Microsoft or Dell or one of Apple's competitors should do the following...

Hire that annoying "Hi, I'm a Mac" guy from the Apple commercials. Have him talking a-mile-a-minute about how great Windows now runs on the Apple Mac. Using Parallels, you can't tell the difference. Now you can run all your Windows software inside your Mac.  On and on about how, now that Mac users have access to Windows, they will never need anything else.

And after all that, have the poor "Hi, I'm a PC" guy say, "So why don't you just get a Windows PC then?"

Seriously, I just listened to a podcast on Parallels by Steve Gibson and Leo Laporte (Security Now), and they went on and on about how fast Windows now runs on a Mac.... but didn't spend any time discussing the downsides to such a setup. "So if you need a fast version of Windows so badly, why don't you just get a Windows PC?" was all I could think at the end.

OK, perhaps saying "too easy" in my last post was asking for trouble. Because trouble is what I have.

I have a weird situation and I need some help figuring out what the solution is... Any clues, please email me or leave a comment.

We have a "base page" that inherits from System.UI.Web.Page. My base page is called "BasePage.aspx".

All other web pages of our application inherit from that. So "default.aspx" inherits from BasePage.aspx.

So my problem is that the codebehind page for BasePage, called BasePage.aspx.cs, has been saved in the App_Code folder. Now I don't know this folder or what it's for, thats my first problem.

Now all the pages that inherit BasePage.aspx now fail because the class doesn't exist.

If I move BasePage.aspx.cs out and put it at the root folder, then another page breaks about the BasePage class (one of our User Web Controls).

Ugh. I am sure once I figure the underlying problem out, fixing all pages won't be a problem. This isn't a problem relating to the number of pages, it's a problem every web site would have during a migration with base pages, web controls, etc.

So here is the status of my .NET 2.0 application conversion. I converted the 71568 lines of C# code and 38452 lines of VB code in the business object and services tier with little error. No errors really, just a few things that Visual Studio 2005 warns about that Visual Studio 2003 did not.

When I got to the web portion, I started finding lots of interesting errors. It's going to take a full day or so to go through them all.

All in all, it's not as scary a proposition as I thought. It's the kind of conversion that can be completed in a couple of days. I thought it might have required weeks before.

We have a rather large project that currently runs in ASP.NET 1.1, and we want to convert it to 2.0. Now it's not huge compared to some projects, but it's pretty big.

71568 lines of component C# code

38452 lines of component VB code

100401 lines of ASPX C# code

It should be fun seeing how the .NET conversion wizard handles this... I'll document my progress here as I go through it for the sake of posterity and the benefit of anyone else contemplating a rather large conversion.

Microsoft recently released a toolkit called Atlas at http://atlas.asp.net/

Atlas is an AJAX framework for ASP.NET 2.0. If you don't know what AJAX is, it is best described as a web programming technique where the page appears to update itself (or actually does update itself) without a full refresh of the page by the server. This can be automatic or at the request of the user viewing the page.

For instance, the up-arrow images on the left side of this page are AJAX-type controls that allow the user to alter their view of this page. This happens without a trip back to the web server.

An example of automatic updating would be a page with stock quotes embedded inside a news article (like http://www.marketwatch.com/news/story/5x12BB8tQmtNw49x3nlsrfJ?siteid=mktw&dist=TNMostRead ), and every few seconds the stock quotes update themselves with the latest values, but the rest of the page stays the same. In this case, the web browser is actually going back to the server to get a new quote, and updating a small part of the web page with the new value. The reason this is useful is that full page refreshes are slow, and annoy the user by making the page flicker or become unreadable.

For a live example of Atlas, check out the official samples page at http://atlas.asp.net/atlastoolkit/

Look at the Accordion control. As you click the headers, all other headers collapse and only the one selected is open. Or try the slider control - a control not normally available in HTML web pages. These controls are cross-browser compatible with IE and Firefox.

I saw a demo of this last week and it was very cool. http://www.msdnevents.com/ 

Crank is a movie that stars Jason Statham, the lead actor from The Transporter movies. It also stars Amy Smart as his girlfriend, who I swear I know from somewhere before. But looking at her resume on IMDB, I haven't seen anything she's been in... Maybe just her name sounds familiar.

Anyways...

The movie has an interesting plot: a hitman is poisoned in his sleep, and only has an hour or two to live. He quickly discovers that adrenaline can prolong his life, so while trying to exact his revenge he keeps having to find ways to pump up his adrenaline.

OK, let's start with the good bits of this movie.

* It was funny. There were several places where the audience broke out laughing. So for an action movie, some genuinely funny moments is a nice twist. 

* The cinematography was innovative. If you want to see an example of bad cinematography, take a look at Bourne Identity 2 - the camera was jumpy, and the action sequences were blurry. I thought Crank had some interesting camera shots. In one scene, Statham runs down an alley while one of his killers is on the phone with him. He actually runs past a video projection of that person on the phone. I thought that was a cool technique I hadn't seen before.

* It delivers on the action. So I went to this movie hoping to see some fights and car chases, and that's what I got.

Now let's be honest and talk about the bad bits too:

* Some plot holes. Well, let's be honest. The guy is given a poison that should kill him within an hour. Even though he discovers that when his heart is pumping and his adrenaline is running, it reduces the effect of the poison... there were plenty of places in the movie where his adrenaline should be lower. He's sitting in a restaurant with his girlfriend having lunch --- why isn't he keeled over on the floor from the effects of this poison? Lots of driving, lots of cab rides... The times when the poison affected him were not consistent...

* The poison itself. If you're going a kill a guy, kill him. Shoot him in the head or something. But why inject him with poison and leave a DVD movie for him to learn about his impending fate? Now I guess some bad guys have a huge ego, and love torturing people a little bit or rubbing it in someone's face that they are about to die. The movie never really explains the "why" of it's central plot theme.

(Well, I guess there wouldn't be a movie if they killed him in his sleep... that's a factor.)

Anyways, this is a decent movie. See it on DVD. Or download it. 6.5 out of 10.

I wonder if anyone has done this type of analysis on the show "Deal or No Deal".

If you haven't seen it, the premise of the show is that there are 26 suitcases with various dollar amounts in side, from $0.01 to $1,000,000. The contestant picks 1 case at the beginning, and the dollar amount inside is what they're playing for. They don't get to see this amount til the end.

Then they eliminate each of the cases a few at a time, and the banker makes them offers to get them to give up "their" case. If the contestant is eliminating low dollar cases, the offer can be high. But if the contestant eliminates high dollar cases, the offer is low.

So the first thing you have to notice is that the contestant can only win one of two amounts: the amount that is in their case (determined at the start of the game), and the amount offered by the banker. The amounts in unpicked cases are not available to the player.

The other thing to notice is that (presumably) the banker has all the information while the player has none. The banker knows the dollar amount in the players chosen case, and has a computer system to tell them the statistical average amount outstanding. And this goes into how much to offer the player.

But the banker doesn't care about the statistical average - he knows what the player stands to win, he knows what's in the case. And he controls the offers. So if the player only has $10,000 in his case, the banker would be stupid to ever offer the player more than that to get them to quit.

But of course the banker needs to keep up an appearance of fairness. His offer has to be within a reasonable limit of the statistical average. For instance, if there are 5 amounts left on the board, $1,000  $100,000 $250,000 and $500,000 and the banker offers $800, it will be a bit obvious to the player and the audience what the player has in his case.

But that doesn't mean that the banker has to be at or above the average. So in the above instance, a player with $1,000 in his chosen case might get an offer of $70,000, while a player with $500,000 in his case might get an offer of $200,000.

Now this is where I'd love to see real life examples of this. Can we get copies of past shows, and see where two players both have the same amounts left and different offers? Then we can see how the offer can give the player a clue as to what is in there case.

And if anyone knows where I can find more details on how the banker calculates the offer, put a link in the comments.