sysd

So I've been playing a lot of "Race for the Galaxy" lately. RftG is a pretty addictive little card game where you build a galactic empire and score points by balancing construction, exploration, production and consumption. (If you're interested and have never played, the rules can be found at the Rio Grande Games website; an excellent, free, online version with bots that'll kick your ass is available at keldon.net ).

One of the nice things about RftG is that it's quick - a game is over when someone fills out his empire's allotment of twelve cards or when all the loose points have been allocated. When playing against the AIs, the game is usually over in under 10 minutes. This is actually difficult for me, because it works against two of my personal weak points when playing games.

My first weak point is that I tend not to pay terribly much attention to what the other players are doing. This means that I find myself surprised when the game suddenly ends because an AI fills out its empire (which should be obvious, you can only expand your empire one or two cards max per turn), or that I'm waiting to get a card that another player already has played out in its empire.

My second weak point is that I'm a pretty good tactician, but a lousy strategist. The goal of playing a game is to win; since games don't play forever, in any reasonably complicated game your endgame tactics should differ from the your midgame tactics. For example, you shouldn't waste resources paying for a cost-reducing development at the end of the game anymore than you should tune up the engine of car that you're about to donate for scrap. I tend to think in terms of long-term improvements, which makes me a nice guy and a good designer, but a poor gamer. My gamer friends kick my ass regularly because they know that strategies are ultimately finite. They see the end and plan accordingly; I see a perpetual tomorrow and don't plan, but just build.

So one reason I play games like this is not only because I enjoy them, but also because they'll hopefully help me to think strategically, not just tactically. As I get older, I've gotten tired of simply reacting to my surroundings; you won't change the world playing defense. This is why, for instance, I became an architect at my job; it's often a hassle, and less enjoyable than straight-up development, but I'm responsible for shaping our environment and hopefully making it better for everyone involved. This is why I'm pushing the Housers to move beyond just building shelters and towards connecting our clients to other services afterwards - we don't want a hut to be someone's final destination in life!

Incidentally, this relates to a phenomenon I've noticed with some of our homeless clients. Occasionally, I'll run across a person that seems basically all right. But the first meeting's deceptive that way; it's after the third, the fifth, the tenth meeting that I realize that although this person appears rational and articulate, they have been saying the same things over and over again without actually making any progress. There may be lots of activity, and plenty of drama - but no movement.

I might have a hard time strategizing, but these folks cannot plan, period. My MUD friends referred me to 'executive function', which can be affected by brain trauma - concussions, drug abuse, microstrokes, etc. That sure sounds like the issue, all right. This implies that our fledgling efforts to try and help move our clients out of the huts may become significantly more involved; do we become involved in goal-setting? Or is there some other organization that'll do casework like this?

Explanation. Due date apparently late June/early July.

I sent an email to friends announcing rosindust 's pregnancy, and in it I described how we choose names for our offspring. I don't think I've described it here, so here it is.

Nowadays, many modern couples don't want to have the wife change her surname to her husband's name. This leads to interesting contortions as they attempt to figure out naming conventions for the family and for the children.

The most common approach is the hyphenated name. This works, but only marginally. By 'marginally' I mean that if it were adopted by the public at large, it would quickly break down after the second generation. There are options to let the kids drop a name etc etc, but if the goal is to avoid making people change their names, this really falls short in my opinion.

You also occasionally hear of folks using a completely new last name or a compound name made from elements of the parent's previous surnames - for instance, since my last name is Hess and rosindust's is Woodard, we might become the Woods. This is a little avant-garde for my tastes, and once again, you're changing a name.

So we decided to simply not change anyone's name. Well, then what about the children? Here's what we came up with:

• The first child's surname would be that of its same-sex parent. Subsequent children's surnames will alternate between the parents regardless of sex.
• The parent who does not contribute the surname picks the first name.
• Middle names are reached by agreement from both parents.

Our first kid is a boy, so he's a Hess. Tracy picked the name Karl to honor a great-uncle of hers, a very interesting and decent man by all accounts. We agreed on Arlen, my father's name, as a middle name. Karl Arlen Hess.

This next kid is a Woodard. I've decided to name it after my much-beloved maternal grandmother, Ada, so if it's a girl, she'll be Ada, and if it's a boy, Adam ^*. rosindust wants to give her middle name, Kathleen, to a girl child, which I'm fine with, and we're still going over options for a boy. So it's either going to be Ada Kathleen Woodard or Adam Woodard.

So with this convention, nobody needs to change a name and neither family 'loses' a name in a marriage (at least, not in a marriage with two or more kids). The only difficulty is explaining a single family with two surnames to the various powers-that-be, but in this age of remarriage and blended families, the PTB will simply have to cope.

* Technically, the closest male version to the name Ada is Adolf. Uh, no - some folks already give us funny looks for Karl with a K. I've tinkered with creating the name 'Ado', but that's a long shot. Adam is fine.

I've noticed some troubling changes with rosindust lately. I'm worried that perhaps she's undergoing some sort of seasonal something or other - I mean, she's been telling me that it's my fault that she's getting kind of plump, when all I can see is that she's sleeping and eating a lot more, even though she's also been complaining about tummy troubles. I tried to talk with her about losing weight, but she told me that she's going to gain weight until late springtime/early summer, but that then she'll lose somewhere between seven to ten pounds all at once. I dunno, that sounds like a pretty crazy diet to me.  Fortunately, she's also gone to a doctor - but get this - she's going to an Ob-Gyn instead of a regular doctor. I'm like, is this some sort of crazy chick hormonal thing? and she just looks at me like I'm retarded.   What gives?

Oh well. I'm sure this will all pass. At least her boobs are growing.

(This post is going to be a little rambling and whiny, but I'm focusing more on getting it out than getting it perfect.)( core dump )

I think that there's some relief on the horizon. In the current crop of volunteers, I've got a few guys who seem to be interested in leading builds. I've pledged to cut back on builds and do more training next year, so that should hopefully free up time from hut building to organization building.  

Ultimately, I think that our fitful progress represents the awkwardness of growth. I do know that I, personally, need to work on delegating more often and learning how to (ugh) manage. I can lead teams, but managing is something different - on a team, you have a discrete, visible goal for your team (build that hut!), whereas managing is about shaping the growth and activities of your group to make it more capable. To condense into an epigram: team leadership is outwardly directed; management is inwardly directed.

Friends dying. Friends getting divorced. Friends breaking up. It's been a tough month for our social circle. KNOCK IT OFF!

When things go right, they're OK. When things go wrong... they're dreadful.

Here's how I explained the situation to the MUD:
( Probably NSFW, unless you work for Larry Flynt )

I've worked for six months - or longer, depending on where you start looking - on the new weather.com that launched this morning. Now that it's out, I can finally talk about it.

For starters, it's much more than a "redesign". We've actually redesigned weather.com at least four times over the past ten years, changing elements of layout, design, and navigation. What you see with the new weather.com is a ground-up rewrite - new application framework, new data layer, new page design, new caching strategy, new development methodology - everything. Some elements had been developed earlier and thus were "off-the-shelf" for us, but we'd never been allowed the luxury of simply ignoring what we had done before and starting over. We took the opportunity to learn from our past mistakes and do it right this time.

That said, the new framework is only on four pages - the home page, the main 36-hour weather page (/weather/today), and a couple of navigation pages. All the rest of the pages are the old framework with new headers and footers.  Over the next year or two, we will transition our various pages into the new framework *. So for folks who complain that the redesign is broken, you're probably looking at an older page which is having an issue with the new header/footer scheme. We're working on it, believe me.

A few new things about  the new site that please me -

• "Trupoint" locations. If you look underneath the search box, you'll see a link saying 'try it' for something called Trupoint. This does address-level resolution for anywhere in the United States - for instance, here's the weather for the White House.  This is tying into a system we have called HiRAD, which does sophisticated modeling to determine the weather at a 2.5 kilometer grid across the continental US. Lemme tell you, it's an amazing amount of data.

Caveat - Trupoint locations are only on the new site (those four pages). Thus, if you create a custom location like the White House, then navigate to a legacy page, and then go back to the overview weather page, you're going to be looking at the ZIP code location instead of the fine-grained Trupoint location. The old system simply isn't engineered to handle TruPoint natively. This is a problem that'll only get better over time, as more pages are migrated into the new system.

• Saved locations. This is nice - upper left  right (thanks, Yakka!) corner of the header, you can save up to three locations, including Trupoint locations, and we'll show the current conditions for those locations everywhere on the site.
• The apps. This is below the weather section of the 36 hour page (/weather/today). We're always interested in trying to match up user's interests with the weather - it keeps folks interested and it offers better targeting for our advertisers.  Previously, we'd asked folks to declare a primary interest, such as sports, and we'd offer a tiny bit of extra content on the weather page. Why would someone even bother?  But the apps allow the users to pick content that they're interested in, so that if you have more than one interest, you can have more than one app. So you can add, say, pollen forecasts to your weather page if you're allergic, and see what the local golf course conditions are like if you play golf, and etc. That's the right way to do it; instead of asking 'what are your interests' and showing you content, we allow the user to pick the content and determine your interests from there.
• Render speed. weather.com is not known for being a speedy site, but we're working on that. The new pages are designed to render the HTML first off, and then add the various bits of Javascript functionality after the page loads.  Furthermore, customized content, such as the saved locations and user apps, are all loaded asynchronously, allowing us to cache pages instead of generating a custom page on our servers with every request.  Once again - this is only on four pages so far, so the vast number of pages are still kind of slow. However, the most visited pages should be noticeably faster, especially after your first page view.
• Under the hood. Let me tell you, the legacy site was a complete bear to debug and maintain. The new site has much cleaner and clearer separations of concerns, making it a joy to work with - data, control, and view are all clearly demarcated. This makes coding faster and debugging simpler.

Undoubtedly, there are things that we could have done better and things that we'll look back on and say 'what were we thinking'? But I'm very proud of this project and of the team that put it together. weather.com is going exciting places, and I'm happy to be along for the ride!


* This will also give us the opportunity to review our product offerings, update what needs fixing and eliminating things that simply don't work.   Weather.com is ludicrously overstuffed with products - did you know we have an outdoor grilling section? How about a vehicle safety guide? A flight tracker?  I mean, all of this is potentially useful information, but is it really in our mission, and do people go to it? "You know, I'm thinking about  buying a car. Hey, let's check weather.com!"