Visit jaxn.org

My other blog

jaxn — Thu, 21 Dec 2006 06:15:56 +0000

My blog is available at jaxn.org.

2004 Electoral College Projections

jaxn — Fri, 29 Oct 2004 18:10:11 +0000

I have been doing a research online and have compiled my projections for the 2004 Electoral College votes. There is a reasonable chance that the 2004 presidential could be an electoral college tie on November 2nd, though that is not my projection. I thought I would post a summary of how this could happen.

There is no question whether or not the 2004 presidential election is going to be close. Like many people in the world, I am almost consumed by the potential outcome of this election. Lately I have started to feel a little more like Kerry is actually going to win this one, but it is just too close to call. Of course, that doesn’t stop me from trying.

I took a little time and researched the polls and the electoral college. I have compiled my predictions, based on the most recent polls, of how the electoral college votes will go after the November 2nd election.

Maine and Nevada allow their electoral college votes to be divided based on a district system. The other 48 states and the District of Columbia have a winner take all system for electoral college votes. Colorado has a measure on the ballot to move to a district based system. Maine could go either way. I gave the state to Bush but gave one district to Kerry, this gives Bush 3 votes and Kerry 1. I don’t think that Nevada will be close enough to split it’s vote. Since Colorado is currently a "winner takes all" state I gave all 9 of Colorado’s votes to the current winner (Bush).

Out of the 51 electoral districts (50 states + DC), Bush is projected to win 29, Kerry will win 21, and Maine will be split. Yet, out of a total of 538 votes it looks like there is no way the final tally can be more than 15 votes (less than 3%). My projection has Kerry winning the electoral college by a mere 8 votes (less than 1.5%).

It is likely that there will be recounts and court battles after this election. It could get ugly. So, for extra credit I am going to also make predictions as to where it is going to get ugly.

Alaska (3) and Hawaii (4) will be the last states to report since they have the latest timezone. However Alaska is solidly Bush and Hawaii is solidly Kerry, so there shouldn’t be a cliff-hanger in that sense. Following along the same lines of timezone drama, Nevada (5) and New Mexico (5) are the most westward swing states. I don’t think anyone will be able to make any attempt at calling the election until those states are reporting. They both could go either way. They both have 5 Electoral votes. They both could be the deciding factor on election night.

The 2000 election taught us that it is not all over on election night though. While I don’t think Florida will be ground zero again, voting procedures and policies will most likely be the focus of lawsuits in multiple states after the polls have closed and the results have been estimated. The next president could well be decided by the Supreme Court again, though not until we watch lawsuit after lawsuit and appeal after appeal.

Many states already have lawsuits and appeals related to the election active in the courts. The most notable of which is probably Colorado (9). A citizen has filed a lawsuit to try and block the measure on the ballot that would move Colorado’s electoral distribution from a “winner takes all” system to a district system. Under the proposed measure, 5 of the 9 votes would go to the candidate who wins 55% of the popular vote in the state. the rest of the votes would be divied up by district results. MoveOn.org has targeted a last minute ad campaign in Colorado to try and gain some more votes there. If the election is close, Colorado could be very pivitol.

I also have high hopes for Pennsylvania (21). It is a divided state that is currently a statistical tie. Pennsylvania has Philadelphia, and Philly has an inner-city population that is known for organizing (KWRU). It is also the state I think could be most effective by Sean (Puffy) Combs’ Vote or Die campaign. Philly is hip-hop, and Philly is undecided. There is also a large population of college students in Philly that may be first-time voters. I don’t know as much about Pittsburgh, but I am giving Pennsylvania to Kerry.

So here is the summary of my predictions:

John Kerry will win the Electoral College by less than 5%
The election results will be contested in multiple states
Maine will split it’s electoral college votes
Colorado is the state most likely to end up in the Supreme Court
Pennsylvania is the state most likely to get a visit from Jesse Jackson
Other Tennessean’s will be wrong

There you have it. Only time will tell, but at least this will be in Google’s cache.

2004 Electoral College
Kerry	State	Bush
	Alabama	9
	Alaska	3
	Arizona	10
	Arkansas	6
55	California
	Colorado**	9
7	Connecticut
3	Delaware
	Florida	27
	Georgia	15
4	Hawaii
	Idaho	4
21	Illinois
	Indiana	11
	Iowa	7
	Kansas	6
	Kentucky	8
	Louisiana	9
1	Maine*	3
10	Maryland
12	Massachutusetts
17	Michigan
10	Minnesota
	Mississippi	6
	Missouri	11
	Montana	3
	Nebraska*	5
5	Nevada
4	New Hampshire
15	New Jersey
	New Mexico	5
31	New York
	North Carolina	15
	North Dakota	3
20	Ohio
	Oklahoma	7
7	Oregon
21	Pennsylvania
4	Rhode Island
8	South Carolina
	South Dakota	3
	Tennessee	11
	Texas	34
	Utah	5
3	Vermont
	Virginia	13
11	Washington
3	Washington, DC
	West Virginia	5
	Wisconsin	10
	Wyoming	3
272	total	266

Mosh

jaxn — Fri, 29 Oct 2004 16:14:31 +0000

Last night on some pseudo-news show on ABC there was a story about a new political music video/song from Eminem. There were many clips of the video, but they did not play it all the way through (granted, for language reasons they couldn’t). A couple of weeks ago Marcus told me that he heard some new Eminem song that was sick, but I hadn’t heard it. After seeing the story about it, I had to go get the video.

Moby is even putting his hard feelings behind him and supporting Eminem by encouraging his fans to Watch the video.

I liked what he has to say in the song. I hope it has some influence on voter turnout.

Of course Miguel has something to say about it. Mosh is #1 on mtv.com today. Salon.com has an article. Yahoo details the popular growth of Mosh. Google search results for Mosh.

Congrats BoSox!

jaxn — Thu, 28 Oct 2004 14:55:24 +0000

I am not a huge baseball fan, but it is cool that the Red Sox _finally_ won another World Series. The last time the Red Sox won a World Series it was 86 years ago.

Comparing Tapestry to CEP

jaxn — Wed, 27 Oct 2004 17:04:26 +0000

I am sure my co-worker as well as my cohort are sick and tired of hearing my comparisons, but as I start developing with Tapestry I can’t help but see some striking similarities.

The first similarity is that both Tapestry and CEP set out to provide web developers a way to develop component-ized/modular applications that emphasizes code re-use. CEP was born out of a desire to develop components out of components that can then be re-used in multiple parts of the application by just calling the component. My very first implementation of this was on the now-defunct cataBlog in 2001.

About a year ago I registered Core Enterprise PHP at Sourceforge so that Marcus and I could implement the next generation of this modular “framework”. We have been developing CEP and developing in CEP ever since.

CEP turned out to be vastly different from cataBlog. CEP gave us a clear way to separate our code. We developed a component life-cycle that took place for every module/component that was active for a request. We were able to stop thinking about URLs and pages and start to think about events and components. One of the most important aspects of CEP was that was that it was object oriented, and this allow us to hide much of the complexity from the application developers. From what I hear, this complexity mask has be very helpful in allowing developers to work on a CEP project without having to know much more than the business problem and how to write PHP.

We have decided to give Java a try for some projects at work. When we first started looking at Java I was very unsatisfied with the development methodologies that were feasible for a small team. I wanted to be able to develop modular applications in a object oriented way, and I didn’t want to think in terms of URLs and top to bottom page rendering. I didn’t even want to think about pages. When I first looked at Tapestry I saw a tag-based WYSIWYG language similar to ASP, JSP, and most PHP. We even started to implement a project in JSP, which was very frustrating for me because it just doesn’t allow me to develop the way I have become accustomed (which I also feel is vastly superior to the way most “web apps” are developed. After taking a deeper look at Tapestry we decided it would be close enough and started to give it a try.

So what are these similarities you ask?

Re-usable component based development
Tapestry’s Engine and Visit objects provide similar functionality to the “Local Classes” in CEP which allow application-wide behavior to be defined/implemented once
Tapestry may do a better job of it by providing listeners, but CEP also masked URLs using cepCreateUri()
Components can include other components
Application flow redirection without sending http headers
Similar steps in the request processing (cepPreRender() vs pageBeginRender()

I will list more if/when I find them.

I should say this. CEP modules may be a little easier to develop than Tapestry components, and there is less of a distinction between CEP modules and pages than there is in Tapestry (meaning modules can be pages and pages can be modules with no alteration in CEP). Tapestry is hands down a more mature system.

There is also one thing I would like to see implemented in Tapestry (or figure out how to do it if it is already there). I would like for there to be an equivalent to EventState methods. In CEP, EventState methods are a way in which just be define a method you can add functionality to a particular state of a module/component. A good example is an add/edit form. In CEP the form can be defined in the module constructor and the in the edit state a renderEdit method would be called if it exists. This allows CEP developers to populate the form. Obviously there are ways to do this in Tapestry, I just think that the elegance of simply defining a method is simple and correct.

Having said that, I look forward to being able to post more things I like about Tapestry in the near future.

-Jackson

My Blogger Code

jaxn — Tue, 26 Oct 2004 17:04:27 +0000

I usually don’t do online surveys, and I don’t put any stock in them. However, I saw someone’s “blogger code” on their blog and thought I would generate mine. Here it is, for the whole world to see, my “blogger code”:

B9 d++ t+ k s+ u- f- i+ o x- e+ l- c– (decode it!)

Introduction to Tapestry

jaxn — Tue, 26 Oct 2004 13:49:56 +0000

I am playing with Tapestry at work. Today is the day that I am going to build a working Tapestry “app”. Unfortunately I can’t find a good introduction. Throughout the day as I find resources I am going to post them here so that I don’t have to look for them over and over again.

If PHP does one thing better than all other languages, it is the core documentation. Easy to find, includes examples, includes user comments.

Tapestry Documentation
Designing Tapestry Mega-Components @ OnJava.com is a two page tutorial that talks about using “The Palette”.
Illustrating Tapestry
Brownbag Developer Series:
- Creating a CRUD application (pt. 1)
- Enhancing our application using Hibernate (pt. 2)
- Enhancing our application using Spring (coming soon)

Who's Getting Your Vote

jaxn — Thu, 21 Oct 2004 17:46:09 +0000

I found a really interesting article today thanks to memigo.com. It led me to the website reason which is from the magazine reason. I think I may order the magazine.

Anyway, the article is a collection of responses to the following questions:

Who are you going to vote for in 2004?
Who did you vote for in 2000?
What was your most embarrasing vote?
Who was your favorite president?

Read Who’s Getting Your Vote to see some interesting responses.

I found the answers to be pretty informative. It was like reading a month’s worth of OP-ED pieces in one sitting.

-Jackson

PocketPC moblogging

jaxn — Thu, 21 Oct 2004 03:25:17 +0000

Blogging from my iPaq pretty much sucks. Iwould really have to want to share something fast to blog this way. All of the nice javascript stuff doesn’t work either. I have already committed to enough projects so I won’t even pretend that I may one day have time to create a .NET CF blogging app.

What I Think PHP is Missing

jaxn — Wed, 20 Oct 2004 18:05:33 +0000

namespaces: I agree that no implementation is better than a bad implementation, but I think this should be priority #1.
PECL extensions: What I mean by this is that there are things in the core that should be in PECL. I read a convincing argument of why PEAR/PECL package developers should support PEAR/PECL and not try and get everything in the PHP distribution. There is already too much core functionality available, and it would be difficult to move it out of the core, but I think that would be good.
better install: PHP has gotten to be a PITA to compile. Everytime I have to go through all the configure options, write/copy the ones I want, and then run configure. Often times I have to include things I _think_ I may want later (pcntl), but then find I have failed dependencies. That should be fixed.
return hinting: I think type hinting on methods was a great addition. I like strict, but I also like flexibility. I also think declaring the return type in the method declaration makes for very readable code (especially since developers don’t remember to document 100% of the time). I would never want to force it. It should always be optional. “mixed” should be a valid return type. And no, it is not just trying to be like Java!

I think everything else that PHP is missing could/should be provied with extensions. I may be missing some other language design stuff that I am not thinking of right now, but functionality should be in extensions.

This was in response to: "what do you think PHP is missing".