Jeff and .NET

The .NET musings of Jeff Putz



My Sites


July 2006 - Posts

Atlas is too hard

Actually, Atlas is relatively easy to use, but the configuration sucks.

First off, what is wrong with the people in Redmond that they can't just give us a zip full of files. Why, in a world of Xcopy deployment, do I need these stupid installers just to get an assembly out?

Second, the configuration is borderline ridiculous. I realize we're in CTP mode right now, but why, oh why, are there dozens of lines that we need to add to web.config just to get this stuff to run? It's too much work. Extra work is bad.

The tools are pretty cool, but the deployment needs to get much more simple to be really useful. 

Posted: Jul 28 2006, 11:50 PM by Jeff | with 13 comment(s)
Filed under: , ,
Config versus Database

I was thinking a little about the pros and cons of storing configuration data in config files versus using a database. I have to tell you, that for all of the learning and work I did for POP Forums (like two years ago!) seemed pretty slick at the time, but now I'm not so sure that using config files is the way to go for most things.

I guess the change of heart came around because back in the day, and by that I mean the ASP.old days, we didn't have caching, and I had no idea what a static anything was. Sure, I stored stuff in the application object (*shudder*), but I still didn't really think the way I do now.

Config pros:

  • Simple text changes change config
  • Notepad r0X0rz teh big one!!!!1111

Config cons:

  • If you don't externalize the file from web.config, your app restarts on changes
  • Simple config sections suck for large blocks of text (like TOS or e-mail templates)
  • Who wants to wade through text files and not UI? (yeah, I know you can programmatically alter them)

Database pros:

  • Cache your junk
  • Big giant text field capable

Database cons:

  • Not as trendy as .NET config files
  • Potentially big load on app start (though I don't know it's worse than config files)

 I guess in the end it doesn't matter, but what's starting to concern me is that the more crap you have in config files, the harder they are to manage. We already have to put all of our HttpModules and HttpHandlers there, Atlas adds a huge ton of crap, sometimes you still need those lame AppSettings things, etc. It's just so non-tidy.

Yet another .NET vs. Java thread

People really get a fire in their panties about this kind of thing. I'm not sure why. Check this one...

Granted, having a 'Softie inflame people doesn't help. But really, .NET is pretty awesome. I think it would be even more popular if there wasn't senseless hate toward Microsoft, and if Microsoft had a single clue about how to market it. They're getting better though by using anti-marketing tactics. Having Scott Guthrie blogging, for example, is enormously helpful.

The way I see it, the haters don't know what they're missing, and that means more job opportunities for me.

Posted: Jul 28 2006, 02:44 PM by Jeff | with no comments
Filed under: , ,
Comment away!
It turns out that anon commenting was turned off on the blog when they upgraded. Boo. I wondered why no one ever commented anymore.
Posted: Jul 27 2006, 10:18 AM by Jeff | with 1 comment(s)
Filed under:
Missing TechTV

Leo says he'd love to do a TechTV reunion. God do I miss that channel. I sat down to watch The Screen Savers pretty much every weekday for almost three years. I couldn't get enough of it! The personalities on that show were top notch geeks. Over the years we always had Leo and Patrick, then later Kevin, plus Megan, Jessica and Sarah. Yoshi, Roger and Robert contributed good stuff, and even Martin was entertaining to some degree.

But G4 and Comcast hosed the channel, and they deserve the crappy ratings they're getting.

Pirillo and Scoble have differing views on what made TechTV great, and what is good about life after the network. I tend to agree with Scoble. It might be sad that I don't get The Screen Savers in my living room every night, but I don't think it's hard to argue that This Week in Tech, Digital Life TV and Digg owe their roots to the demise of the network. The sum of these things are far greater than TechTV ever was, or likely could have been.

I'm an experimental innovator

You know, the July issue of Wired was just full of really good stuff. It has been ages since I've read nearly every story in a magazine.

If you read my blog, you know that one of my big struggles in life is, well, struggling through life. I really do think that I'm a pretty smart guy, maybe a genius, but I look around and see people my age or younger who have done far more than I have. It bothers me. If I'm so ****ing smart, then what's holding me back?

So this economist, David Galenson, has this theory that there are two personality types in terms of how your genius comes to fruition. The first type is the "conceptual innovator," a person who bucks trends and creates great things early in life, then spends decades, well, not. The other type is the "experimental innovator," who goes through life in a long trial and error mode, but eventually does great things. Of course there are people who will always be a useless piece of ***, and the rare life-long contributor, but historically, and demonstrated through Galenson's data, people tend to fall into one category or the other.

I wouldn't say that I'm unhappy or pissed off or whatever that I haven't achieved what I think I want to achieve (seeing as how I don't even know what that is), but I do get frustrated relative to people in the tech world doing amazing things. (Also, some people tell me I've already achieved big things, whatever that means.) Still, this research really makes me feel a lot more positive, because as an experimental innovator, clearly my time will come. I need to sit back and enjoy the ride a little.

After I read the article, I was thinking about which I would want to be if I could choose. To peak early in life might be kind of lame. Maybe that's why creative geniuses who do it in their 20's kill themselves. If life's a journey and you actually reach a destination, what the hell is next?

37signals takes money from Jeff Bezos

The company that prides itself on being profitable and not taking money from investors took money from one of the biggest investors period.

Check it here.

Staggering self-publishing numbers

37singals posted some numbers on their self-published book today. I don't think I need to really add anything, except to say you should look at the numbers versus their dead tree publication.

And I understand that the company is popular now, but still. That's pretty amazing stuff.

A four-hour software challenge
I was thinking about stuff that I need to code for my sites, and was thinking about how I'd like a place for the public to track bugs. Yeah, stuff like that already exists, but it'd be nice to have something more basic.

Then I realized it might be fun to crank something out like that for myself. As fast as humanly possible. Like four hours or less. Could I do it?

I started to ask myself what it is that takes the longest for me. It's almost always the data access code, and more specifically, the test code to make sure it all works. I know that I test wrong by normal standards, in that my data access code is always dependent on the business objects that call it as containers. So a method that gets data from a table of users is called by a User object, and expects a User object from the data call. Most "experts" would say that's bad design.

So if I can let go of the intense testing, maybe I'll give it a try.
Posted: Jul 20 2006, 01:23 PM by Jeff | with no comments
Filed under: ,
Anyone using Visual Studio in Parallels on a MacBook Pro?
Anyone using Visual Studio in Parallels on a MacBook Pro? Right now I'm doing the BootCamp thing, but obviously virtualization would be a better solution.
Posted: Jul 20 2006, 11:06 AM by Jeff | with 45 comment(s)
Filed under: , ,
More Posts Next page »