I often want to see the hands I won or lost in the last 15 minutes or so. The online poker sites supported by Poker Copilot have excellent hand replayers, but they are too interactive for what I had in mind. I had the idea of an ever updating list showing at a glance how I fared.
On Sunday afternoon I coded it, in time for the beta release of v 1.2. Here’s a screenshot:
(The screenshot suggests that I won $2,085 in my most recent hand. It was only tournament chips, although I did win the tournament and $4.50.)
It’s quickly become one of my favourite parts of Poker Copilot. Judging by the feedback I’m getting there’s others who find it helpful too. It’s also a great tool for finding bugs in the hand history parser.
It was much easier to code than I expected. I discovered that there are unicode characters – supported by Java – for the four suits, saving me from having to find or create suitable images. Java text labels (Swing’s JLabel) support HTML, so I use HTML to colour the cards. A monospace font gives me the alignment. The description of the hands – “Nines and Treys”, for example – are generated by Moritz Hammer’s outstanding open source poker hand evaluator. Moritz’s software is excellent in more ways than I can describe.
Everything else I needed – the iTunes-like table rendering, the statistics gathering – was already in the Poker Copilot codebase. I simply had to stitch the various pieces together.
The problem: I’m maintaining a list of requested features for Poker Copilot. These mostly come in through e-mail or discussion forums. It’s getting unwieldy and I’m running multiple e-mail conversations and forum threads discussing these features. People are requesting things others have already requested. The same bugs are reported multiple times. I’d like the list of requested features and bug reports to be public. I’d like this to be as painless as possible, for myself, and especially for my customers.
The solution (maybe?): Get Satisfaction. I’ve been considering this site for customer support. Before I take the plunge, would anybody like to offer strong opinions for or against using this service?
It seems I’ve found the right way to attract beta testers. Since announcing the open beta, I’ve been inundated with feedback and bug reports. The beta test has revealed a couple of regression bugs, a bug handling PokerStars hand histories, problems with handling accented characters used in non-English names, and some usability problems.
A big thank-you to everyone who has already taken active part in the beta test. I have to give a special thank-you to Pedro, who not only found an important bug, but also isolated it, determined the problem, and gave me specific steps to fix it. What more could a one-person software company ask of users?
After dithering around I’ve decided to make the PokerCopilot 1.2 beta an open process. Anyone can download and install the beta here. This is the version that adds, amongst other things, support for PokerStars.
The offer of free licenses for beta testers still stands. The deal is: 1) try out PokerCopilot 1.2 with PokerStars. 2) email [feedback at pokercopilot.com] with some useful feedback, especially bug reports. 3) you’ll get your free license at the close of the beta process.
Naturally, my judgement on what constitutes “useful feedback” is final, but you’ll find that I’m quite reasonable on this count.
While adding PokerStars support I tried to ensure that Poker Copilot remains the easiest hand history analysis tool to configure and use. What I’m especially avoiding is the need for users to manually locate and “import” hand history files.
In light of this, I added a new start-up dialog for first-time users. It checks for the existence of both Full Tilt Poker and PokerStars, then reads their own preferences to locate the hand history folders. To ensure that the user knows what is happening, Poker Copilot presents the list of folders that will be imported.
Additionally it informs the user how to add folders later, perhaps containing a friend’s hand history files. The complete Poker Copilot instruction manual in a sentence, if you will.
The implementation was easy, but designing a workable, elegant, and simple solution was a challenge.