Monday, April 28, 2008

Science Schmience

I know for the most part that science is not exact. You need to gather facts and come to conclusions based on those facts and that it's sometimes just a series of educated guesses. And I have no problem with that, especially when these "conclusions" are fully disclosed and backed up. But sometimes, studies can be just complete non-sense.
Case in point, this article that links breakfast cereal to the sex of your baby. Come on.
The survey pool was 740 first-time mothers, of which 56% had a boy if they consumed a high calorie diet. From this, the "scientist" further drilled down to breakfast and then to breakfast cereal. Now really.....I think that's pretty close to 50-50 in a fairly small sampling.
I do realize that 6% in the larger scheme of things is fairly significant, but where'd they get breakfast cereal? And what about other factors? What about just plain old dumb luck?
And now that I'm completely into this rant.....I think scientist that release these studies should be held accountable for them. If their findings are disproved later, they should rack up one in the Loss column....or something.
I'll wrap this up by saying that my comments are not based in fact either. The study may be completely correct and I have no proof that says that this is not the case. I just find it rather disturbing at how some things are stated like they are truths rather than the educated guesses that they are.

Sunday, April 27, 2008

Baseball Website

I created a website for my son's baseball team. Used Google pages to create/host the website. I learned a couple of things while I did this.
1) Google is pretty cool. The Google page creator is really easy to use and it allows you to do a lot of things....much more than I was expecting.
2) I should have used CSS to make formatting changes easier.
3) Compiling baseball stats is not as easy as you would think.

Anyway, here is the site if you are interested in checking it out.

Wednesday, April 23, 2008

What's A Database Layer Good For?

I've spent the past couple of days refactoring my database schema. Nothing really major, just removing a categorization layer to make it less complex. It was determined that the extra layer would not be needed. I would have just left it but it wasn't my call. We also modified some table and field names to keep things consistent.
My manager seemed rather shocked that this change would entail changes in the database layer all the way to the UI. He seemed to think that having a database code layer would protect you from these changes and isolate the code that needs to be modified.
But in fact a database layer does not necessarily do this. If there is a new field, for example, that a user needs to supply data for, or in this case an entire table that does not need to be supported. The field needs to be added to the UI and then it needs to be processed and passed to the database layer. Therefore all aspects of the code that deal with this data need to be modified.
A database layer IS very useful and should be used when dealing with anything but very simple database table structures. It gives you the ability to isolate database functionality by hiding the database specific stuff from the rest of your code. These include things like connecting to the database, transaction handling, mapping object data to table fields, and other functions that deal directly with the database. Most of these changes can be easily modified without impacting other sections of your code.
Just make sure that everyone on your team, especially managers and project planners understand what you mean when you say you are implementing a Database Layer and that they know how certain changes to the database will impact the code.

Monday, April 14, 2008

Iolani Fair

This weekend is the Iolani Fair. Friday and Saturday from noon to 10:30pm. Come check it out. I'll be there working at the Produce booth under one of the big tents (the one without the games). I'll be there both days....all day long. Come check it out.

Thursday, April 10, 2008

Advanced Searching

I put together a couple of custom searches. One using Google Custom Search and another one with Swicki. They are both set to search websites that contain java programming pages. I seeded them both with a set of urls. I'm going to start using them to see if it will give me a better set of search results when I look for coding help. Hopefully I'll be patient enough to use them and compare the differences when I really need to find something.
The Swicki search is in my sidebar, the Google search didn't fit very well, so you can check it out here.
The way the searches work is that the Google custom search will search the entire web and put my url finds at the top of the search result list. The Swicki search is supposed to learn from other Swicki searches and refine itself over time. It also uses my seeded urls as a base and also uses a set of keywords. The set of keywords that I started the Swicki with was: coding, examples, java, programming, search, software.

Wednesday, April 9, 2008

ShoeBox v0.1

Here are the first set of prototype screens for the ShoeBox. It's basically functioning. The rest of the development will be fine tuning it to make it work better for me.

This is the Link screen. You can drag links from the browser to the list on the upper left. The space on the upper right displays the web page. The rest of the fields are description fields. Right now I'm storing the links on del.icio.us so I'm using the Topic Notes section as tags.



These are the link posts in del.icio.us. I'm hoping I'll be able to add these to my search component. The del.icio.us api is really limited but I think it'll work.




This is the search screen. There is a combo box at the top that will allow the user to switch between the various custom searches. For now I will be creating searches using the Google Custom Search tool.




This is the feeds page. For now I'll be displaying the Google Reader page since I'm planning to point all feeds here using Yahoo Pipes if necessary. I'm not really sure if this page is necessary. I was thinking of integrating this page with the Save Links page but I'm not sure yet.

Monday, April 7, 2008

Adobe Air

I tried out Adobe Air this past weekend. Spent a couple of hours prototyping my new project. By the way, I've named it ShoeBox for now.
Anyway, I have 3 basic screens and I created them in Air in no time at all. Of course I have Flex Builder so that helped with the layout, but still it was pretty quick. I also got some basic functionality done pretty quick as well.

Sunday, April 6, 2008

Search App

This is an overview of my next project. I basically want an application that will (1) help me store links and notes so that I can find them easily again when/if I need them, (2) allow me to do smarter searches or get search results that have more of the things that I am interested in, (3) show me more stuff (links) that i might want to see, like a suped-up google reader.

I've decided to create the application front end in Adobe Air. Just to see how it works. It seems like a good platform, although I'm not too keen on using proprietary stuff. I have also decided to use as many existing tools that I can. So I will be using things like del.icio.us, yahoo pipes, google reader, google custom search, and whatever else I can find.

My next steps will be to setup all the tools that I want to use, create the application ui, and come up with a name for this app.

Tuesday, April 1, 2008

Drafting is Done

Finally, all my baseball drafts are done. I can now get back to some real life stuff....like filing my tax returns.
I used a different method for my drafting this year. My picks turned out to be a little more off the wall then some of my past seasons. I'm placing a whole lot of hope on players like Nick Markakis, Josh Hamilton, and Matt Kemp. I got those players in multiple leagues. We'll see how the season goes.
But now that the draft prep is done, I can now start focusing on some side projects. First up is my Information storing/sharing/searching tool. More on that later.