More opinions on SOA
Submitted by Chaz on Sat, 12/17/2005 - 2:24am. Links | SOAGregg Wonderly has some of the same doubts I do about the Web Services/SOA model.
Textile Testing
Submitted by Chaz on Fri, 12/16/2005 - 8:04pm. Drupal | TextileThis is a test of using Textile.
This demonstrates how you can mix the formatting options.
Here's what this looks like when I type it in.
_This ["demonstrates":google.com/search?q=demonstrates] *^how^ you ~can~ mix* +the formatting options+_.
Here is a link to some demonstrations of how to enter Textile information.
You vs. Process
Submitted by Chaz on Wed, 12/14/2005 - 7:05am. IT Process | LinksHere's an interesting article from Tristan Yates. This is the man who in September put up an interesting blog about how his IT bosses went with the safe choice. I believe they spent a quarter of a million dollars a week on IBM "talent" at $325 an hour which was hired from Monster and were being payed about $75.
Anyway, I think I like this article even better.
Why you're smarter than your process
I'll have more to say on it later. No time now.
This is my new blog.
Submitted by Chaz on Tue, 12/13/2005 - 11:06pm. Blog | Drupal | Spam | WikiAfter I became disenchanted with Movable Type, I experimented with Wikis.
I have determined that wikis kick ass, and yet they're not quite what I want.
What I want is
a) to be able to blog without thinking about anything but content.
b) to be able to add other pages and navigation with only a little more difficulty than blogging.
c) anybody can comment, but only if they are verified by email first.
d) I liked the usability of a wiki tremendously, but didn't care for the publicly visible history. Particularly once spam surfaced.
So I'm trying out Drupal (www.drupal.org), a free Content Management System that may help.
Visual Studio 2005
Submitted by Chaz on Mon, 12/12/2005 - 10:00pm. .NET | IT Process | JetBrainsI went to a Microsoft event today to see their presentation on the new Visual Studio 2005. I have to tell you, I've been using the new C# for months now in beta form - gotta love the generics support - and I love it. That didn't get a mention today. But I did learn a couple of things. And, hey, they gave me a free Visual Studio (standard), a free SQL Server (standard) and some variety of BizTalk thing. Incidentally, there was a point when they asked how many people in the audience used BizTalk, and I think two people raised their hands. From the various hand-raisings, I would wager about 60% of the audience was web developers. I found that disappointing. I'm all about the smart client, but my clients don't want to be the first ones experimenting there.
John Vlissides
Submitted by Chaz on Tue, 11/29/2005 - 10:00pm.John Vlissides passed away on Thursday, November 24th. I never met the man and haven't read anything of his in years, but I recall fondly his C++ Report articles from a few years back. And of course there was the Gang of Four. This guy was one of the real talents.
Ward Cunningham's wiki has a very touching tribute from many of his peers and colleagues and readers.
The new Yoda language
Submitted by Chaz on Wed, 10/12/2005 - 9:00pm. HumorThis is a joy to read. If you're amazed at how well Microsoft set us up for the new LINQ features in C# and want to know what's coming next, then you'll find this damn funny: Yoda the Programming Language.
Consultants Gone Wild
Submitted by Chaz on Sun, 10/02/2005 - 10:05pm. IT ProcessSome days, I wonder why anybody trusts consultants at all.
Here's an interesting unverified story and I aim to read the whole thing later and will probably comment more then.
Is it really low risk to go with the vendor who can do everything? You decide.
The Adventures of Action Item!
Submitted by Chaz on Sun, 10/02/2005 - 9:48pm. Humor | IT ProcessGreat comic at http://www.fatalexception.org/action_item.html.
Acceptance/Unit Testing Revisited
Submitted by Chaz on Fri, 09/09/2005 - 5:24am. Acceptance Testing | In Memory Database | Test DrivenMichael Feathers seems to have everyone talking about his latest pronouncement on what defines a Unit Test. He's quite correct. But there's still a lot to talk about.
The idea's simple. If your unit tests are fast and don't depend on anything slow like a disk or a network then:
In a few seconds, the compiler tells you that every function in your code code compiles.
In a few seconds more, the unit tests tell you that every function in your code still returns the expected results.
Simple, immediate feedback. You can't beat that. And it's more useful than the compiler's feedback.