The Waiting is the Hardest Part

It was 11:00 PM on 8 April that the refreshed Bakers Game went to Apple for review.

It is still awaiting that review.

Fortunately I've managed to keep busy.  I've identified a couple of ideas for new projects and selected one. I've pulled down the data necessary for it.  I've got some code written, and the code has spawned newer ideas.  One mega-project has now suddenly morphed into two mega-projects.

And still I find myself refreshing the "Waiting for Review" page.

It's not like this is my first rodeo.  I had one app take two weeks from submission to publication.  I've witnessed apps from other developers taking months.  One week is nothing.  It's a feeble drop in the bucket, especially when you see hundreds of apps going live every day on the App Store.  You just know your ticket number probably resembles this year's deficit figure.

BRB ... nope, still "Waiting for Review".

I've a ton of work to be done on my advertising campaign for Bakers Game.  I've got copy to write, I've got reviews to solicit, I've got to select my targets, and I've got to finalize my budget.  If Apple suddenly approved my app today I would be woefully unprepared.  It's not a bad thing they've taken this long; I've been very bad about getting my stuff together!

There just aren't enough hours in the day to get everything done.

But there sure would be more if I wasn't "Waiting For Review".

Monkeying about with Amazon EC2

There are a lot of new boards required for the Bakers Game refresh. And I do mean A LOT. We're jumping from about 5,500 easy boards to 12,000. And we're more than tripling the number of standard boards - and it's the standard boards which take FOREVER to compute.

So I needed processor power - gobs of it.

And thus Amazon steps in. Amazon rents out processors by the hour, and for cheap! For 17 cents an hour I can run two solvers on Amazon's hardware, so although their hardware isn't as fast as mine, they have much more of it. And solving winnable boards is practically Parallel Programming 101 ...

I've been running the solvers now on Amazon's hardware for about 72 hours. And in that time I've managed to generate more than enough easy boards and am nearly complete with the standard boards. And for just a little over $22 - which includes my experimentations, too.

With any luck I'll be able to wrap everything up and post for Apple's review team this weekend! Follow me on Twitter (@lgehrig1) and you'll see exactly when.

Keep coding!