Friday, October 25, 2013

Morning JOLT

National Review

Today on NRO
JONAH GOLDBERG: With every passing day, it's less likely the ACA's disastrous rollout can be fixed.Obamacare's Ticking Clock.
CHARLES C. W. COOKE: The president is all things to all people, and concerned about all of it. Obama the Avatar.
KEVIN D. WILLIAMSON: Obamacare's cost controls are economic and political fantasy.Obamacare's Doubtful Deficit Reduction.
DEROY MURDOCK: The sinking of S.S.Obamacare will vindicate GOP "extremists." Not So Extreme After All.
ANDREW STILES: What is the administration doing to fix the insurance exchanges? The HHS secretary won't say.Sebelius Ducks.
PATRICK BRENNAN: Medicaid expansion through the health-care exchanges is running into real problems. Also Not Going So Well.
Morning Jolt
. . . with Jim Geraghty
October 25, 2013
So Far under Obamacare, More People Have Lost Insurance than Enrolled
Campaign Spot's increasingly regular contributor, IT project management expert Bruce Webster, writes in again:
Lots of people are going around quoting Fred Brooks now -- both the 'mythical man-month' concept as well as Brooks's Law ("Adding manpower to a late project makes it later.").
But there's something else Brooks said with regards to projects that are running late that is directly relevant in the context of the 'tech surge': "Take no small slips." In other words, if you know a project is going to be late, or if it already is, and you have to come up with a new anticipated release date, you should significantly over-estimate how much time it's going to take to get it right. (In essence, a recasting of the old engineering maxim to 'under-promise and over-deliver'.) It is far better to estimate that you'll need an extra six months and deliver in four or five, than to estimate that you need one or two months and then deliver in four or five.
I know that rule, or a version of it. I'm just used to seeing it attributed to chief engineer Scotty from Star Trek.
Of course, the natural tendency on the part of HHS & the Administration will be to minimize the estimates of how long it's going to take to fix things -- and those estimates will almost certainly be wrong. So what we may see is the 'Never-Ending Story' pattern, where for several months they're perpetually 4-6 weeks away from having working properly.
If I were in charge? I'd pull the plug completely and give no completion date at all until the website reconstruction was at a point where I felt comfortable opening it up for public alpha testing. Based on how the alpha testing went, I might announce a subsequent date for beta testing; and if that went well, then and only then would I announce a planned date to go live. (Here's some background on alpha test/beta test/release:

No comments: