airlines. But both actions spawned unimaginable innovations. They brought unheard-of products and services to millions. Airlines were deregulated in 1978, the government-sanctioned phone monopoly in 1980. Thirty five years later, and we’re all experiencing an ongoing arms race among carriers, content providers, phone manufacturers and software developers to bring better and faster and more innovative products and services.

So what is the problem? Why would the administration push an allegedly independent FCC so hard for the “strongest possible” regulation of something that is such a roaring engine of jobs and wealth and innovation? Who asked the federal government for this? The FCC majority says this is to protect the future openness of the Internet. But you have to question it when any set of regulators gives itself immense power over something unfettered in the name of keeping it unfettered, and promises if won’t use those powers very much in its “modernized, light-touch approach.” As Anton Chekhov is often thought to have said, if there’s a gun on the wall in the first act, in the third act it’s going to go off.

The issue isn’t regulation per se. Some things have benefited from regulation. I’ve flown two million miles or so in my career. The explosion in air service, planes, traffic — none of it could happen without one of regulation’s great successes. The cooperative arrangement between the FAA, airplane manufacturers, carriers and other responsible parties has produced a logarithmic increase in safety since lumbering, prop-driven airliners ruled the skies. When those planes plowed into one another regularly, they made a good case for new rules and the technology investments to back them up. That sort of regulation doesn’t sound like anything the FCC is selling, with words like “just,”  “reasonable,” “addressing concerns” and similar vagary. Ultimately it cites a 1934 law that, 35 years ago, was found to be stifling communications technology and services. Now that law is invoked to box in the greatest communications innovation since moveable type.

Will CIO Tony Scott Make Feds Eat Dog Food?

February 15, 2015 Leave a comment

The most interesting fact about the new federal CIO, Tony Scott, is that he owns a Cirrus SR22 airplane. At a half million dollars a copy, the SR22 is popular with enthusiasts and even small commercial operators. It’s faster than a lot of other single-engine planes, and it’s made of modern composite materials. But it’s not in the class of single-engine planes with turbine engines and near-jet speeds. Those start a more than a million dollars.

One more thing. The SR22 has a built in parachute so that should the engine quit, the plane can still make it to the ground with the pilot and passengers in one piece.

All in all, not a bad metaphor for federal information technology. It should be up-to-date, perform well, come in at reasonable cost and have a safe way to bail when projects don’t go right.

Since the appointment, all of the trade media have dutifully trotted out the usual parade of experts, giving their opinions on what the new CIO has to do first, what his agenda should be, what he must accomplish in two years. But the White House was pretty clear about what it wants Scott to do: drive value in IT investments, delivering world-class digital services, and protecting federal IT assets and information. In other words, stopping the billions spent on failures, getting online services up to date, and cybersecurity. Pretty much what federal CIOs and their equivalents going back 20 years have been trying to do.

By the blog accounts, Scott left Microsoft in 2013 amicably. For example, Geekwire‘s Todd Bishop reported at the time that Scott left because of family needs following the death of his father, and because he wanted time to complete his instrument rating for his airplane. One tidbit from that post gives a clue to how Scott might approach his White House job. To wit, dogfooding.

These things can be murky, but Microsoft is apparently where the term, eating your own dogfood, first came into the high technology industry in the 1980s. When one group develops new software, Microsoft employees use it internally to see if it’s any good. The practice continued while Tony Scott was CIO. (Aside: That does raise the question of how in Hades the company felt it was safe to publish Windows 8.)

How could dogfooding help federal IT? Consider how the federal IT market has bifurcated. On one side you find the apps and data people represented by the Presidential Innovation Fellows. This side focuses on the quick developments, using agile techniques and multiple data sources. Example: SmokeFreeGov, which is a web site, app, and Twitter handle all at once. Up the food chain a bit is something like the Labor Department’s Enforcement Database. This pulls information from five databases, each developed separately, so they are more searchable and useful.

On the other side are the traditional large-scale developments driven by program managers that still take years and billions. Case in point: The next generation of electronic health records, torturously pursued by the lumbering Veterans Affairs and Defense Departments. Maybe Scott can help each side adopt what it good about the other. Serving each a little taste of the other would be a great strategy. Send a starry-eyed Fellow over to the VA CIO shop for a few weeks. And send some DOD health functionary over to a few hack sessions.

This wouldn’t be dogfooding in the classic sense, more like giving the dry kibble eaters a taste of canned, and vice versa.

By the way, if you find dogfooding unpalatable, so did Tony Scott. Thanks again to Todd Bishop, we know from the Puget Sound Business Journal back in 2009 that Scott replaced the term dogfooding with ice-creaming. Feds, pick up your spoons.