Now, here's a switch

I spent the weekend hearing various folks talk about the latest (illegal) immigration flap. It was a little disconcerting to hear the left against helping poor people (e.g. “these people are stealing jobs from honest americans”) and the right for helping poor people (e.g. temporary work visas and some program to “legalize” those already here).

Frankly, I think there’s an ugly racist subtext to the whole debate. Almost everyone is talking about aliens crossing the Mexican border and almost no one is talking about: 1. The fact that potential terrorists have crossed the Canadian border and 2. The fact that a significant chunk of illegal immigrants (that is, aliens without a valid visa or work permit) come from Europe. So far, the only person I’ve heard talk about that is John McCain.

So here’s an intellectual puzzle for you: Why is it okay for trade to cross the borders without restriction (i.e. NAFTA), but not workers? Discuss amongst yourselves

 

4/3/2006 9:44:30 AM (Pacific Daylight Time, UTC-07:00)      Comments [0]  politics

Alistair Cockburn has an interesting article at his site...

Alistair Cockburn has an interesting article at his site: Are Iterations Hazardous to Your Project?

Hmm. It turns out that this article isn’t particularly new (from 9/9/2005), but still, very interesting (I found out about this from Sam Gentle’s site).

I took away, two bits of information from this article:

First, it shows how people can go wrong when trying to implement an agile development strategy, and (in fact) how they could succeed without doing agile development per se. It helps to underscore the important distinctions between “how to success” and “how to fail” beyond just “do agile” (what bad grammar!).

The second bit of information was more intriguing: The article is essentially a “fine tuning” of agile development. Instead of talking about “why agile development is good / better than waterfalling”, it talks specifically about how to be successful within agile development. This shows just how much agile development is moving to be mainstream.

A recent SDTimes article said that of surveyed companies, 19% were doing agile development. It sounds pretty low, but I wonder what other categories people answered with. How many were doing RUP-style I/I development, how many were doing Waterfall-style development, and how many were not really following any formal development strategy (I suspect the last category was the largest)? I think that the next 5 years are going to see a dramatic change in how mainstream software is developed.

3/27/2006 4:25:44 PM (Pacific Daylight Time, UTC-07:00)      Comments [0]  development

Interesting legacies

I just finished listening to Leo Laporte’s This Week in Tech (TWIT) and Scott Hanselman’s Hanselminutes. I think its interesting that both of these “cutting edge” “technology shows” include theme music. This is sort of a hold over from radio shows, and now television (though some television shows play with the format, placing the theme music later and later in the program. The theme music doesn’t really add anything to the program, since its self contained anyway. It’s not like we need something to inform us that the show is coming. And yet, the programs would seem less professional if they were missing.

As an aside: yes, this is one of the ways I get my tech news. I think its fantastic that I can simply download these shows to my Palm LifeDrive without the hassle of dealing with TV or radio networks.

3/24/2006 4:43:09 PM (Pacific Daylight Time, UTC-07:00)      Comments [1]  general

Open Source

Scott Hanselman has an interesting post, commenting on the Economist’s article on Open Source. One interesting quote is

“The contributors are typically motivated less by altruism than by self-interest.”

Well, duh.

Sure, the Open Source marketing guys (and they do exist) want to world to think that participants in Open Source are all selfless saints, but the ugly fact is that we all participate in Open Source development to get something out of it. Whether its a working system or experience or even the good feeling you get when you see your name in an About box, its all about self interest.

The “beauty” of Open Source isn’t altruism, its the high degree of collaboration that’s possible. There’s no worries from the sponsors about the secrecy of the intellectual property; it all open for gosh sake! That’s the idea.

Again, contrary to what the Open Source marketers like to say, Open Source software isn’t intrinsically better than non-Open Source software, its just cheaper (and I don’t mean free, though it can be that too). Therefore the price/performance curve can be radically different. If I have to pay $400 for a piece of software, it darn well better work satisfactorily. However, if the software just costs $25, I can live with some glitches while the folks that produce it learn from their mistakes. This is what Open Source is really all about. Okay, so, technically, I could dig through the source code, find the “problem area” and fix it (then submit the fix back to the community), but (as any experienced developer should be able to tell you) this a lot easier to say than to do!

I think Open Source is here to stay, and continues to have an impact on how software is developed today, just as Borland’s low price models in the ‘80s changed the way software pricing is done. However, the thing I don’t know is what form Open Source development will take in the future. I sure had no idea that it would be where it is today when I was playing in the sandbox fifteen years ago.

3/23/2006 4:06:06 PM (Pacific Daylight Time, UTC-07:00)      Comments [0]  development

I've made the change

Well, I’ve gone and done it.

I’ve installed Comcast’s DVR. I haven’t retired my ReplayTV units, instead I’ve added the Comcast unit. While not as good as ReplayTV in some ways, it’s much better in others (like, being available for instance  ). It handles conflicts much more rationally than the ReplayTV software does. And, as an extra bonus, I can watch both HD channels and OnDemand again (for some reason, they wouldn’t work with my ReplayTV plugged in to the cable box).

As soon as I get my new 8 way amplified splitter, I’ll be in hog heaven with a total of 5 (five!) recorders hooked up (the Comcast unit can actually record 2 shows at the same time!) I’m not missing anything now!

3/22/2006 7:20:10 PM (Pacific Daylight Time, UTC-07:00)      Comments [0]  TV | general

So, what's smaller than monogamy?

I’m reading a lot of “chatter” about polygamy on the internet, in large part thanks to HBO’s new show, “Big Love”. I think its interesting that a lot of the “anti-polygamy” sites are connecting legalized polygamy to legalized gay marriage. Now, to be fair, I’m all for gay marriage; but I’ve mentioned the same argument. After all, if you’re for gay marriage based on the “privacy” issue (what consenting adults do in the privacy of their own home is their business, and the government should stay out of it) or the “religious” issue (separation of church and state prohibits the government from criminalizing something simply because one or more religions consider it sinful), then what about polygamy or (consensual) incest? Why should it be legal for two unrelated guys to be married but not a brother and a sister? Or, for that matter, two brothers (no “genetic” problem there)?

That’s not really related to the title of this piece.

What is related, is the (supposed) fact that (as of 2000), 16% of all households (in America, I assume) are single parent ones. Over the years, single parenthood has become more mainstream. In fact, we consider single parent families “legitimate” today. So, what kind of marriage is this? It’s not monogamy, and certainly not polygamy. Just where do these families fall in the taxonomy? And is it relevant?

As for me, I’m all for heterosexual marriage, gay marriage, polygamy, etc. The issue is about stable families, not who’s hopping into bed with whom. What about you?

3/22/2006 4:26:05 PM (Pacific Daylight Time, UTC-07:00)      Comments [1]  general

Microsoft: changing direction without changing direction

In Computerworld today, Bill Gates is quoted as saying that MS waited too long for a new browser release. Yeah, I guess it’s “too long” when you start working on the next release after you’ve announced that there will be no new versions of the browser  . Okay, okay. They said “stand alone versions” of the browser. Of course, IE 7 will be available for XP. That sounds pretty stand-alone to me.

Anyway.

I think it’s interesting that this time MS is spinning this as a late, but intended change rather than a huge shift in strategy, like their move to the internet was in the ‘90s. This is a company that spent two years trying to convince the development community that web development was dead, and their “one click” technology would replace it. Of course, they changed their tune when the community started beating the AJAX drum. They quickly announced their own “AJAX product”: Atlas.

More and more, I’m thinking that Microsoft’s 15 minutes of technological leadership are over. However, I’m more than a little nervous about what/who comes next.

3/21/2006 9:29:01 AM (Pacific Daylight Time, UTC-07:00)      Comments [0]  development | general

Daria sighting

Okay, maybe sighting isn’t the right word, since she was heard, not seen, but…

I just got a voice message from Alan. He thought that he’d heard Daria O’Neal on 105.1. I just checked out their web site, and sure enough, Daria’s back on the radio! Bravo to The Buzz for recognizing real talent.

3/18/2006 7:23:43 PM (Pacific Daylight Time, UTC-07:00)      Comments [0]  general

SD Expo: Its over (again)

Now that SD Expo ‘06 is over, I’d better take the time to finish writing up some experiences

Wednesday, Day 1 of Classes:

I started off Wednesday by hearing Scott Ambler’s talk on Database Refactoring. Scott’s worked out a series of refactoring tasks, similar to Martin Fowler’s code refactoring that Martin covered in his book by the same name. Like Martin, Scott now has a book on the topic. I’d already pre-ordered the book (it should be waiting for me at home as I write this), so I didn’t take a lot of notes; I”m planning on reading the book when I get back.

During the second part of the morning, I went to a roundtable discussion on the future of Web application development run by Christian Gross. I know I’ve heard Christian before, his accent is unmistakable, but I can’t remember when. I’m sure it was a previous SD Expo. Christian mainly wanted to talk about REST, with a little side trip to AJAX. Nothing substantive came from the session, but I did get interested in REST. Since Christian had two sessions on AJAX and REST on Thursday, I made a note to attend those and get updated on this stuff.

During lunch, I attended a panel discussion on Model Driven Architecture (MDA) and Model Driven Design (MDD). The MDA concept has been floating around for at least 3 years now. It’s OMG’s next big thing after UML, and they’ve been pushing it hard. Until now, It’s been mainly smoke and mirrors. I’ve been skeptical about it, but I figured that I’d give them another chance to make their case. Besides, Scott Ambler was on the panel (definitely anti-MDA), so it looked like it would be entertaining at least.I couldn’t have been more wrong. At the panel, Scott pointed out several issues with MDA (for instance, most teams don’t all have modelling skills), but the rest of the panel just ignored him. The other panelists were obviously there simply to push the technologies that their companies were selling, not to really discuss the pros and cons of MDA. In fact, the Microsoft guy (who said he really wasn’t an MDA guy, though he cur knew how to talk the talk) went so far to say that anyone who didn’t know how to model should be looking for new work at Home Depot. I couldn’t help wondering how many developers Microsoft has laid off because they didn’t know how to model. I also wondered just how much of Office 12 existed in a model. This was a major waste of my time. The one good thing that came from this was that, as far as I’m concerned this was the last straw. MDA is, in my humble opinion, just a scam to sell consulting at high prices, the old Schlaer-Mellor scam of building dedicated translators for each company. It’s no longer worth my time.

After lunch, I went to a couple of ASP.NET 2.0 sessions. Once on Personalization and one on Web Parts. My study group has been dabbling in ASP.NET 2.0 with very little success, so I though I’d see if the experts could shed any light on the subject. I was pleasantly surprised to see that they were. The Personalization feature in ASP.NET seems to be pretty darned good. By making use of the web.config file and a back-end provider (the default is SQL Server 2005 Express), the system makes it easy to add type-safe properties to the system. I’m all set to share this with the study group! Web Parts turned out to be the reverse. It turned out to not be what I thought it was. It’s similar to the “web parts” in Sharepoint (yuck); I’d thought that it was more of a way to build specialized HTML emitters, like is done in custom controls in 1.1. By the way, the web parts built with ASP.NET 2.0 is incompatible with the web parts in Sharepoint. It won’t be until the next version of Sharepoint when Microsoft unifies them. Nevertheless, I’m feeling a little better about ASP.NET 2.0 than I’d felt earlier. I wish I’d gone to the previous presentation on the security features in ASP.NET 2.0, since that was what we had such a hard time with in my study group. Live and learn.

Wednesday night was the annual Jolt awards. I wasn’t able to take notes on the runners up, but I was able to note down the winners when I got back to my room:

Book, General: Prefactoring by Ken Pugh
I’ve known Ken for a while. He’s a veteran of SD. I remember sitting with him in a Birds of a Feather session back in 1997 on whether Java was going to go anywhere or not  . I actually bought the book at the conference. Any other books, I just ordered online from BN.COM (boo Amazon! boo!). That way, I saved the California sales tax and the hassle of carting the books home.

Books, Technical: Agile Web Development with Rails by Dave Thomas et al
Ruby on Rails is hot right now, and it showed here. This is sure a different attitude toward Ruby than when I first read about it back in the late ‘90s. In those days, no one cared; they were too enamored with Python. Rails changed all that. I just might have to revisit Ruby, though (apparently) Rails doesn’t play well with IIS (I’m sure it’s all Microsoft’s fault  ).

Enterprise Project Management: WelcomRisk 2.6 from Welcom
I forgot what went here. Had to do some research on the web. I think it’s interesting that no one has posted the winners yet, other than that the individual winners have posted blurbs announcing their own award. I wonder why not. Anyway, the winner is some sort of risk management tool. I don’t know why we have two project management categories. By the way, this is the only category that Microsoft was nominated in where they didn’t take home the jolt award.

Database Engines and Data Tools: SQL Server 2005 by Microsoft (you remember them)
This was the start of a trend. Can you think of anything else that Microsoft released over the past year? Yes, they’re coming up  .

Defect Tracking, Change, & Configuration Mgt: Perforce SCM from Perforce
I know absolutely nothing about Perforce, except that they’ve gone on the record, saying that commercial tools are superior to open source tools. One interesting sidelight: last year, FogBugz 3.1 won the Defect Tracking award (that was rolled into this new category). This year, FogBugz 4.0 was a runner up.

Design Tools & Modelling: Lattix LDM from Lattix
This is another tool I’d never heard of before. They beat out Borland’s Together 2006 for Eclipse (Borland’s walking dead anyway) and Altova’s UModel 2005. Altova’s product looks interesting to me (they just added C# support). It’s a lot cheaper than the others, which I think should be the trend here. Hopefully, they’ll do better next year.

Development Environments: Visual Studio Team System 2005 from Microsoft (here they are again)
Yup, they did it again. VSTS is a hugely expensive product, practically a CASE tool with a built in compiler. I guess if you’re going to buy a CASE tool anyway, you might as well get a compiler and code editor with it. Again, I’m waiting for the Open Source version.

Libraries, Frameworks, and Components: .NET 2.0 from Microsoft (detecting a trend here?)
Nothing else on the list came close to the impact that .NET 2.0 will have, so this one was a no brainer. Even given my qualms about ASP.NET 2.0 and VS 2005, the framework is a real winner, with its superior robustness, support for generics, and support for dynamic languages.

Mobile Development Tools: Crossfire 5.6 from AppForge
I’d be using this myself if it wasn’t so darned expensive. It lets you use C#/VS for cross platform mobile development: Palm, Nokia, MS CE, Symbian, and now Blackberry. Cool stuff.

Quality Project Management: Rally 5.6 from Rally Software Development
Rally is an Agile project management system, so I was very interested in this. I’d seen their ads for quite a while. I kept wondering just how much “management” an agile project required. As it turns out (according to my recent experience), a nice management tool can help. I stopped by their booth at the show. The tool looks pretty good, but they use a hosted, subscription model. Probably too rich for my blood. Too bad. It looks like they’re onto something. I’ll just wait for the Open Source version  .

Security Tools: Elemental Compliance from Elemental
Beats the heck outta me what this is. One of those new security thingies that everybody is sighing over.

Testing Tools: VMTN Subscription from VMWare
They really deserve this one. VMWare’s VM Workstation is the best addition to our test system we’ve made, and VMTN is the MSDN of VMWare. Very very cool.

Utilities: Camtasia Studio from TechSmith
This is an application recording tool, excellent for building training and marketing materials. I’ve used their screen capture product.

Web Development Tools: Rails 1.0 from rubyonrails.org
The one truly Open Source winner this year. Good thing ASP.NET 2.0 wasn’t up for the award  .

So, Microsoft had a near-perfect sweep of awards this year. They even won the Hall of Fame award for VS Professional. The folks at SD were saying that this was the end of VS Pro’s reign as the IDE of choice (implying that this was going to pass to VSTS), but I’m not convinced. Frankly I don’t see a whole lot of companies paying the big price for VSTS when VS Pro plus a few open source tools gets you everything you need. We’ll see…

Anyway, that was the Jolt awards.

Thursday:

On Thursday I spent the day at several AJAX presentations. The one thing I took away from the sessions was that a lot of people wanted to know about AJAX, and a lot of people promised to talk about it, then didn’t.

There were 2 hot topics this year: Ruby on Rails and AJAX. Of the two, AJAX was the leader by far. But no one had anything original to say about it. I think there are two problems with AJAX right now:
1. The basic technique is nothing new, and the underlying technologies have been around for a loooong time
2. AJAX as a “product” is still not ready for prime time. This reminds me of the first SD where XML was being talked up. No tools existed yet and it was all “promise”. It’s still something to follow though.

Friday:

I spent Friday morning at a tour of Microsoft Research. They’ve got a lot of neat stuff going on there. I’ll have to expound on this in detail later, because it really is some nice stuff, for .NET developers…

At lunch, Bob Martin was the keynote speaker. He called his talk the Prime Directive: Always Make Progress, Never be Blocked. As usual, he was excellent. Bob has to be the best speaker out there today. He got me all fired up again. It reminded me that I don’t just come to SD to see what’s going on in the industry; I also come here to get all fired up again about software development. His talk really got me going again. Can’t wait to get back in the saddle next week

In the afternoon, I went to Ken Pugh’s talk on HTTP. You know, I thought I’d pretty much understood the HTTP protocol. Boy, was I wrong. This turned out to be another one that’s going to have to take some digesting on my part.

So, that was SD 06. I’d been on the fence whether this conference was really worth coming to again. I completely skipped it last year. It turned out that I was right to come back. I had a great time, and have some ideas on how to move my development project forward…

3/17/2006 10:10:15 PM (Pacific Daylight Time, UTC-07:00)      Comments [0]  development