Ken Circeo: The View From Building 17

Thanks for dialing up my all-new weblog where I'll focus on the speech industry and speech technologies, particularly as they relate to Microsoft.

Getting Kicked Upstairs
I'm a writer with Microsoft. Redmond campus. I've been with the company for just over four years now, and in that time I've had eight offices. Just this month, they moved me to my current office in Building 17 third floor, but who knows for how long? I used to work for Novell, so I understand that large corporations like to move employees around just for agitation. I don't disagree with the concept, but between you and me, I hope I get to stay in this office for awhile because of its proximity to three places that I need to get to often: the copy room, the kitchen, and the bathroom. And to top it off my buddy, Phil the Coffee Junkie, is now right across the hall from me. Perfect. I used to have to walk down two halls just to get to his office for my afternoon espresso. But now it's just a few steps. So the office location is good. In fact, I couldn't have picked a better spot. (If my manager, Bob, is reading this and thinks I'm being sarcastic, I'm not.)

On Technical Writerzzzzzz...
As a writer, I put together Help documents, whitepapers, and the occasional web article for the company. All designed to sell and support Microsoft Speech Server. Yes, it's technical writing, technically. But, like most technical writers, if I had to sit in front of a PC and write "Step 1 - Step 2 - Step 3" for 2,000 hours a year, my mind would be fried like Keith Richard at a frat party.

So I don't do that. Instead, I approach it like a journalist for the People's Press, trying to figure out what kind of documentation is going to help customers the most, and then delivering it. If that seems noble to you, well, that's just the kind of guy I am.

v2 Solutions
For our next version of Speech Server, I'm concentrating on writing end-to-end solutions docs, which tell you everything you need to do from the time you open the Speech Server box to the time you go live with your speech solution, all based on deployed real-world case studies. Sounds impressive, doesn't it? It better be. It's turning out to be a dang sight harder than I thought it would be. The case studies aren't the problem. We released Speech Server v1 about a year ago, so there are plenty of good applications around that are built on v1. But remember that I'm writing my doc for the v2 product, which is different from v1. So I'm having to retrofit the solutions into the v2 structure.

Doable? Yes. Easy? No.

Thankfully, I don't have to wrestle with any code, which might have pushed the solutions project into the "purt' near impossible" category. We've got a good library of code samples that we deliver with the product. But, while delivering code as part of the solutions docs would be nice, I'm thinking that providing speech elements like the call flow and dialogs, as well as security and tuning practices will help people stay on the right track throughout the dev and admin process, particularly if they're new to the whole speech writing biz. At least, I hope so. (Otherwise, I might find myself getting traded back to Novell for a user doc writer and an intern to be named later.)

Cup o' Joe
Today's coffee flavor: Pegasus Espresso mixed with Peet's House Blend
Result: Quite good