New mouse gesture: “slide-slide-double-smack”

I'm spending some time in a cube (ugghh, I know) next to a woman who seems really nice, except for one thing. When she's away from her desk for a while, her PC goes to sleep, and she's convinced that the way she needs to wake it up is to slide the mouse back and forth a couple of times, and then pick it up and smack it on the desk twice. Exactly twice.


Apparently, by the time she's through executing "smack" number 2, her computer has been sufficiently jolted to be able to respond to her, because she ceases the beating at that point.

I'm sure there's a usability lesson in here somewhere, but it escapes me at the moment. My head hurts too much to think about that.

Ford: Demise of an industry

You've probably seen all the announcements about Ford's recent record-setting performance. They managed to cough up $12.7B -- that's the big "B", folks -- last year. This is tempered only by GM's $10.6B loss last year ("all the kids are doing it, Mom!").

I don't know about you, but the first few times these staggering numbers crossed my desk, I managed to brush them off. For one thing, $12.7B is large enough to go into my mental "Federal deficit" bucket -- the number's just too big for any kind of normal comprehension. After you cross the billion dollar threshold, I start to get a little glassy-eyed. Maybe we should measure Ford's loss in Gates' -- as in, "Ford lost a quarter of a Gates last year."

The other problem I have with these numbers is that it's just way too easy to chalk it up to Ford's management just having been numskulls last year. "Get your head out of your ashtray and build a car that people want," or "build a car that doesn't fall apart on its way out of the dealer's parking lot." Easy scapegoat.

But it turns out the problem isn't quite that simple. Take a look at this article from CNN Money. This is a great introduction to the hole that our big-three automakers find themselves in. The article points out that our nation's automakers are giving away a $2900 handicap per car to foreign rivals. Of this total, the biggest hits come from labor costs (especially health care), a weak dollar, and a negative brand premium.

So here's where this article started to get really interesting. The blame for the brand premium problem falls squarely on the shoulders of the bean-counters in Detroit. These people are every bit as capable as any other business executive of reading the tea leaves. Gas is going to jump every time the weather is colder than normal, or warmer than normal, or windier or rainier than normal. Gas is going to jump any time anyone hears any loud noise coming from the general direction of Mecca, or if a camel does his business on a pipeline. Any auto executive that doesn't get this should be stood up in front of Donald Trump -- end of story.

Same thing goes for styling. As in, get some.

But the rest of their problems -- who do you blame for those?

How about health care costs? What, *exactly*, is it going to take before our government collectively gets its act together and starts to make this better? Our health care problems clearly affect every aspect of our nation, including its overall economic performance.

How about labor costs? Most people have heard the quote, "What's good for General Motors is good for the rest of America", attributed to GM Chairman Charlie Wilson in his 1955 appearance before the U.S. Senate.

In those days, GM was considered invincible. The extent of their power and dominance was even greater than Microsoft's today, in part because of the big-capital, industrial nature of their product. They kept a whole lot of other businesses churning away, and this effect was very visible. You see a big factory cranking out steel a lot easier than you see the impacts of computer software.

The big automakers, then, tended not to get a lot of sympathy when the UAW came knocking. Contracts got bigger, workers gained protection, and pension funds were established. Now, of course, these decisions have become gifts that just keep on giving.

I have a feeling that there aren't too many executives at the big three who were around in the 50's. And once that ball got rolling, I can completely imagine that it was hard to slow it down.

Who do you blame?

Do you blame today's executives? Who's going to be the one to go toe to toe with the unions and weather a long, dry strike?

Do you blame the unions? Which union leader is going to go back to his guys and say, "Yeah, so it turns out we're asking for way too much from them. We're going to end up with no jobs at all if we keep this up, and we've got to reduce our costs."

I wouldn't want to bet on who's going to blink first.

What about the weak dollar? It turns out this is effectively a subsidy for foreign automakers to the tune of around a grand per car. No matter how good our automakers get, they're at a thousand-dollar disadvantage.

Who do you blame? Again, the problem here isn't GM or Ford. It's an international economic and political issue for the United States, and the solution has to come from Washington.

What if GM and Ford are the canaries in our coal mines?

Finally excited about AJAX

I've known about AJAX since the acronym was first coined. I saw the buzz when Google first loosed "suggest" on an unsuspecting community of developers. A palpable murmur swept through online communities as thousands of developers brought up on GETS and POSTS asked, "How are they doing that?"

It turns out that the way Google was doing it was sophisticated, but not revolutionary. In fact, it was a great use of a collection of technologies that had been around for a bit; Google just managed to make them sing together.

The only bad news about AJAX was that it was ugly. No, the apps were beautiful -- it was the code that was ugly. So AJAX went on my personal radar screen of things that were going to change the way I do development... as soon as the kinks got worked out a bit.

The last 6 to 12 months have seen the tempo really start to pick up. In no particular order, we've seen:

  • AJAX-enabled component suites have become common. There are really solid releases from well-known vendors like Infragistics, as well as strong newcomers like ComponentArt.
  • Microsoft's Atlas is nearing release. Like many of their products, they're not breaking too much new ground, but their presence provides a much-needed center of gravity for developers and vendors alike.
  • There has been an explosion of "2.0" sites using AJAX. They're literally everywhere you turn.

I've played with some of these control suites (Atlas, Infragistics, ComponentArt). They're nice. A whole lot nicer than that first Google JavaScript code. But they're still a fair bit of extra work, relative to making the same UI without the whizbang.

So I was just a bit skeptical when I saw a link pop up in some Google search I did about an open-source framework that claimed to produce AJAX-enabled web applications from Windows Forms projects.


But it's true. Almost. It's almost true, and that's a whole lot higher than the current bar. The project is called Visual WebGui, and it damned near does exactly that. You develop in a WinForms application using a set of controls from Visual WebGUI (that's the first catch), and when you run it, you've got an application that looks impressively like a WinForms application, but it's running in a browser. And no, not just IE -- I created an app in less than two minutes and ran it in both IE and FireFox.

This project has some serious chops, and they're executing on more than just technology. Visit the web site, and you'll see a well-done web presence, complete with documentation and an impressive collection of video tutorials. Watch one or two and see for yourself if this is all I've made it out to be.

Is it ready for prime time? Not quite, if you want my honest opinion. In playing with this toolset for a couple days, I ran into a few too many cases where I had to blow a form away and start over because I got myself into an unrecoverable error state. This sort of problem isn't a deal breaker if it happens every once in a while, but at the frequency I saw problems, it would be a problem.

I can assure you, though, I'm keeping my eye on this project. I've seen hints that they're going to try to enable Mono support, too, which would score extra points in my book. I doubt this would ever be a mainstream deployment platform, but I think it's important to have that option.

So go take a look - I think you'll be impressed.

Why do I read blogs?

If you've visited LinkedIn lately, you've noticed their new Answers section. I just finished posting an answer to Al Tepper's question: "Why do you read blogs?"

Al asks:
Why do you read blogs?
What is the biggest draw? What keeps you coming back? What gets you commenting? What gets a blog in your RSS reader? What makes you talk to others about a blog?

I began typing what I believed would be a one- or two-line answer, and the next thing you know, I'd rattled off a couple of things that do, in fact, bug me about some bloggers. Here's my reply to Al:

Before RSS became popular, the number 1 factor for me was frequency of updates. Now, however, I'm far more concerned with quality and relevancy. It costs me nothing to add a feed to my reader, and as long as the author respects my time as a reader, I'll leave the feed in my list.

What do I mean by respecting my time? Here are some things that'll cause me to drop a feed:

  • Lack of full-text RSS. I hate "teaser" feeds.
  • Repeating a link without adding any content. If there's a popular story going around, I don't need another link to it. I want you to add value. Explain why you care. If you want to just share links, use Google reader to share it and publish your link blog.
  • Adding a blog entry just for the sake of publishing something. Some of the most valuable feeds in my list publish very infrequently. If you don't have something to say every day, then don't.

You'll notice I keep referring to "feeds" rather than blogs. That's purely intentional. If you're not doing RSS, and doing it right, you're probably off my radar. That's also how I share with others, by the way -- I use Google reader, and I publish my shared feeds.

Related Links -- The Campaign against partial RSS feeds
My Google shared links feed

Yahoo! buys MyBlogLog

You've probably noticed the MyBlogLog sidebar on the left of this site. In case you haven't checked into it, this is a social networking tool for bloggers, linking like-minded blog writers and readers, as well as introducing readers to blogs they might not have noticed otherwise. I've been a member for almost two months, and I really like it.

So does Yahoo! -- they just announced that they bought MyBlogLog for a little over $10M -- read more here.

Windows Home Server – Good or Evil?

If you're watching any of the news from CES, you've seen announcements about Microsoft's new Windows Home Server. When I initially heard about this, I was really excited, but when I started to learn more, my excitement faded, to be replaced with nervousness and even bitterness.

What changed my mind after my initial take, and will I ever come back to the Windows Home Server Camp?

First, let me catch you up if you haven't seen the announcements (links at the end of this story). The idea here is that Microsoft wants to own the home-NAS (Network Attached Storage) market. They understand that just as storage became a problem of critical proportions for businesses, it's also becoming a problem for homes.

This is true, and I applaud Microsoft for recognizing the problem.

To be fair, there are a handful of really, really nice features in this product, most of which are available elsewhere, but in true Microsoft fashion, are not available in an integrated platform that I could reasonably expect my parents, for instance, to operate. My favorite features:

  • Backup. Plug in the device, load a client on your windows PCs, and they get backed up.
  • Restore. (Makes sense, doesn't it?) Pop a restore CD into your lobotomized client and restore from the Home Server.
  • Media streaming. A good thing, but I'm not crystal clear on what breadth of client will really be able to use this.
  • Remote access. Get to your files, and even your desktop via the Home Server using your Windows Live ID.

Ok, so it's all good, right? What could I possibly be nervous about?

Let's start with the easy stuff.

  • Windows client. I suppose I should have seen this one coming, but it's still damned disappointing. There's absolutely no technical reason why this couldn't be an open platform. Despite the olive branch that Microsoft appeared to extend when they made their recent marketing deal with Novell, this is where the rubber hits to road, and there's nothing open about this platform. Maybe I'll be proved wrong before this thing releases. I hope so.
  • Live ID. Here's another winner. Why, oh why, do I need a Windows Live ID to access my files? Anybody? Buehler?? I'd be ok if Windows Live wanted to provide something like DynDNS support so Mom & Dad don't have to figure out how to configure dynamic address support. But please, please don't make this the only way to access my files. Given my earlier experience with Windows Live services, I'm cringing at the very thought of Microsoft holding my files hostage.

Now, some concerns that may be a little less obvious:

  • Network speed. Not mentioned in any of the press releases I've seen, the price of admission here is Gigabit speed, hopefully with jumbo packet support for quick file transfers.
  • "RAID is for bugs." I actually saw this in an interview with on of the product management people for this product! I'm not opposed to the idea of technical innovation  I really like what these guys are doing for backups, for instance, by storing each distinct file only once no matter how many PC's it's found on.  I really don't like the attitude that makes Microsoft feel they can completely discount the technical merits of RAID without explaining what they're doing differently. Read the datasheets, for example, on Infrant's X-Raid (below), and you'll be able to see precisely how they improve on RAID without throwing out the parts that work. Microsoft needs to come clean on this before I'm going to start uploading tax returns and baby pictures.
  • Adding external disks. This could be a really great idea, but again, I'd like to see some technical background. Specifically, I'm concerned about speed here. The "plug-in" support appears to be USB-based, and not SATA. Please, publish some benchmarks so I can be assured that I'm not going to slow my NAS to a crawl if I need to expand it.

It's early yet, and there's a while to go before we'll know the whole story. In the mean time, I really hope we see some more information not just marketing hype about the tech that makes this product go.

Related Links

Paul Thurrott's SuperSite for Windows: Windows Home Server Preview. This is, by far, the most complete breakdown of Windows Home Server I've seen so far. Read this article first if you haven't heard anything about this product.

Windows Home Server cures Digital Amnesia (sorry - Microsoft has since taken this site down). By contrast, here's Microsoft's marketing site for Windows Home Server (built in collaboration with HP and AMD). This site is long on fluff and sparkle, and very, very short on information.

Engadget : Windows Home Server. As always, Engadget is a great source of breaking news, and they're once again covering the big announcements well.

cek.log : Windows Home Server This is Charlie Kindel's blog. Charlie is the General Manager for Windows Home Server, and he plans to release some more info on Home Server soon via his blog. I'll watch this space for more news.

AMD Positioning Itself For Home Server Market The HP product that's featured in Microsoft's announcement is an implementation of AMD's reference architecture for this product. AMD has an important toe-hold with this product, and I'd like to seem them succeed, if only to keep the processor market kicking with competition.

Infrant ReadyNAS+. Here's a glimpse of what this product looks like when it's built on open standards. I'm a big fan of Infrant -- until Home Server came along, I felt they were hands-down the class act in home and SMB-oriented NAS. You could also check out Buffalo, but Infrant strikes me as a bit more solid and mature.

Steve Lacey -- Hands-on with Infrant ReadyNAS Here's Steve Lacey's experience unboxing and setting up an Infrant NAS. This is the competition.

VM Products – you really do get what you pay for

Virtualization is the greatest thing since sliced bread, except when it's a big pain in the neck. We've got a VM host that our IT administrator set up to support development and testing machines. It's a real nice box -- a Dell server with a pair of Xeons and 32GB of RAM.

Unfortunately, we spent money on the metal, but skimped on the software. Our guy used Microsoft's Virtual Server on this box, largely because it's free and "it's from Microsoft, so it must be good."


Yes, in general, it's good. Performance has been fine, and management is tolerable, though not great. I ran into a brick wall today, though, that really highlights why you need to consider a higher-end product for a large-scale server (our machine currently hosts around 30 VM's).

It turns out that Microsoft's Virtual Server won't let you overcommit memory. What's that mean? It means that if you have 10 VM's, each configured to use a Gig of RAM, you're going to need 10Gb of RAM on the host, plus a little for overhead. Even if each of these 10 VM's is only really using half of its allocated RAM, MS Virtual Server still needs to reserve all of it in physical RAM.

This, in a word, bites.

VMWare's ESX server, by contrast, lets you set up a minimum size and a maximum size for each VM. The minimum size is still physically reserved, but the difference between the min size and the max size is allocated dynamically when needed (and available). If you stop to think about it, the other resources on the server are already managed this way -- CPUs are shared, disk space is allocated on demand, and so on. Why would you want to be required to supply physical RAM for all of your VM's?

In practice, this limitation is a nagging problem for small-scale servers and workstations, but if you're planning a server installation, you'd be well-advised to weigh the cost of additional RAM for your server when you look at the relatively higher cost of products like ESX server.

Louderback on Today

I turned on the Today show this morning, and Jim Louderback was doing a little bit where he compared devices of a few years ago to those we have today. He looked at early PCs, laptops, and so on.

Since I've seen Jim speak before (going back to the 90's -- yikes!), I was a little disappointed in the content. He seemed to really be trying to tone it down for the crowd this morning, which may have gone over a little better with some of the crowd.

The best part of the interview, though, was him trying to demonstrate a Lenovo convertible laptop / tablet PC. He started with the laptop opened up normally, then showed how it converted into a tablet, and then demonstrated the touch capabilities. He was standing above and behind the tablet, though, so he really wasn't too nimble with the handwriting. The best he could manage was to scribble on the tablet from behind, explaining that you can draw "circles and stuff."

Not really a big leap forward for public awareness of tablet PCs.

I'd have thought he'd have been further off to grab a remote desktop view of the tablet on another PC and pipe that view directly into the video feed so the show could cut to that view while he wrote. It would have been nice to see a real demonstration of the capabilities of the tablet, but we'll have to wait until next time.