Hacker News Reader

If this guy is such a rockstar at cleaning up the mess other people leave behind, why is he not leading these projects from the beginning? Or, since he is only one person, training and mentoring a staff of people who can manage the various projects in a similar fashion?

It's harder to convince someone to pay you to prevent a problem that hasn't yet occurred. But once it's occurred, it's a crisis, and no amount of money is too much if you can clean it up.

(Think about oil spills or heart disease.)

it'd be nice to have a full team of heroes leading all of the projects mentioned, instead of calling one in after millions of dollars have been spent on a failed system.

In my experience, it usually goes something like this:

1. Developers say 'we need X, Y and Z to make this happen'

2. Managers deliver Y and Z late, and never provide X

3. Developers plow ahead anyway, project is a catastrophe

4. A senior VP is appointed to fix the problem, asks the developers "What do we need to get this fixed?". Developers say, "Well, we need X".

5. The mere involvement of a VP removes all impediments to obtaining X, and the project is completed as originally designed.

Large projects have many gatekeepers (that have feedback, sometimes conflicting), integration issues (I'll get to this one in a second), and scheduling issues (this one seems to be arbitrary based on political deadlines).

Consider the integration issues for this website. Multiple insurance companies, after decades of working in isolation (insurance companies were early to utilize computing), are now expected to bend their offerings into the lowest common denominator between themselves and other companies (so a computer can do analysis on them).

Think about the headache that would be for managers, programmers, and domain experts. All with political taskmasters cracking the whip with unrealistic expectations.

On top of that, bad code means that money might be lost by third parties. This code has high lawsuit potential (no matter how well agreements are written).

My point being, public or private, this project was destined to be a difficult one.

I want to expand on a simple point of government and large corporation programming that might not be obvious to the casual observer. There is a lot of inertia in the system. I've found bugs older than me (in code that runs nightly). I once worked for a company that had code written in the 60s still going (thankfully I wasn't on that project).

Old code isn't going anywhere soon, and trust me, you don't want to deal with it. Even though you can make a career out of it, a lot of effort is needed to make even the smallest changes. I don't care how good a programmer you are, if your tools or environment suck, you're in trouble.

And this is one of those cases where being elite doesn't beat above average with experience. Not even close. I'd much rather the person who can say "We did something similar back in '95, and it caused problems in these places" than the best programmer on the planet. Because for these types of jobs, the best programmer on the planet is usually the person who knows the system best.

Any team of people, no matter how talented, will be spending the majority of their time understanding what is in place.

Note that I haven't gotten into the procurement process, which can turn a difficult project into an impossible one. But that's a problem that also lacks a silver bullet, and is a larger discussion.

> If this guy is such a rockstar at cleaning up the mess other people leave behind, why is he not leading these projects from the beginning?

Because he's usually tapped to lead things bigger than these individual projects.

> We have a strong national military... what's to stop us from having a strong national development team?

Competition from the private sector for talent. If we had as much demand for private troops in the US as for private developers, and private troops were paid as well as private developers, we couldn't afford a strong national military.

(We don't actually need as many developers as troops to have a strong national dev team, so this isn't an insurmountable problem, but if you look at the civil service salary schedule against private sector dev salaries, there is a pretty radical disconnect.)

"Because he's usually tapped to lead things bigger than these individual projects."

Playing devil's advocate here: what's bigger than an overhaul of an entire nation's healthcare system?

> Playing devil's advocate here: what's bigger than an overhaul of an entire nation's healthcare system?

High-level leadership positions in offices whose scope is the entire executive branch not just the healthcare piece: his position for the entirety of the Obama Administration has been Chief Performance Officer and Deputy Director for Management of the Office of Management and Budget (and about a year and a quarter of that he spent as Acting Director of OMB, as well).

Or, being responsible for leadership on the entirety of the nations economic policy -- he's scheduled to move into a position as Director of the National Economic Council on Jan. 1, 2014.

So, basically you want this - http://digital.cabinetoffice.gov.uk/about/

It's worked so far. It remains to be seen how it will age - maybe in ten years time it will be just as buried in bureaucracy as the rest of the offices.

I understand the (very successful) www.gov.uk site was largely done by direct government employees, working for the various departments the website covers.

On the other hand, a number of huge, subcontracted efforts (NHS system, tax IT consolidation) seem to be going much worse.

I don't think this has to do with the skill of the people involved. Rather, I think when the lack of flexibility in huge government contracts meets the demand of a new government to change everything, the project fails.

Yes, www.gov.uk is the product of the new Digital Government Service which I linked above. They haven't been around for very long yet and they are currently only responsible for the government web-services but I hope that their success will eventually pave the way for in-housing much bigger project likes NHS systems.

Rescueing a fail[ed|ing] project takes a very different skillset than managing one properly in the first place. Planning and executing is a hell of a lot easier than firefighting through a storm of finger-pointing and blame-shifting. Once things have gone wrong, most people involved lockdown and stop being helpful for fear of being seen as the one who made it go wrong. Managing that is a rare skill.

Ideally no one would risk projects going awry by employing "average" project managers and then having a single "rockstar" guy to sort out the mess if they do go wrong. It'd be far preferable to have brilliant project managers for every project. But pragmatically, that's never going to happen, so we need the "rockstar".

Does anybody else feel that having a track record in solving beaucracy issues can not really be applied that much in getting a legacy IT system fixed that seems right before cardiac arrest?

about that national development team, IMHO developers don't want to work for the government or on government jobs for a very good reason, hence the good ones are normally in the position to skip such jobs. Quality and management of IT projects is much worse in government projects vs private sector from my limited experience.

> We have a strong national military... what's to stop us from having a strong national development team? I'm 100%-for private industry and letting private companies bid and fight to build the best product for the job ...

How would you then fix the obvious misalignment of interests? Party A gives party B $X to perform a job. It is in party B:s best interest to spend as little effort as possible on the job, and/or to maximize $X. Even if B is staffed only by rockstar programmers who do amazing work, their incentive is to work as little as possible and introduce as many bugs as possible be because they get paid for fixing bugs.

Only way I can see to solve that is for the governments to do their own software development in house.

>We have a strong national military... what's to stop us from having a strong national development team?

not having another $600-700B/year to spend?

Considering one would improve local healthcare and the other one is designed to destroy the health in remote places... the priorities on the existing $600-700B/year could be adjusted. Unless the global health of the world is treated as a zero-sum game...

not true, the debt crisis is a myth, and the more efficient functioning of government services across the board could end up paying for itself

I really want to know why Sebellus still has her job. If as she claims the President did not know of the issues until after launch then why wasn't she informing them of the situation. If he did, then why did he permit this to occur?

It is beyond belief that the president was not aware of the problems. This is THE centerpiece of his presidency.

You think because he got the legislation passed he was elbow deep in it's implementation?

Do you think Nixon was personally testing drugs or testing water supplies, or Reagan was writing software for the Star Wars project?

He'll get status reports in the same way he gets status reports from each department. Implementation details aren't passed up the chain.

If anything, they are only lied up the chain.

This is landmark legislation, he goes around the country campaigning about it. Yes I damn well expect him to be fully aware of what is going on with it. We are talking about one of the biggest changes in how America operates in nearly forty years, if not more.

What irks me more is the refusal of officials to even provide real numbers on applicants and successful enrolls. Really guys, you work for us, quit the lies; and they wonder why so many Americans do not trust their government.

> If this guy is such a rockstar at cleaning up the mess other people leave behind, why is he not leading these projects from the beginning? Or, since he is only one person, training and mentoring a staff of people who can manage the various projects in a similar fashion?

My guess is that there's a lot more money in his line of work than the other.