othello is cookin
I've been taking a bit of a break from working on this site because I've been working on a basic python version of Othello/Reversi. I think I've been doing well, and sharpening up my skills to boot. I plan on making it open source/available pretty soon through svn even though it's a ways from being full featured. I would really like to find a way to get a bunch of people to look at it and give me some feedback. I can read all day long about the right things to do with code, but it's harder to pick out the things that I'm doing wrong - especially with something like games programming. However, figuring out the svn settings to make only what I want public is not the sort of thing I'm interested in doing right this moment, but soon!

It has been quite a while since I worked on some code that was not somehow website related - and it's nice. A lot less time is spent tweaking random little things and much more is spent plugging away at logic. Granted, I haven't gotten very far into the graphics part, but so far, mostly logic. Aside from some ASCII representations of the board, I plan on doing something with curses, and eventually if I feel ambitious, maybe pygame. Othello is appealing because I've always liked the game, and have tried to implement it before in python, and c++ or java - I forget which. I guess it really depends on whether I tried in highschool or college. Either way, the fact that I've never finished the project makes it that much more appealing to me, and I think the possibility of developing different grahical frontends and possibly game playing AI offers good opportunity to work on my chops.

So anyway - Hopefully svn access soon so I can get some feedback.

Foodsperiments and complaints
Last night Hillary and I were facing the eternal question every couple faces in the evening: "What are we going to eat tonight?"

After a bit of debate, we decided on some frozen tortellini that we had lying around - it had served us well when we made the first half of the package. However, tortellini alone does not a dinner make. At least not a very complete one. The go to for pasta-related side dishes for us is typically garlic bread. Hillary makes a mean garlic bread with some mozzarella and adobo as well as other things that don't immediately come to mind.
Lacking proper bread and mozzarella (and with no one wanting to go to the store) Hillary set her sights on the only bread product available - some tortillas left from making quesadillas a previous evening. With the concept of garlicy tortilla goodness in mind, my lady went to google and found a few recipes that fit the bill. One involving butter and garlic cloves, another involving oil and garlic salt.
I guess she found both lacking, so what follows is hybrid she created, single handedly saving dinner:

Garlic Tortilla Bread

• 3 tortillas, cut in quarters
• about 1/4 stick butter
• one clove garlic, minced
• salt and pepper
• olive oil
• parmesan cheese

Instructions: Preheat oven to 400 degrees. Melt butter in a bowl. Add all ingredients that are not tortilla or cheese into melted butter (just a reasonable amount of the olive oil). Spread this mixture on top of tortilla quarters, placed on some sort of baking sheet. Bake somewhere around 5 minutes or until the pieces look like they are crisping up a bit (our time was closer to 3 minutes I think). Then flip tortillas and put parmesan cheese on the upward side - bake another 3-5 minutes until looking crispy.

The resulting pieces are quite good - combining a bit of chip crispiness with a bit of bready give and a garlic bread like flavor. Hillary thought next time we should substitute garlic salt for the clove, since in the flipping we lost a lot of garlic. What a lady - and she says she has no inclination toward cooking!

Last night we also discussed the idea for a machine called the complain-o-tron, which would listen to complaints and reply "Tough noogies." While not practical, it has a certain elegance to it. It gave me the idea for a website with a single form for inputting your complaint - and then one (or maybe more!) responses. The interesting part would actually be storing the complaints (anonymously, naturally) and then providing some fun statistics about them, like "80% of complaints contained the word 'Buns'". It would be a nice oppurtunity to work on some text parsing scripts.
The whole idea might be a little too close to a certain achewood comic. We'll see.

from symfony to django
Work continues on the site, trying to get the CSS just how I want it, also tweaking the code base so that things are properly seperated, i.e. low cohesion between different elements of the code. I've been working through various parts of the django tutorials and the book (which is on its way from amazon) and find myself getting ahead of the tutorials quite often. This has a lot to do with the fact that I'm used to working with the Symfony framework for php. Working with a framework, things like separating logic and presentation, seperating urls from actions (or views, depending on what framework), etc are constantly hammered into your head. The django tutorial starts you out with high cohesion and gradually works apart the various pieces. So instead of working straight through the tutorial, I would find myself hopping right into the documentation to look for the django implementation of a feature I use all the time in symfony. Not the best way to work - however, today on reddit, I stumbled accross Top 10 tips to a new django developer. Great set of tips, not so much for the technical aspects, but it quickly pointed out a number of things I had been looking for.

Gotta go, but more on this later as well as database issues!

thanks django syndication
I've added an rss feed to the site, since every good blog really needs a way for people to keep track of it... I would forget more than half of the blogs that I regularly read if not for the google reader item I see every morning on my homepage. Of course, this isn't an rss feed rolled from scratch. The minutiate of writing and testing your very own xml creation is not very much fun. Xml is a necessary evil in many cases, but I try to stay on the parsing side of things whenever possible. I have many thanks to give to the Django Syndication Framework.

The question this brings to mind is - If I'm trying to learn python through django, how much of the code should I be writing from scratch and how much of it should be contributed add ons? Tough question. From what I gather, a lot of python coding is "from x import y". I watched a talk from PyCon where Steve, who wrote reddit (twice) talked about how every time they need something new, they use the standard python implementation until such a time as they either need or have time to roll their own version of a feature. For me, up to this point, included stuff seems to be a good way to get past the stuff with annoying implementations: rss and comments so far. I'm thinking I can revisit the stuff I want to learn more in depth once I get past the "need to get done" stuff for the site. So far, even implementing the included stuff has taught me a ton.

Next post: django database migrations and the advantages of knowing another framework

the pain of web design
I've swapped the colors around bit... the original green/yellow scheme that I decided on really did not sit well after a month or so on my hard drive, and really became an issue when I actually made this site live. It's fine for something to bother me when it is just on private display, not so much when on public.

Putting together the nuts and bolts of a website is a task I like. Working on the paint job, however, is a painful, irritating process.

Naturally, I sought an easier way through the old standby, google. I was not disappointed. Color Wizard is a tool that lets you input a base color and provides you with a whole set of colors that complement it based on real color theory. Utterly perfect for me, because minute shade tweaking is not a lot of fun, and because I seem to have a great deal of trouble perceiving what colors actually go together and which colors look like an unspeakable horror.

I'd like to say that was the end of it. I inputted my base color, and machine, logic, and science triumph over the evils of aesthetic design and voila! New color scheme. (I love it when that happens!)

Sadly, not the case. I switched around the colors, but still had to go through about 20 iterations of turning the screen towards my girlfriend and saying "How about this?" before we reached a consensus on what didn't look awful. Thanks for the kickstart "machine, logic and science", but the point still has to go to aesthetic. In the future, this task may just need to be completely outsourced to the girlfriend.