Antares Trader Blog

Ars Technica's Ken Fisher recently penned an article about Why Ad Blocking is devastating to the sites you love. As an avid Ars Technica reader (via their RSS feed) I take an interest in the health of the site, and as a user of Ad Block Plus I am just just the kind of user to which Fisher is speaking. Here is why I still block ads.

With the FCC considering the rules for the next decade of internet access, I wanted to put in my two-cents about SureWest Communications. They have just hooked me up with a Fiber Optice 3Mbps connection for about the same as AT&T wanted to charge for 768Kbps and for less then a third of Comcast's asking price for the same speed. If you have not heard of SureWest, it is because they are only in two places in the country, and I just happen to be lucky enough to live in one of them. Because of this I don't have to deal with stealth filtering (Comcast) degraded ssl connections (Comcast and AT&T) and less then half of advertised speed (AT&T). The real point here is that much faster higher quality internet can be delivered at the current price point. When AT&T wants the FCC to call 350Kbps highspeed and Comcast says that they need filtering to maintain quality, the FCC needs to point to SureWest and ask why they can do it.

Here's to more competition in the ISP market place!

Here is another one of those head scratching moments that makes a lot of sense once you know what is going on. If you have installed the Rails 3 pre release, you may have inadvertently broken a number of gems that depend on ActiveSupport.

Your humble and not very consistent author has survived the Christmas Rush and the purchase of our first home with very few social causalities. One of these few has been blogging. SureWest is going to come and hook up my fiber connection on Friday, until then I am at the mercy of places that advertise "Free Wifi" but don't actually have a live connection. (I'm looking at you LaBou Bakery of S.Natomas)

This doesn't mean I haven't been hacking. I have started working with Mongo and MongoMapper both of which are really cool and deserve a post written with out the distraction of a fast food joint. I'm also looking at Redis. These alternatives to SQL databases make the complex data relations found in Antares Trader much easier.

I'm also enjoying learning from the folks at the Sacramento Ruby Meet-up. More will flow from this blog.

Finally, I'm starting on Rails 3 sence I know that it will be the version I will want to which I will port Antares Trader. I thought Merb was great, but it will also be nice to be back in the main stream again.

More will flow from this blog. Hopefully starting Friday

Github has recently posted a wonderful article about their history with job queues. It is a superb read and I recomend looking both at it and the README file for Resque, GitHub's new Redis back job queue. It is replacing the venerable delayed_job. In this post I am going to briefly look at some of the concepts of Worker Queues in an attempt to better explain my hopes for Updater, my delayed_job clone for DataMapper.

The factory_girl Gem is a great tool for creating valid test instances on the fly. There are lots of posts about why this is so much better then fixtures. Unfortunately for us DataMapper fans, the official version is not quite compatible with our favorite ORM. I've done my best to remedy that situation. There is now a fork of factory_girl that I have made mostly compatible with both DataMapper and ActiveRecord. You can also pull the gem from GemCutter as dm-factory_girl.

The game loop is an essential part of what makes a game work. To implement one for Antares Trader, I wrote a class that would repeatedly call the update method on DataMapper resources. It was a quick hack that got my mini-game sketches working. Eventually, I needed to pull the code out from under the web server and give it it's own process. For a while it was just a class in the lib directory. Eventually, updater grew to need it's own table in the database and now there was Updater code in too many different places.

At this point I went looking for a better solution. I picked up delayed_job and liked what I saw. It did a lot of what I needed. It had a cool API, and GitHub used it so I knew that it had been tested in the real world. On the other hand there were a couple of features missing, and more importantly, it depended on ActiveRecord. So it was back to Updater.

The Factory Girl gem from thoughtbot is the way to go if you have to test ORM classes. It has been featured in Ryan Bates Railscast, and by good fortune, it also works with DataMapper, my ORM of choice. When I upgraded to the latest release of DataMapper however, all my user tests broke. in particular the crypted_password and salt fields were not being set when the record was saved. I'm using the salted password mixin from Merb and I assumed that it was to blame having not gotten up to speed with some of the new API changes in DataMapper. It turned out instead to be an incompatibility between Factory Girl and Datamapper that had existed for a long time but was now showing up. I've modified factory girl to maintain compatibility and my changes are at http://github.com/startrader/factory_girl.

Internet Explorer is the bane of every web developer's existence. Google being the one of the largest web content producers finally got fed up and has developed a work around for everything that is wrong with IE. They have developed a browser plug-in that completely guts IE and replaces its rendering engine with Google Chrome. The Google Chrome Frame looks like a great way to reach all the cooperate users and grandmas that cannot or will not upgrade to a new browser.

In my last game related post I wrote about the mechanics of making a game work. Just because a game works does not mean it is enjoyable to play. In this post I am going to talk about the reward cycle and why some games are so addictive and other either get boring, or were never fun to begin with.

The key concept her is the action-reward cycle. A player makes a decision and then gets some feedback on that decision. Good decisions get rewards while poor ones are either ignored or reduce future rewards. The thing that makes a game fun to play is the balance and value of these rewards.