[pdeuchler]

>> "It is also a story of an Obama Administration obsessed with health care reform policy but above the nitty-gritty of implementing it. No one in the White House meetings leading up to the launch had any idea whether the technology worked."

I used to be this guy. The guy with the lofty ideas, but who thought the implementation was "beneath me". The guy who would sit around, waxing poetic about various features, user acquisition, header alignment, etc. Don't get me wrong I had serious technical chops, but fixing that annoying localization bug? Blegh. Form encoding off? Don't wanna get my hands dirty. I had "big ideas"! I was going to change the world! People who change the world don't do the dirty work! So I'm very empathetic to the Obama Administration.

Like them, I needed a real wake up call. In my case, a friend who had implemented an idea I had sold the software for a lot of money. When I confronted him about sharing the profits he started running git blame on files across the project. My name came up maybe once or twice, across a multi-k LOC project, and even then on nearly inconsequential lines. It hit me then that while ideas may have value, the implementation usurps all of it. An idea alone is powerful, but once it's implemented the idea becomes worthless. At that point it's all about rolling up your sleeves and getting shit done. When you focus on that problems like an "ID generator" becoming a bottleneck (I had to read that bit several times over... apparently I need to start raising my rates to the hundreds of millions) disappear. It's a hard lesson I had to learn, and it's one the Obama administration has hopefully learned as well. Of course, I had just turned 17 when I learned my lesson, and Obama is now a lame duck with less than 2 years left on his final term. I guess this exemplifies my greatest struggle with the Obama legacy, in that it has become one defined by squandered potential.

[JonFish85]

>It hit me then that while ideas may have value, the implementation usurps all of it. An idea alone is powerful, but once it's implemented the idea becomes worthless.

You need both. Implementation people aren't all that useful without a vision. A vision isn't all that useful without implementation. It's a symbiotic relationship.

Steve Jobs without a Steve Wozniak probably wouldn't have been as successful. Steve Wozniak, without Steve Jobs, probably wouldn't have been as successful. But the two put together made some great things happen.

Same with the moon landing. As much as it took a tremendous engineering effort to put a man on the moon, it took someone with the vision and power to make it all work. JFK didn't get involved in the details, I'm sure, but he really helped to set the tone of the whole effort.

There's a fine line between looking at implementation as "beneath" your position and knowing when you're being more of a hindrance than a help. In my experience, nothing has driven me more crazy than a person above me who, while being a great project manager or whatever, tries to get involved with things that end up hindering the effort. If you take a weekend course on programming in Java, that's great! But don't start giving out "helpful" tips in something that is not your domain.

[saraid216]

> You need both. Implementation people aren't all that useful without a vision. A vision isn't all that useful without implementation. It's a symbiotic relationship.

37signals' take on this was the best one I've seen (paraphrased):

Ideas are a multiplier. If your execution is good, a good idea will multiply that. If your execution is minimal, a good idea will scrape by. If your execution is abyssal, a good idea will make it worse.

I'd link the actual page in the book, but I have always had trouble finding it.

[alexhaar]

Derek Sivers talks about this idea as well: http://sivers.org/multiply

[saraid216]

Actually, that's exactly what I remember. I think I have been misattributing it this entire time.

Damn it. Thanks. :P

[abecedarius]

Jobs and Wozniak don't match this discussion very well: Woz was responsible for both the idea and the implementation, to first order. It's true that without Jobs there wouldn't have been a company.

[jbooth]

You're not wrong, but anyone who actually has an office in the physical whitehouse, or anyone within, say, 5 layers of the President, actually shouldn't be concerned with the implementation of the technology. They should be focused on getting money and resources to the people who do the tech and running interference for them to help prevent them from being sabotaged by political opponents. In between their other million non-technical job responsibilities, many of which involve a lot of legal, policy and political details.

That said, obviously they should have brought in the campaign website team to oversee this instead of doing the standard federal government contracting route, which leads to a late, over-budget project literally every single time.

[jowiar]

> That said, obviously they should have brought in the campaign website team...

That's called cronyism, and, for the most part, it's either against the law, or will get you destroyed in the media. One of the biggest challenges in government is that, for those of us in startup-land, the process goes something like: "I need to hire someone, I'll call my friend who I know does good work". You can't really do that in the government, or everyone who isn't your friend won't be happy.

The standard federal contracting route is certainly a mess as well. The underlying belief that everything can be reduced to a series of checkboxes, and whoever can check the boxes the cheapest wins leads to a disaster.

Hopefully, one of the outcomes of all of this is we rethink how the government software (and other sorts of procurement) process is done.

[cynusx]

Procurement process design is not only a problem in public bodies, there is a lot of corruption in private companies as well where a procurement person would strike a deal with a sales person. This deal can range from "a nice watch" to "I like a swimming pool", depending on the contract value.

That said, I also think that the current procurement methods are broken and that a less rigorous procurement process would eliminate contractors that specialise in government procurement.

One way would be to implement a randomised audit where the deal would be re-examined and the procurement officer would have to defend every choice he made. if 5% of the deals were audited then bad actors would be caught very rapidly.

An other approach would be to adapt the jury-trial approach to government procurement and have them review every procurement contract.

Third approach would be to combine this to reduce the workload.

[jbooth]

All true. That being said, they got destroyed in the media anyways.

[hga]

"They should be focused on getting money and resources to the people who do the tech and running interference for them to help prevent them from being sabotaged by political opponents."

Can you suggest some effective mechanisms by which "political opponents" have been able to "sabotage" their efforts? Do these people get depressed when they hear an eeeevil Republican saying what they're doing is wrong, or even evil?

[jbooth]

Shutting down the government for a few weeks probably complicated some things. Can always call non-stop hearings and bring people in to talk instead of work, as well.

What, you think nobody plays hardball in DC?

[hga]

For the former, we've heard that nobody important stopped working during the "shutdown". The latter was entirely a hypothetical ... before the disaster. After, I don't recall anyone truly important being detained in hearings for long (more than a day? with presumably 1-2 days of prep time at most before?), and the Administration has had no compunctions about not sending people to Congressional hearings if they don't want to.

"What, you think nobody plays hardball in DC?"

National level Republicans as of late have been playing badminton at best.

That you cannot cite "some effective mechanisms" after making such a broad claim suggests to me that you're approaching this discussion as a political, not a technological, one.

[jbooth]

Shutting down the government because you don't get your way isn't hardball? I suspect if we flipped some D and R labels around, like say this was President Romney's healthcare plan, you'd have a completely different opinion about every detail, up to and including the severity of shutting down the government to try and defund something.

[hga]

Not these faux "shutdowns".

Especially the latest one, where the Park Service went to great efforts, employing a lot more armed Rangers to keep people out of anything they could claim as their turf, including things open 24x7 without corresponding 24x7 coverage, like the WWII Memorial. But not, curiously, a rally for immigration "reform".

Look, I spent a dozen years "inside the Beltway", I know political theater when I see it. It's just that, not "hardball". Heck, they even made the furloughed employees whole, as they have in times past.

[Tloewald]

If Obama's only legacies are preventing (or at least not causing) a second Great Depression and Obamacare (assuming Obamacare works worth a damn) that's not too shabby.

And by my calculations, since we're several months out from the mid-terms, Obama has well over two years left. Not that I'm expecting much from them.

[odonnellryan]

That's interesting, what was the software?

[seivan]

Those exist. They are called UX-"Experts", Interaction "designers" and etc.

[AutoCorrect]

No, an idea is worthless UNTIL it is implemented.