The Far Corner of the EEC

I took some of the things in Randy Pausch's Time Management lecture (see related post) to heart and I'm trying to do at least a couple of his suggestions. One of the things that I am trying this week is keeping my email inbox empty. I've always had sub-folders where I let Outlook autosort email from big distribution lists, but I've let my inbox be my catch-all for everything else. I started this week by moving the ~6500 items from my inbox into a sub-folder and have since been attempting, successfully so far, to keep my inbox at "0 items".

I have to admit that it feels a bit strange. I switch to Outlook and there's an empty inbox. I can't pretend like I'm working as I scroll through the list. Everything that has come in has either been handled, deleted, or filed into a sub-folder. I know immediately how may "email-action-items" I have (3 currently) that require me to do something and respond. When I left work yesterday I didn't feel like I needed to go through my email "one last time" before leaving...it was empty, nothing to go through and I knew the one action item I had was waiting on a response.

Getting to work this morning I had 42 (meaning of life, the universe, and everything) messages waiting for me. Even spending the time to explore the links that interested me from the trade-emails that I get, I was back down to an empty inbox in about twenty minutes. Throughout the day, I don't feel like I need to check every email as it comes in, when I choose to look at my inbox the number of items is small and I can take care of them on my schedule. Before the clean inbox, I found myself looking as each email came in, there was so much in the inbox that if I didn't keep watch, I might miss something that actually needed my attention (and of course, this method meant that I would miss things that came in when I wasn't sitting at my computer).

My perspective on email as a tool rather than a job is already changing. As my colleague Josh said, "It takes two weeks to establish a habit"...I'm looking forward to hitting the two week mark with my inbox still empty.

I love working at Microsoft and I especially love getting to work on the operating system. One of the the things I love about working on the OS is that I get to run the operating system as soon as it's built. If there's a new feature, I can have it on my laptop the day it's in the official build (and if I'm really interested in something in particular, even before it's in an official build). I also get to see the evolution of certain features. Something that looked great in a controlled environment with just a couple users might not work as well when it's spread across a division or two², in those cases (and for many other reasons) you get to see a feature evolve; sometimes minutely, sometimes drastically. You get to see a creation that you have a part in grow up.

There is a downside however. Right now I'm running a recent build and while there are a lot of similarities to Vista, it's not Vista. The OS reports its version differently, the browser is a different version, and other changes that affect my computing experience are close, but not exactly like Vista. Imagine having a key to your car that is close, but not exactly shaped like the key that is meant for your car. There are web sites that just won't work properly and it's often not that they wouldn't work, but the developers were careful and made sure that if they didn't recognize the platform and browser then you were redirected to a polite page that told you that your configuration was not supported. Internally we often have "hacks" to get around such issues so we can test against these sites. application compatibility is key to our success, but it's like that key that doesn't quite fit right; if you jiggle it around for a bit you can usually get it to work, but it's a pain that you have to jiggle it.

The other downside is when I'm trying to help someone that is running "the latest software from Microsoft", usually my parents or siblings. These usually involve talking the relative through something over the phone and I often "follow along" as much as possible on my computer. But now things don't quite match. Microsoft spends a lot of time trying to make things simpler and more straight-forward for users² and this is goodness, but it also means that my "cheat sheet" (following along on my computer) no longer matches and I get to listen to complicated dialogs being read out a bit at a time while I'm trying to help solve some problem. A while ago I actually got my dad, mom, and sister computers that were identical to my home machine so that I could handle these situations more easily. Now if I could only convince my family to only have computer problems when I'm at home rather than at work!

Overall, the upsides of selfhosting far outweigh the downsides. It can be a bit frustrating now and again, especially when I want to try out the new cool thing "RIGHT NOW!" (today's RN! was Microsoft Live Lab's Photosynth)...but I'd rather be selfhosting than not have the opportunity to do so.

¹I also find it interesting why it didn't work when we go to larger scale...quite often it's not that the feature works all that differently for a bunch of users, it's that there are groups of users that use their machines differently or have different hardware or work remotely, etc.
²yes, you can probably provide the counter-example that bugged you, but take a second and think about the last time you needed to enter an IP-mask while setting up a netcard...and I could list many more examples where we have gotten it right

Occasionally, I get so wrapped up in other things that I totally miss something big (I didn't hear about the Columbia Shuttle disaster until almost a year later). The most recent "big" thing I missed was the buzz about Dr. Randy Pausch's The Last Lecture. When the buzz went around shortly after the WSJ article I didn't pick up on the significant bits of the conversation and filed this into my brain as "just another last lecture" and it went into the "if I happen to have time while it's still in my brain, I may go look for it" bucket. Needless to say, I didn't have time then.

Then the book came out and I had a second opportunity to miss the boat.

A couple of weeks ago I noticed it was available for the Kindle I had recently purchased and so I purchased a copy...but I was already reading something else (W.E.B. Griffin's The Shooters), so it sat on the digital bookshelf for a bit.

Finally, last week, I was talking with a friend and she mentioned having seen it on YouTube. "It's on YouTube?" I said. "Duh!", she replied (okay, she didn't reply exactly that way). So I launched my browser (IE8!) and watched it. That night (luckily this was last Friday, so I was able to stay up late) I proceeded to also watch quite a few of the related videos. This week I read the book and watched Dr. Pausch's lecture on Time Management.

And over the last week, I've been asking every friend and colleague if they have watched the lecture and encouraging those that have not seen it to do so. Now I'm recommending it to anyone who takes the time to read my (sporadic) blog. Both lectures will inspire you both personally and professionally and have applicability to both halves of your life. If you haven't watched it, find an hour and sixteen minutes during this weekend and do so. If you are married, watch it with your spouse. If you have teenage or older children, watch it with your kids. In case I'm not clear, if you haven't seen The Last Lecture, it's worth giving up sleep to watch it.

Bit of a disclaimer: I think I would be as passionate about this even without the personal touch, but it has touched me extra as about a month ago my Mother was diagnosed with cancer and a couple of years ago my best friend and skydiving buddy died of pancreatic cancer. I really choked up while reading the book when it mentioned Randy asking his doctor if he could SCUBA...my buddy had the same conversation with his doc but with skydiving rather than SCUBA diving.

Here's the links:

• Randy Pausch Last Lecture: Achieving Your Childhood Dreams
• Randy Pausch Lecture: Time Management
• Alice: Programming for a New Generation
• Book: The Last Lecture
• www.randypausch.com

Just read an article on WSJ about custom textbooks in colleges (here's the link, not sure if this will still be free by the time you click it). The basic idea is that there are several colleges (or departments within them) that are requiring custom printed textbooks. Interestingly, though the article didn't mention this, I'm sure that some of these custom textbooks are able to be made because of the computer/web technologies (getting the material to the publisher quick enough, editing large sections of the text in soft copy, etc. Note: The specific example they gave, University of Alabama's A Writer's Reference would not need heavy editing, but I think my general thought holds true.) This irks me as we are also hitting the technological point where we could easily reduce the cost of texts for students. The article's premise is that the prime mover for the custom textbooks is the publishers wanting to dent the used-book market and the university's desire to make a royalty on the books.

One use for the money that the royalty generates is for scholarship funds. Hmmm, make the students pay more for their books so you can give money back to students. Is this supposed to be a "intro to government spending" tactic preparing the students for life after college?

 A technology that I would love to see being used sometime soon in schools is textbooks on a Kindle (or similar device). I hate reading anything long on screen, it's hard on the eyes and the mechanisms of how we read are not supported by the comparatively low resolution (a poor laser printer does about 300 dots per inch while a really good monitor does about 120 (and it's a n^2 type of thing)...this means that reading enhancers like serifs are generally bad ideas on screen), but the electronic paper or eInk in the Kindle provides a high enough effective resolution that reading it is very similar to the experience of reading a book. The eInk screen is reflective instead of emittive (you can read it in bright sunlight, you can't read it in a dark closet) which I personally think also reduces eyestrain. The cost of publishing a textbook to the Kindle or similar device is pennies (cost of writing still needs to be recovered). If we (the world) really wants to make things better for students, this would probably be a good place to start looking. All that said, I think the Kindle (or Sony's eReader) still have a ways to go before we could do all textbooks on them. The interface is decent for serial reading, but not so great for flipping back and forth between pages...and in college I often had multiple books and articles spread out at once...an idea that works on the computer screen, but not so well with eInk.

 The Kindle (I've had mine for about three weeks now) has also made me want more eInk devices. It has a very slow refresh (I haven't timed it, but it feels like about 1.5 seconds), but the effective resolution makes it awesome for reading long articles. My current wish would be to have an auxiliary screen that was eInk that I could display documents, long emails, or reference material on...anything that I need to read, especially in depth, rather than edit. Until the technology gets a lot faster, editing is still better on the LCD.

I received an email yesterday requesting that I post something new in this blog. So, here goes an attempt to put something of interest here on a more regular basis.

Since my last post (WSOP2007...I busted out of the tourney in just under four hours...still a bit of scar tissue on that one), things have moved in several directions for me. Shipping Windows Server 2008 earlier this year was one of those things that I'll remember forever as a "life achievement". In my mind it's the best Server operating system ever shipped and the amount of testing that went into it was unparalleled in the history of Windows Server. After shipping WS08, I changed jobs and moved from the Windows Server Build Team to the Windows Build Engineering Program Management team. While this may not sound like different things, those outside of MS could look at it as analogous to a move from driving a car to a position where you plan how we should design the next engine for that car. Changing to Program Management is going to be a good move for me longterm (I think) and it's pretty cool to get back to learning a whole new discipline.

A couple of weeks ago I won a poker tourney and was thus selected to represent my local poker crew in the main event this year. I'll be heading down to Vegas on July 4th and will start playing on the 7th (day 1b). If things go well, I'll post the action as I see it. If things go poorly, I'll post my bad beat story.

A project I lent my face to is now live on the web: The Face of Windows Server 2008. Neat idea, looks like there are a few glitches in the site, but overall it turned out pretty cool.

If you haven't checked out Windows Server 2008 (Formerly code named "Longhorn"), it's worth checking out. Getting Windows Server 2008 built and ready for customers is my day job. I'm more excited about this release than any I've worked on in the past.

Yesterday I spent my afternoon and evening dealing poker for the incoming Microsoft Interns. An added bonus for me was seeing an intern named Richard perform. His control of a deck of cards blew me away such that I informed him that he was not allowed to deal or shuffle in any of the regular poker games I frequent.

Richard runs a website "decknique.net" that is definately worth checking out. The site includes a ton of videos and from the random sampling I've done so far, most of the videos have a higher production standard that much of what's on the net. Spend a few minutes on the site and you'll probably be thinking of grabbing a deck of cards. Fair warning: after Richard demonstrated a particular card flourish, I asked him how many hours it took to learn that and he responded, "Years."

Yesterday I spent the bulk of my afternoon attending the WSD (Windows Server Division) Quarterly Manager's Meeting. The meeting was focused on improving manager's coaching skills, but a side discussion emerged towards the end of the meeting concerning rewarding employees that get the job done without resorting to heroic measures.

It was clear in the discussion that we were not talking about the employee that just puts in his/her time, but rather the employee that through proper planning and execution is able to get exceptional results without having to resort to "fire-drill" mode. It was also clear that identifying these employees is more difficult than identifying those that have a short-term high-impact during a "fire-drill". The exceptional non-heroic employee is not the one in the spotlight and it can take months to realize the business impact of someone who is quietly good at their job.

It struck me that these are employees that we need to reward and retain. The nature of our business is such that we do occasionally need the folks that can jump feet first into the fire, but the exceptional non-heroic work style is the one that is long term sustainable and ultimately the style that will continue to make Windows Server a success.

Took a couple of days off. Spent most of it being lazy or playing poker. Spent very little time reading (work) email. It's amazing how much of a recharge it is to get away email for a couple of days.

Did get a doggie door installed (still working on convincing my girls that they can use the door; they seem to like the increase in treats); so I wasn't 100% lazy.

Also went and saw "Next" starring Nicholas Cage and based on Philip K. Dick's "The Golden Man". Decent movie, some of the scenes justify seeing on the big screen. Worth a matinee price.

Spent a bit of my time today over in Microsoft Studios contributing to a bit of work that should hit the web middle of next month (I'll post a link once it's up). Different experience being on the other side of the camera. Having make-up put on me was not one of my all time favorite experiences, but as none of my friends or co-workers were there to razz me, it wasn't that bad. The photo portion went quick and the photographer was using a Canon (a much nicer one than my aging D-60). I felt a bit awkward during the video bit, but overall I think it went okay.

One of the "prompting" questions asked during the videa was something along the lines of "so what makes Longhorn special?" This one was easy for me to answer. This is the best Beta release I've ever been a part of (and I started with MS back in '95 and have worked on Windows the whole time). This release is rock-solid. I'm not just happy that we hit a milestone, I'm proud of the level of quality across the board for Server. As Iain McDonald said in his Channel 9 interview, this Beta is production quality.

There are a few things we need to fix. There is some fit and finish work that needs to be done. But, this is release is something I'm proud of.

Around 3:10 this afternoon, the final signoffs were completed and we released Beta3. This has been my number one priority (and that of most of the folks I work with) for the past several months. A lot of hard work and many, many, hours of ship rooms, but we are releasing the Beta with a level of quality of which I can be proud. Additionally, we are building Longhorn with Longhorn!

Check out the video in Dave's post. (For those that know me personally, I have a very minor part...you can see me 31 seconds into the video and hear me a bit later when I say something like, "How about we deliver it with the morning newspaper".)

http://ct.enews.eweek.com/rd/cts?d=186-5691-48-550-66509-642524-0-0-0-1

Bit light on the statistics, but interesting nonetheless. In the build world we beat up on hard drives and tend to go through quite a few of them. In the build lab I work at we had a slew of drives die due to some poor configuration in some RAID packs (disruption to the air flow cooling the drives). While we are now past all the failures that resulted from that, our MTBF is still way above the 1% over a single year.