April 2003 - Posts
Mike Sax posted that "The Creative Commons License may ruin your life"and linked to my previous post about wanting a different copyright notice on my blog. Frankly, I didn't see what the big deal was. Then I read his link to Dan Bricklin's post about these "warrant" issues, and Yikes!
It got even closer to home when I followed Dan's link to the Lawrence Lessig keynote from OSCON 2002. This talks about how aggressive the lawyers with "The Simpsons" were about 2 seconds of film from a documentary about education. They wanted $25K!
I actually had a substantial discussion with these same people exactly two years ago. In the PNW, the start of yachting season is the first Saturday in May, when the Seattle Yacht Club hosts the Opening Day ceremonies. I was in charge of the decorated boat from Meydenbauer Bay Yacht Club and we wanted to do a Simpson's theme. You would not believe the legal restrictions. Of course, we had to license the right to use the likenesses. That was something like $10K. And then we had to use costumes that they had approved, which (big surprise) they also rent out at something like $2K each. Plus shipping. And cleaning. And these were professional costumes, which means that they only fit healthy adult males between 5'10" and 6'. And the wearers could only use them for a 45 minute stretch before taking a break. And they had to enter a closed area to remove the mask, since seeing a mask removed would destroy the "reality". And the characters could not speak - since they could only have the "correct" voices. So they could wave. But only in a manner appropriate to each character (there were written descriptions of this too). Of course, they _could_ have thought baloons with character-appropriate sayings. Which we would have to engage one of their editors (at $2K/day) to work on with our "creative team". And, of course, they would need to see every piece of communications that was put put for the event. So we were probably going to have to hire a media consultant.
A bigger bunch of greedy anal fuckwads I have never encountered. I just turned down a speaking opportunity in El Salvador because the State Department warned of "random banditry, carjackings, kidnappings, criminal assaults". Fill in the blank.
The "rest of the story"? Well, we decided on a theme called "A Mir Miss" where dummied up a rocket crashed into the boat. We towed the boat itself, and towed a landing "target" behind the boat. We had firemen and astronauts, fire hoses, and I even hooked up a smoke generator I had lying around (don't ask <g>). We won three awards, including:
- Best Yacht Club Sponsored Decorated (SYC Commodore's Cup)
- Grand Sweepstakes, Best Overall Decorated Yacht (Admiral's Trophy)
We were on local television and even (briefly) on the national news. The trophies were nice, but just thinking of the publicity "The Simpson" lost warms the cockles of my heart.
Keith Ballinger pointed out to me a couple of weeks ago that the "Copyright Keith Pleas" at the bottom of this blog basically doesn't allow anybody to do anything with my blog. Perhaps even quote it. I asked Scott - who runs dotnetweblogs - about adding a customization option for copyrights, but it apparently didn't make it into the new site he rolled out this weekend.
Now I see that Scoble has also gotten the message on this and is now using the Creative Commons license. Which looks like a reasonable solution to me.
Julia talks about
why she runs a .NET user group. I just couldn't resist making the connection to this past Sunday's
Dilbert cartoon about corporate "Customer Support Groups". <vbg>
Chris Kinsman and I taped an episode of the .NET Show on applications architecture with .NET. We did the show with Ed Jerzierski (of PAG) and Mike Burner (of the Platform Strategy Group), and it should be out in May. Lots of stuff to talk about: the architecture community portal I announced last week, new patterns and stuff coming from PAG, and what we learned from our first on-campus architecture event in March and how we're updating things for our next event (now called the "Guided Design .Summit: Architecture and Design for .NET").
Ironically, the coolest architecture stuff we talked about that day (but didn't get taped) were all the architecture blueprints on the walls of MSStudio (in building 127, same as the company store and museum). There were drawings for the Addams Family, the "Richard and Laura Petrie" residence, Bruce Wayne's mansion, Gilligan's Island, and a bunch more. Talk about a "way back" machine!
Another interesting experience was the makup. Turns out my color is "suntan", at least in the line of professional cosmetics they use. Of course, I live here in the PNW so I think it's more a reflection of something I need, rather than have. And the makeup girl darkened my eyebrows, patted my hair a bit (I'd just gotten a haircut that morning), and put on what seemed like a bunch color around my eyes.
Robert Hess was, as always, the host. In fact, this was his 33rd show. Someone joked that he'd been on longer than "Friends"! Anyway, Robert is Scoble's new boss starting (I think) next week, so maybe we'll see Robert on there at some point.
The new Microsoft
portal for .NET architecture and design just went live on GotDotNet. This is the community side of the
MSDN Architecture site and contains patterns, VS.NET enterprise templates, technical articles, discussion groups, and third party content.
As I mentioned in my previous post (emphasized in the comments by Mike), RSS "views" are the vast majority of accesses. And, yes, they tend to inflate numbers dramatically. The aggregator I'm using - SharpReader by Luke Hutteman - by default refreshes every hour.
So here are the detailed stats for my directory (again, for the week starting Sunday). Please forgive my quick "table" formatting <g>.
8808 /kpleas/
16 /kpleas/admin/
83 /kpleas/archive/
22 /kpleas/blogadmin/
10 /kpleas/category/
6 /kpleas/kpleas/
97 /kpleas/posts/
1 /kpleas/services/
Robert is waging a personal war against "traditional" industry pundits (particularly John Dvorak), postulating that they're afraid of the readership that blogs as garnering. So, just how popular are these blogs?
Scott Watermasysk (who runs the dotnetweblogs host that many of us use) has given me access to the statistics. As of today, for this week starting Sunday, I've had 8,709 page views of blog (mostly the RSS feed by a factor of about 10:1). And I've only been blogging for about 2 weeks! The other thing that stands out is that Scott Guthrie has the most page views - almost 47K, about the same as frequent blogger Sam Gentile - even though Scott hasn't posted anything since April 1st. Lots of people apparently waiting for Scott to post something to their RSS aggregator.
So, how many hits does Robert get? The userland statistics page shows he's had 341 page views so far today and a total of 193K page views in the approximately 1.5 years they've been aggregating statistics. Dunno if that includes RSS hits. Also, I just had coffee with Matt Carter (most recently VP of Fawcette's online operations, now - after taking the red pill - in charge of developer titles for MS Press) who said that while Robert's traffic can't be compared to a web site, the number of people who read his blog several years ago (when it was hosted by Fawcette spin-off DevX) was surprisingly high. More importantly, the names of the people who read his blog read like the "Who's who" of the industry. In fact, I've heard that Steve Ballmer has made public reference to his blog.
OK, thanks to some noted bloggers (Tim, Don, and Robert in particular) everybody knows that Chris Sells "took the red pill" and joined Microsoft. No doubt, this has caused some people to wonder about their chances of getting hired by Microsoft.
Well, if you work for Developmentor the chances appear pretty good: Tim, Don, and Chris all were associated with that company. I don't think Robert ever worked for DM, but I imagine he might have come close at least once during his peripatetic travels through the industry - and California - over the last few years.
What other associations stand out? Well, Chris is actually the third INETA Speaker Bureau (SB) speaker to be hired by Microsoft in the last few months (after Yasser and Asli Bilgin). Ironically, Chris is the guy who came up with the cute little "INETA Speaker" logo that a lot of us use on our sites. When you consider that there are only about 30 speakers in the SB, that's a pretty fair hit ratio considering the SB's been around for only 9 months or so.
Also, Chris was an MSDN Regional Director. Of course, lots of RDs do contract work for Microsoft. Chris Kinsman and I are right now working on the ASP.NET Hands-On-Labs for TechEd, for example. And the companies of RDs Tim Huckaby, Scott Stanfield, and Jon Rauschenberger (Interknowlogy. Clarity, and Vertigo respectively) have done tons of work for Microsoft over the years. Strangely enough, while all three of these guys were - collectively - featured in Microsoft's "Iron Developer" contest, none of them have a blog (perhaps "Iron" refers to the "iron age"? ). Are there other RDs who have been hired by Microsoft? There are around 150 of them worldwide, so you'd think there must be others.
So, speaking of the SB and RDs, what's the correlation between the two? Pretty high, in fact: of the current 29 members (subtracting Chris now, of course) of the domestic (US) SB, I count 10 of them as being RDs (though none of them, as far as I know, have blogs other than Chris Kinsman). And the proportion is actually much higher for LATAM: 3 of the 6 LATAM SB members are also RDs.In Europe, only 2 of the 11 SB speakers are RDs (both of them - Clemens Vasters and Christian Weyer - have blogs). Also, Bill Evjen - who runs INETA here in the US - and Christian Nagel (ditto Europe) are RDs (though neither have blogs). And I'm sure there are RDs - other than Scott Hanselman - who have blogs.
Hmm...looks like there's some interesting correlations between the INETA SB, MSDN RDs, blogging, and being hired by Microsoft. Especially if you work(ed) for Developmentor.
I showed a security example at VSLive this week that simply didn't work. It was pretty straightforward: I put a LinkDemand in a component assembly that looked for a key signature in the caller. It's always worked before, so I didn't think to test it ahead of time. Dumb.
So...on stage, I run it and it does not generate the expected security violation. Oops. I re-compile the solution (with three projects), but it blows off at runtime but not with the expected security violation. Hmm. OK, turns out to be VS's propensity for nuking the files in the build directory, easily fixed.
But it still doesn't generate the security exception! Must be a change in the RTM bits for 1.1 (which I'm running). Visions of a huge security bug dancing in my head, I head to bed. And...bolt awake in the middle of the night wondering if I'm somehow turned security off?
Yes! Somewhere along the line I ran something (a test batch file, or from the command line) and turned security off (using "caspol -s off"). I forget when I did this, probably just checking to see if something was truly a security-related issue. But I never re-enabled security. And there is no option to caspol to see if security is enabled. Clearly, turning security off survives a reboot, a re-installation of VS.NET, and even an OS upgrade. Ouch!
This week was VSLive Chicago. Pretty good turnout with lots of late registrations, which is encouraging for the industry.
I had dinner with Rocky Lhotka and Billy Hollis on Sunday night. Billy and I had a workshop on Monday on upgrading VB, Rocky's was on ASP.NET.
As you'd imagine, we talked mostly about developer stuff. And we agreed on almost everything, except for the design basis for putting application-related code into the database. As you probably know, this capability is one of the highly touted features of "Yukon", the next major version of Microsoft SQL Server. In particular, this seems to fly in the face of the loosely-coupled architecture long expoused by Microsoft's Pat Helland (who actually works on the SQL team) who's "Autonomous Computing" (a.k.a. "Emissaries and Fiefdoms") model has been well publicized by Microsoft (including this Webcast).
Clearly, putting code into the database layer implies a high degree of trust. And the loosely-coupled model is all about a high degree of cohesion within an "fiefdom", but a low level of trust between them. This plays well with a service-oriented architecture and Web services, which after all are just passing data around. Remoting, on the other hand, passes objects and reference around and obviously requires a higher degree of trust. In fact, Microsoft added a FilterLevel property to Remoting in version 1.1 to at least partially address this issue.
So, passing data around - which is inherently safe - also means that, in a suspicious world, you have to do data validation at every service interface. As Rocky pointed out, this is a repetitive pain in the butt and is not something that enhances the performance of your application. So, ideally, you'd use Remoting within your application for performance and conveniece, but something like Web services for going outside your "fiefdom". The trick, of course, is to draw those trust boundaries appropriately. And developers moving from a traditional DNA architecture (if there really is such a thing) are used to tighly-coupled designs. Unfortunately, Remoting is likely to be too complex for the average "application" developer (OK, I guess I'm talk about"Mort" here) while Web services are going to rob performance and - realistically - required extra validation code. Clearly, application developers have some difficult choices to make.
More Posts
Next page »