If I can go another hour without working on the book, this will be the first day in ages that I haven’t spent any time working on the book. I spent this morning gathering ingredients, and then a couple hours cooking and prepping, and then from five on a bunch of people came over for dinner.

So:
– I had no idea how much time it can take to entertain. I understand why people spend so much on catering now
– That felt really good
– My friends are cool
– Now that it’s over, I’m about half happy I had a good time and relaxed and half overwhelmed with guilt and dread

Bat corking finished.

~125p of rewrites left, and I leave for London on 8/20. Things are going to get worse before they get better.

After Paper of Record destroyed itself, denying me access to the Sporting News, I’ve had to rely a lot on Proquest which I can access through my local library.

It is, without a doubt, the worst system I’ve ever been forced to use. I remember the bad old days of library systems when you’d have to crazy stuff like
subject:baseball +casey +stengel
hit ‘enter’ and hope for the best. I would love to have one of those systems lately.

If I had to sum it up, I’d say “Proquest doesn’t do what you tell it to and provides radically different results given only slightly different inputs with no feedback as to why.”

Here’s an example. The opening Proquest “Basic Search” page looks like this:

In the search box, operators like “AND” and “OR” don’t work as they’re described in the search tips. More frustrating, it will not return results it obviously has.

The date range doesn’t work at all. Searching within two dates always returns results outside of the range. There’s no message as to why. Did it fail over? Were there no results? No clue.

The “database” dropdown allows me to select the “Proquest Newspaperes” and the New York Times. Once you’ve selected that, the results sometimes improve. If I search for some strings and don’t get any results, narrowing it down to one paper often returns what I’m looking for.

Proquest doesn’t have any of that documented or hint at it. Why would it? “Your comprehensive search failed. Would you like to limit it, search each possibility one at a time, and turn up more results?”

Gosh, that’d be great. Or, better yet, why doesn’t the comprehensive search work like that?

The date limits, I should note, do sometimes work when searching individual sources. Whether or not they’ll work on any given search, even if it’s a source that’s honored date limits in the past, is randomly determined at time of search submission.

Here’s how bad this is: I’ve been using other sources to find references to New York Times articles and then gone to Proquest to search only the NYT only that day for that headline, and my success rate is probably 75%.

As an added bonus, at random times it will throw up a login screen which is impassable. Then I get to start over.

I’d love to use Lexis-Nexis, which I understand is far more difficult to use but 90 times as powerful. Let me at it. Except that it costs (I believe) $900,000 for an indvidual license.

Google, would you please smash these guys and hurry? Researchers everywhere are crying out in pain for a search solution that works. I’ll write you a spec and everything.

After a particularly hard stint of rewrites and reresearch, I had a ton of books out on my desk, and took the opportunity to order my shelves of baseball books by color. It’s a pleasing look — oddly reassuring. However, the fact that I did it makes me wonder if I’ve finally cracked under the stress.

Today was the Hidden Ball Trick and the nth revision of the Black Sox chapter.

I went to the grocery store this morning and the cashier asked me if I had the day off and I told her I was unemployed. “Any prospects?” she asked. I laughed. I don’t know.

Every day I think the chances I go back into IT drop by a percent, half a percent, and I’m more likely to try and stick it out as a writer. Which would certainly make life spicier.

I got up this morning determined not to let the deadline stress get to me. I made breakfast, finished up some new material on the steroids chapter (turned out nicely), and moved on from there. All in all, I wasn’t as productive as some of the other days lately, but I’m a lot happier with the product, and I’m settling in for the evening session with the M’s game on, and I feel pretty good.

I took on a fairly huge chapter today (it runs 50 printed pages double-spaced) which has been in pretty good shape through the whole process. It was part of the initial proposal, which means in some form I’ve had years to work on this.

Anyway, I run through it really quickly, happy that it’s largely light work with a little tweaking, addition, and subtraction. I get to the end and there’s a note that the chapter runs too long (which it does). My editor would like to see 5-7 pages trimmed (and gave some really good advice on how to get there).

I started at the note for a good minute, not really believing I’d just read that. And then I started the second pass, cutting for length.

So my editor mentioned in passing that she “really gave it to me” on two of the chapters.

Here’s what she means.

Every page in the chapter looks like that. On several, the notes continue onto the back.

So I got the edits back on the book and USS Mariner went down. Three days later, it came back up today. A brief chronology:
Wednesday:
– Carl Everett is tossed off the team, and then the M’s trade for Ben Broussard. USSM sees a huge traffic spike that takes down the shared webserver and the shared database server. Our host shuts the site down.
– We eventually get a temporary page where I can start throwing up straight HTML updates. Good thing I was around back in the old days and know how to write those. You want Ajax and Web 2.0? I offer the h2 tag, and use ordered lists.
– Using parts from my last PC and stuff I’ve had hanging around, I build a new server

Thursday:
– At about 2 in the morning, I got up and drove to Enumclaw to volunteer for RAMROD
– At about 6:45, I drove home. This took forever.
– I get home, finish up prepping the server, and drive it Tukwilla. On the way down, I see that 405 North is just as bad as it was when I tried to get hom earlier.
– I take I-5 N back from Tukwilla to I-90. I-5 is a crawl.
– At some point, the hosting guys tell me they expect they’ll have the box ready later that night. That doesn’t happen.
– I go to sleep, exhausted from having slept maybe 2 hours since USSM went down

Friday:
– ETA for the new box is that evening. That turns out to be 8, when I get an email that I should check the box and then call in if it’s cool.
– At about 8:20 I call in and get put on hold. I take the “leave us a message” option, but it blips and says “no message to send” and there’s no option to re-record or anything.
– I file a support request
– I call back in and sit on hold for 45m, during which I bitch to Jeff Shaw on IM
– I give up and go to sleep

Saturday
– At eleven, having heard nothing, I call in and get someone, who promises to look into it and call me back
– I call back sometime after noon, and they’ve made no progress. We both work on it while on the phone and bring it up at about 2

I have a lot of mixed emotions about the whole thing. On the one hand, yeah, we’re around in large part because of digital forest’s largesse, so I understand that we’re getting their tech’s support as it’s available, and got bumped every time a paying customer had any kind of issue. I totally understand.

But from my side, this was three days of troubleshooting, slipped estimates, during which my blood was 90% stress and 10% caffeine. It was three days to getting my box online, and if that hadn’t worked, I don’t know what I would have done next – I was considering buying a rack server from Dell and having it overnighted, and that would have been a couple thousand dollars, maybe.

Or turning out the lights.

Anyway, it’s back to the book revisions. It sucks that I can’t celebrate the resurrection of USSM with a rush of joyous content, but I’ve only got a few weeks before I need this draft turned around, and the book, unfortunately, has to be a priority right now.

Cracked open the manuscript to start revisions on version three. Right now, they look thankfully far less massive compared to the last set, though of course you don’t really know until you get into it.

This ought to be fun.

Tinker with the grind. After some experimentation, I’m getting some really stellar coffee. It really does make a remarkably smooth, sweet cup of coffee, with little of the nasty side effects that come with a french press or other devices.

Check out this review at Dan’s Data for a really good overview.

I’m starting to really see the difference in beans, which is something that largely used to escape me (on the other hand, I once voluntarily drank Keystone Light on a regular basis, but in my defense, I was broke).