Americas New CIO Wants To Disrupt Government and Make It a Startup
An anonymous reader writes "America's new CIO Steven VanRoekel wants to revamp the federal government and make it as agile as a startup. But first he has to get rid of bugs like the Department of Agriculture's 21 different e-mail systems. From the article: '“Too often, we have built closed, monolithic projects that are outdated or no longer needed by the time they launch,” he said. As an example, he mentioned the Defense Department’s human resources management system. Dubbed the “Defense Integrated Military Human Resource System,” the project was meant to take seven years to develop. Instead, it took 10, cost $850 million and had to be scrapped after 10 years of development in 2010 because it ended up being useless.'"
Everyone today wants to be "disruptive". What will end up happening is this CIO will end up creating yet another useless system that is over budget and no one wants. But for 10 times the cost, because it's "disruptive".
Seven puppies were harmed during the making of this post.
That or a lobbyist group behind a specific software group will "donate" money to anyone that can nullify his plans. And since companies are allowed to donate unlimited funds, there is little hope for his plan.
Restore the madness of youth's lechery
A lot of things in the Federal government seem wasteful until you realize the politics behind how they came to be that way. "Why do you have this facility way out here, when it would be cheaper to move it closer?" often doesn't elicit a "Because we're wasteful and stupid" response so much as a "Because we need the support of powerful Senator X and so we built it in his state" response. NASA is notorious for that sort of thing. Almost all of their contracts go to very politically connected contractors with powerful Congressional backing.
That “Defense Integrated Military Human Resource System” was a Northrop Grumman project. If the name Northrop Grumman doesn't mean anything to you, you don't know jackshit about federal politics, or how things REALLY work. Northrop Grumman owns Congress.Tthey have facilities in virtually every state.
SJW: Someone who has run out of real oppression, and has to fake it.
Here's an idea, why don't we just shut down 20 of the 21 sections of the Department of Agriculture so they only have one email system?
We can keep food safety inspections, at least until an adequate private inspection regime is in place (like the one that inspects food and facilities for Kosher and Halal dietary requirements).
Forget trying to make a several-million person organization act like a startup.. I'm not even sure what that would mean in this context. Honestly, sounds like BS. Let's talk about this:
“Defense Integrated Military Human Resource System,” the project was meant to take seven years to develop. Instead, it took 10, cost $850 million and had to be scrapped after 10 years of development in 2010 because it ended up being useless.
Let's say I go to Procter&Gamble, and offer them an HR system. I say to them, it will cost $30 million and 3 years. Then after 3 years, I try to bill them $40 million and say it will take another 2 years to deliver.
I'm pretty sure that's when either:
1) PG sues me for breach of contract, and refuses to pay anything
OR
2) the person at PG in charge of this project gets fired for improperly managing the project (changing requirements, etc).
WHY is that gov contracts never do 1 OR 2. They do just pay it?! Seriously, WTF
That's not something any company would do. Startup or otherwise.
Hmmmm... Most start ups fail and end up collapsing completely within a few years!
Just thought it worth pointing out! ;)
"That's the way to do it" - Punch
Stop letting timelines slip and costs rise. Bring some of the work in house instead of letting contractors rape you. I can get rid of those 21 email systems right quick. Build the new system, migrate folks to it. No user input, no predetermined time table, just a phone call telling them their mail has moved.
But aren't most startups woefully underfunded and don't 97% of them fail? Oh, wait.....
He does not belong in the government... to much logic with this one.
I'm not sure if I'd call having twenty-one different email systems a bug, but it is definitely inefficient. A bug is something that is an error in a program, not an error in the implementation of a program.
I run Ubuntu skinned to look like a Mac on a PC. Go figure.
From TFA:
So ... "disruptive" and "crowd source". Any others?
So the crowd sourced plan will be based on open standards to achieve maximum disruption.
Yes, mobile apps that are crowd sourced should be built on open standards to achieve maximum disruptionability.
Seriously, if you think that people WANT government to be so involved in their lives that they NEED an app to handle their DAILY interaction with it ... fuck you.
He's a CIO that's spouting buzz words.
Often we see people who failed in business try to get into politics. It's time to stop this -- government is not a business.
Let's find people who understand government to run ours.
There's no -1 for "I don't get it."
I didn't realise the Americas were so in sync that they shared a single CIO for both continents.
Spelling mistakes, grammatical errors, and stupid comments are intentional.
So, what he proposes doing is taking 21 systems that currently work, and replacing them with something that, based on history, won't work?
Good rule of thumb - even if it looks inefficient, if it works, LEAVE IT ALONE!
After you've fixed everything that does NOT work, then you can start streamlining the things that work but aren't as efficient as you might like.
"I do not agree with what you say, but I will defend to the death your right to say it"
And having working at a startup, do you know what is the next step? You are correct, outsourcing to China.
So... as agile as say a Google? Which Google? Google the brainstorm of a handful of guys or Google the mega corp with offices all over the world? The agile startup might have been more sexy but it only was capable of things in potentia. It had potential, that was realized as it grew. The snow flake that falls is not an avalanche. Neither can it become one. It can cause one but the moment the snowflake has started on the path to an avalanche it has seized to be a simple snowflake.
I can whip up a fairly complex website for say a job site but the moment it needs to scale I will need more then myself, the more it needs to scale up, the less flexible it will become. Even if the code itself will remain flexible, the support structure around it will become by its growth less flexible. A oak sprout can easily bent but it can only become a great oak by sacrificing its flexibility for sheer size. Then it doesn't have to be flexible anymore to survive being stepped on. Few things can step on a 1000 year old oak.
Also, how agile do you want government to be? Agile means fast, do you WANT government to do its requisition process fast? With no procedures to investigate, file complaints? Nobody told early Google how to buy its hardware but experience has shown that when big orders are involved, oversight is desperately needed and oversight is low.
Government is slow and inefficient because it involves a lot of different interests.
And what is the alternative anyway? For every Google there are a hundred FAILED startups. Good luck explaining that to the voter, the government funding a 100 different projects and getting commercial results of 99 of them failing. See recent fallout over funding for electric cars and solar panels.
And big business? Lets see, the American car industry failed miserable and needed the state to help them out... but no doubt republicans will say that came because of all the regulation. This is proven because regulations were removed from the finance industry and they... oh wait... they failed even more massively didn't they. Gosh... private industry small startups fail left and right... big business fails left and right... compared to that, the state ain't doing so badly.
Anyway, think very carefully what you wish for when you wish for an efficient state. The most efficient form of law and order is to simply kill any offender for any offence. No lengthy trials, no costly jails, no rehabilitation with a near perfect failure rate.
There were efficient governments in the past. People fought tooth and nail to get rid of them.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
And having working at a startup, do you know what is the next step? You are correct, outsourcing to China.
Come on, not all go to China.
You forgot India.
Standards.
And assuming he wants to make it "like a startup" that means small unbureaucratic groups, shoestring budget, and likely to fail. Good luck.
Don't think of it as a flame---it's more like an argument that does 3d6 fire damage
This is the Apple way, and there is some merit to it. If you let people have whatever they want, you'll find you have a lot of incompatible requirements. If you give them something that works, they will find ways to do what they need to do, and in the end they'll spend less time futzing with the little known features they originally wanted. It will also significantly reduce the cost to support.
I scoffed at this way for many years, but now that my hair is a bit grayer I've learned that often the simple tools are the best. Having one system that does everything is very cool, but often it's not practical to build it or economical to maintain it.
Is it just my observation, or are there way too many stupid people in the world?
More power to him, if he can make it happen. That's a big if, though. It's easy to throw around words about how the government should be, but making that actually happen is a different story.
If the constituents of Senator X benefit from his demanding that it be built in his district before he'll vote for it ... then he's doing a good job for his constituents.
This is only "waste" when people outside of his constituency look at it. And only then because it does not directly benefit them.
Which is why people are pissed at "Congress" but the re-election rate for Representatives and Senators is so high.
Get rid of the "bad" people in Congress who are grabbing pork for themselves and their districts ... but keep our "good" Congress Critters who are looking out for the best interests of our district.
Government IT projects usually end up too big to succeed. The other issue is that computers make processes too efficient, and government departments never eliminate jobs.
Sorry, the next step, statistically, is that you fail. Most startups are failures. It's a risky venture. I think this is the wrong approach. It's just political theatre anyhow.
Currently hooked on AMP
I studied such as system in the IRS in school, and have worked with some DoD systems live. One problem is too many stovepipes, often dozens. All the data and business processes have to be integrated into the main system without an interruption of service.
To make it harder, the business processes are often convoluted and the data isn't normalized or even clean. I have seen, literally, layman-made (as in "some dude in the office knew Access and put this together") Access databases holding important information for tens of thousands of people. If the data is about people, even a 1% error rate in conversion means thousands to millions of people complaining. Imagine your tax record is one of the problem records, how it could screw up your life.
To make it even harder, add the political/contracting component, and often powerful users resistant to moving to a new system.
The headline for this item plays into something that's very dangerous in the long term. This guy isn't "America's new CIO." He is the CIO for a bloated, inefficient bureaucracy that runs the GOVERNMENT of this country. He has no power or influence over the country itself. People frequently indulge in the fiction that we elect a president to "run the country" -- and that leads to people having insane expectations and an insane willingness to turn power over to one man. Calling this guy the country's CIO is a small manifestation of the same mistake.
This is the see-saw private industry has been on for 50 years. Do you make each unit independent and agile with its own all-powerful General Manager? Do you consolidate similar support organizations (IT, finance) to HQ thereby giving up uniqueness in favor of standardization? Having spent a lot of time with Mgmt Consultants, I can assure you the current kick is towards consolidation. In 10 years, the consultants will be telling us each organization needs the customization which is only capable by rolling out 20 agile, independent installations. I imagine that this CIO is spending a lot of time with IBM guys with dollar signs in their eyes and pushing their make-work agenda.
What's hilarious is that everyone pretty much understands you give up agility by consolidating back-office functions. The tradeoff is hopefully more cost savings and perhaps better quality/standardization. Saying it will be MORE agile is pretty much a bald-faced lie.
Most startups spend more then they take in and then finish by going bankrupt. Maybe the federal government is already a startup.
Lots of people believe this type of behavior and campaign contributions are bribes. I think they are more like extortion payments.
Give us a cut or nothing gets done.
Except for ending slavery, the Nazis, communism, & securing American independence, war has never solved anything.
Most startups fail.
Awesome furniture, accessories and cabinetry in Santa Rosa, CA: http://humanity-home.com/
He needs to synergize the efficiencies of the current group dynamic to maximize ROI within a mobile framework
of outsourced in-scope cloud computing over the coming disruptive quarterly strategic marketing blitz.
Service guarantees Citizenship! Questions Guarantee GITMO.... Amerika Uber Alles!
sure, but we're angry they're not pissing money our way
Government is for handling the jobs that business is not efficient or effective enough to handle.
Business is good for running things that can turn a profit in a competitive environment.
Do not confuse the two.
Have to eliminate the government's owners and operators along with those massive agencies, otherwise they'll just buy and corrupt what's left.
I've seen this sort of problem before in bigger organizations before - many branch offices run like their own companies, have their own data center (a bunch of servers in a cube).
Granted its a bigger problem in public institutions mainly because good technicians who know how to setup top level IT services like centralized email services and the authentication/directory systems tied to them are working at places that pay better.
Having worked for the State of Oregon - its quite common here, but getting better (because there are a lot of really qualified IT people who can use work, and are willing to work at lower wages the State pays).
Fixing this doesn't mean "disruptive, startup like development" - it demands someone centralize authentication and identity (that would be the hard part really), cleanup namespace collisions that are inevitable with merging 21 email servers, setting up aliases and mail routing so stuff doesn't get bounced from deprecated domains, and migrating all that mail to a new cluster of machines. There - I made a plan for some enterprising new project manager for the department of agriculture.
But where will all the retirees work after they are determined to be useless? Isn't a government job just suppose to be a part-time retirement home?
Worked in a start up for a brief while. It really was a whole other world compared to the average cubical farm IT office. It's a bit more than installing some retro arcade machines, designer couches and having a bar serving liqour all day. It's all about people. There was a certain kind of person in the work place and the work place was conducive to a certain kind of creative think-on-your-feet attitude. Without all the process and procedure of a corporate IT, there was a lot less paper pushing and a lot more getting stuff done for the client. The big boss even had a "making work is not making money" policy and encouraged sparing use of conventional administrative process. Everyone was motivated, stuff got done. It all was a bit of an ad hoc mess that would not scale well to a larger office with some adjustment, but it was bloody brilliant.
If you want to have your large enterprise or government as agile and efficienct as a start up you need a complete overhaul of how people think and act in your organisational culture. It's not impossible but it's bloody difficult, as you have to throw out 90% of how everything is done right now.
After logging in slashdot still does not take you back to the page you were on. It's been that way for 20 years.
We worked with the government and industry partners to develop a federal secure messaging standard called Nationwide Health Information Network (NwHIN) Direct. For this initiative, which the Office of the National Coordinator for Health IT (ONC) announced in early 2010, we were able to go from inception to production in less than a year. For a federal IT standard, that has to be a record. The reason for the pace and success was an open source approach to the problem.
I hope we continue to see more of this.
Indeed. Someone got $850 million in 10 years. Sounds like a good definition of success to me.
You are in a maze of twisty little passages, all alike.
Let's find people who understand government to run ours.
Those people are called "lobbyists" and they already run the government, because ours is a system where corporate executives and government policy makers cooperate for mutual advantage. And so long as there is a financially rewarding level of political power out there to wield, this will continue to be the arrangement.
Space game using normal deck of cards: http://BattleCards.org
How about this for an idea, Create a Raft of Open Source Projects ultimately representing 99.99% of the operating software upon which the government will run. Implement it for each of the Governments many departments resources. Have them all sit on an Open Source Information Framework which efficiently allows the vast government bureaucracy to interact and interrelate with ease and simplicity. Have that resource designed to easily provide transparency, accessibility and communication with the Citizens of the United States.
Close the 0.01% of the government's operating software to develop a security application which is proprietary and provides the government with the ability to protect national secrets and critical national infrastructure (it is a worthwhile endeavor to protect key pieces of national infrastructure from cyber-attack or malicious hacking.) Do this activity last (up to that point fire-walling and isolating classified material and resources from the rest of the system) and have the developers for that software come from the pool of "Best of Breed" proven developers from the Open Source first 99.99% projects. Then once implemented, create a small nonpartisan committee whose job is to make certain that national secrets are indeed national secrets and not just cover-ups for congressional and/or executive misconduct.
Finally, pay the top 10% of the developers and managers with increasing tax incentives for their contribution.Then when US-GOV 1.0 is released, publish the names of the top managers and developers as national heroes, and hold a PBS televised gala at the Kennedy Center in their honor for their patriotism and contribution to all Americans.
You'll save a couple hundred billion dollars, get the job done in one percent of the time, have true transparency in government and be able to endlessly improve the operating software (as it should be), as technology improves and new talent arrives to take up the challenge. Best of all you connect the government with the people, democratize the government's operation, allowing the people to fully participate in its function and performance at the level of infrastructure.
I can hardly imagine a greater opportunity, or a better way to accomplish such a sweeping social project. Just as a side note, pick a ring master for this circus with a little experience, a man whose already accomplished something comparable. Perhaps someone whose first name is Linus?
"Monolithic." I don't think that word means what you think it means.
I thought we already had Government 2.0.
Is this a new release or just a bug fix?
Americas New CIO Wants To Disrupt Government and Make It a Startup
In other words he wants the VCs to take over and run the place into the ground, cook the books, sell out, and finally retire to a private island.
Rare to see such honesty from a man in government. Sounds paleo-conservative since thats how the govt has been run all my life...
"Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
and or many systems doing the same thing and Consolidation is hard to make it work right. When working with Big implementations it's very easy to hit road blocks / cut corners and fail to plan for all out comes.
Also contractors / sub contractors add lot's of over head and buck passing that just slows stuff down even more.
Look at comcast they are made of many other systems that over the years be came part of one big cable system and even then things are very differnt in each region so it's easy to say we can save by Consolidation but you can try and fail, Do it and then find out things worked better the old way, and end up just pushing it back when the first round of test roll outs fail.
Or you make a good product with a single killer feature, establish a decent customer base and get acquired by a larger monolithic company.
'tis but a scratch.
I want to see him synergize the potentials of win-win scenarios to maximize ROI on buzzword ideation... just like a real CIO. Obligatory Dilbert, and excuse me while I vomit for a little while.
Those who can make you believe absurdities can make you commit atrocities. - Voltaire
America's New CIO is a Buzzword-Slinging Douche
So, Steven VanRoekel wants to run the federal government's IT infrastructure more like a startup.
Is he aware that the majority of startups fail?
On top of that, Federal worker bees have been under a pay/hiring freeze for several years, and generally do not receive competitive pay even in good times. To make up for this, they do have stronger job security than their counterparts in the private sector. But this is antithetical to how startups work. With startups, everyone knows going in that it is a high-risk, high-reward proposition. Our federal government cannot and should not operate that way.
the project was meant to take seven years to develop. Instead, it took 10, cost $850 million and had to be scrapped after 10 years of development in 2010 because it ended up being useless.'"
soooo just like in corporations then?
The Kruger Dunning explains most post on
Ok, then here is the simple and easy question: How you stop either of the 2 big parties from getting Ca 50% of the cotes?
Most of the people in the goverment stays in the goverment because the amount of seats barely changes.
Protip; Unless there is a political reform, you CAN'T change those seats, because nobody in poltics is accountable, and nobody of the people voting has a memory lasting longer than half a year.
Disruptive would be telling the IP Lobby to GTFO. But that won't happen, so he's not really planning on being disruptive.
My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
what's wrong in using dreamhost for email?
Just put the whole Government on Google Apps. Problem solved.
As population increases we need improvements in agriculture, in fact that's the only reason the "population bomb" situation didn't already happen becuase the Chinese and Indians got their agricultural act together. Private enterprise is not enough to improve that situation alone.
Like a startup 'eh?
So we'll be letting any 20-something with a roll of duct tape and an unreasonably high opinion of their "skillz" build our national infrastructure based essentially on ideas gleaned from blog postings and google search results? I guess we'll also be spending billions on smoke, mirrors and fast-talking slick executives in a bid to be acquired by China at all costs?
hot damn! sign me up! I'm an expert at this shit!
Government contractors are public robbery... Not sure what to call what the banks do to us...
These business MBA people are worse than the religious nuts-- unfortunately we never established a separation of business and state! All our big problems today in the world are business linked -- in the past religion was a huge problem for governments so people learned to separate them; I wonder if we'll learn this next lesson??
A real CIO solution:
Create the Computer Core (or some other name, obvious geek squad is taken) with a student worker program, all open source and all free.
Hire the best and brightest at stable and HIGH pay - instead of the going rate with good benefits.
A government non-profit similar to PBS or like the USPS was but funded by charging government offices for their services; it can also run at a loss (since it serves gov and is payed by gov it's at a loss anyhow.)
It would be barred from being forced to pay pensions 50+ years in advance (which the GOP did to the US postal service to create a fake debt so they can try to break it up; the USPS is not in financial trouble, BTW.)
A formal process would be devised for software specifications; because 1 of the big problems with soft dev and government is that the requirements are a moving target! Gov clients would be contractually locked in; no moving targets; in fact, most design work should also be removed from those lawyers too.
Student lawyer group; gain experience suing gov offices for bad contracts and corruption. Tech scams are rampant and tech makes great cover for old fashioned corrupt contract deals.
Since management is MOSTLY the problem in any organization public or private; it would be difficult to design a system by which to attract and keep good management-- one method that seems to work ok is to only promote from within. I also don't see why we can't have democratically run organizations; where the workers get to vote -- as opposed to a top-down authoritarian model we seem to love so much (more each year.)
Your city, your state, your elementary schools--- all have the SAME website needs. Much of their other software as well. WHY do their office PCs need upgrades?? they don't.
Democracy Now! - uncensored, anti-establishment news
I am a developer who has worked at several Agencies in USDA and been thru a variety of e-mail system changes. It's the same old same old. The folks at the top want standardization without wasting any time on requirements. I remember once when they took WordPerfect away because USDA was going completely to Word. Good idea, except legislators in State governments and in Congress demanded WordPerfect attachments, not Word attachments and the legislators didn't really care about some lame USDA memo from some lame USDA CIO wannabe.
Comment removed based on user account deletion
Comment removed based on user account deletion