wxBanker, your (hopefully) favorite lightweight personal finance application, has recently been accepted into Jaunty! Aren’t familiar with it? Check out the screenshots! It took a few months of getting over the debian packaging learning curve, and about as much work getting and responding to the reviews from MOTUs, but I did it. I plan on doing a quick point release this week and releasing wxBanker 0.4.0.3, which will sport updated translations in 15 languages (thanks translators!) as well as a minor bug fix or three. Once I get that out and into Jaunty, I’ll turn my complete focus (I hope) to the 0.5 series. If you’d like to translate wxBanker to your own language or improve a few of the lacking existing translations (Bosnian, Dutch, and Portuguese particularly), I’d love it! Head over to https://launchpad.net/~wxbanker-translators and join the fun!

The 0.5 series is a refactor and a bit of a painful one, as I didn’t have a great handle on how to efficiently and smoothly refactor. I could probably do a much better job managing it now, but that’s how experience works. All in all it is a much cleaner structure and led to more and better tests, which should allow more agile development and protect against future regressions. Some of the main upcoming features I’d like to get in are transaction tagging, recurring transactions, reporting, online syncing (via mint.com), and csv imports. I’d love for at least a few of those to make it into 0.5, and there is good progress on some of them including csv imports thanks to an impressive branch from Karel. On the other hand 0.5 has already been ongoing for about 3 months and I might cut a release into a PPA and get feedback while I add new features, sorting out any 0.5 issues and leading to a robust 0.6 release.

Overall I have learned a TON from this project, in no small part thanks to Launchpad. wxBanker started out as a terminal application which stored everything in a pickled linked tuple that I used for myself and added features as I needed them. Eventually I added a GUI (in wxPython) and registered the project on Launchpad, to get free hosted version control as well as more formal bug tracking (instead of a text file :). The combination of Launchpad and [wx]Python being cross-platform made it accessible to everyone, and took it from a project used only by myself to a project available in 15 languages with code contributions from multiple people.

So in conclusion, thanks everyone, enjoy wxBanker 0.4.0.3 in Jaunty, and look forward to future versions in my PPA. If you’re not on Jaunty yet, you can install 0.4 from my PPA, which I’ll be updating to 0.4.0.3 as soon as I release it. I’d love for any of your contributions, suggestions, questions, or criticisms to end up on Launchpad. I’d love to make it as usable and intuitive as possible, so anything that is unclear or confusing would be awesome to hear.

Spelunky is an awesomely addictive indie game, which I highly recommend checking out. It is sort of like Mario meets Nethack. As a side-scrolling, cave-adventuring character, you must venture deeper into the cave, defeating enemies, saving damsels, and collecting money and other fun items. A unique aspect that it takes from roguelike games is that there is really no concept of saving; each “play” typically lasts only a few minutes (much less at first), but you can get shortcuts to different zones to skip ahead once you get better. Also like Nethack, it is a surprisingly rich, detailed, and polished game for all its simplicity.



The unfortunate thing about Spelunky is that only runs in Windows. My initial attempts at running it in Wine proved unsuccessful, but after a while (and playing it on a friend’s machine which gave me enough to interest to keep trying), I figured out just the right combination of tricks to make it quite playable in Ubuntu. Here’s what you need to do:

• download Spelunky, make sure you are using compiz, and install wine and compizconfig-settings-manager.
• set wine to run in an emulated desktop window. To do this, run “winecfg”, go to the Graphics tab, and check “Emulate a virtual desktop”. The size isn’t really important; I use 800x600.
• extract Spelunky and create a file (if it doesn’t exist) called “settings.cfg” and put these contents in it:

1
1
0
0
1
15
15

This is the default settings, tweaked to the only setting that allows it to run in Wine. You can’t configure it normally since it only works in this way, so I had to have my friend change settings on his Windows box and watch what numbers changed. Thankfully for you I have done the hard work.

• Run Spelunky! Double-clicking it should probably work, but if you want it to play nicely with other applications using audio at the same time, force it to use pulseaudio by running “padsp wine spelunky_0_99_5.exe”. It should show a configuration window, but you can’t really interact with it. Just click in it and hit enter.
• Now you can play Spelunky, but it is at a 1x zoom level, the only one that plays nicely with Wine. This isn’t very usable! Run “ccsm” (System -> Preferences -> CompizConfig Settings Manager”) and enable “Enhanced Zoom Desktop” from the second group, “Accessibility”. Now you can zoom in by holding your Super key (Windows/Apple/Ubuntu key) and scrolling your mouse wheel in or out. Using this technique you can make Spelunky effectively full screen and if you sit it right in the corner, you will forget you are doing anything strange within a few minutes.

Okay, that’s it! Some initial setup is involved but once you’ve gotten it to work the first time, all you need to do in the future is run Spelunky and zoom in. As for help playing Spelunky, navigate over to the tutorial area once in the game, and you’ll learn everything you need to know! Let me know if it works for you and how much you love Spelunky (it’s worth it, I promise!) and don’t forget to check out more fun indie games on tigdb.

Jonah wrote a post about how awful the band Brokencyde is, and I mostly agree with the points mentioned. However I would also like to propose that if you aren’t particularly talented, it is more profitable to be epically bad than it is to be just moderately bad or even mediocre. I also think it is more entertaining and useful to society as a whole, thus being the utilitarian venture to embark upon if you aren’t traditionally talented. So let’s begin my defense.

I suspect the magnitude of profitability can perhaps be simplified as the multiplication of two factors: the number of people exposed to your product and the probability that a random person would purchase your product if exposed to it. In this model, a very ubiquitous band that people love is the biggest winner; lots of people are exposed to the product and a good percentage buy it. Mediocre bands do alright but not nearly as well. The percentage of people willing to purchase is on the same order, perhaps half as much or so, but the exposure is WAY less. The market is saturated with mediocre to decent bands and no one has time to find or listen to them all. As a result, these bands aren’t nearly as ubiquitous, leading to sales and thus profits which are orders of magnitude less. As you get worse and worse, your market is more and more saturated, you are less and less interesting, and profits continue to drop, in a roughly linear to inversely square fashion. But it doesn’t approach zero; once you cross a certain threshold of awfulness something magical happens:



You become interesting again! The market actually becomes LESS saturated as it becomes challenging to be worse than that threshold. You have become so awful that you are fascinating and captivating, entertaining and hilarious! Sure, the chance that a random person buys your product has now dropped by an order of magnitude or two, but your exposure increased by many more. You aren’t quite playing with the big dogs, but you can sell leaps and bounds more than the average guy. All for being notably worse than most other bands.

So clearly it can make sense selfishly and financially to be a Brokencyde or a Williang Hung, but are you harming society in the process? I don’t think so. Surely a few people will legitimately be offended and wish such bands didn’t exist, but I think the majority of us are at least entertained by their existence which makes us laugh or smile, have an interesting discussion with friends, or at least have a great gag gift (another unique sales niche these bands get in on). As a result, individual Brokencyde’s of the world increase the overall happiness of society more than an individual “average” band. Sure they’re bad, but would we (or they) want it any other way?

If your laptop has a fingerprint reader installed in it, there’s a decent chance you can set it up very easily in Ubuntu to login and [gk]sudo. Since the manpage isn’t particularly helpful, I’ll guide you through setting it up with the ThinkFinger library, which is compatible with most popular readers installed in Lenovo/Thinkpads, Dells, and Toshibas.

1. Install the necessary libraries: sudo apt-get install thinkfinger-tools libpam-thinkfinger
2. Integrate thinkfinger with PAM (Pluggable Authentication Modules): sudo /usr/lib/pam-thinkfinger/pam-thinkfinger-enable
3. Now acquire your fingerprint: run tf-tool –acquire. If you get an error here (not a failed swipe, you just need to swipe better), running it with sudo might be necessary. If you still get an error that thinkfinger can’t interact with your reader, it probably isn’t supported, sorry! Otherwise, keep swiping your finger until you get two successful swipes.
4. Finally, make sure it worked: run tf-tool –verify and swipe your finger. Try this a few times, and if it doesn’t have a good success rate, do another acquire (the previous step), perhaps slower and more intentionally.

Now you can log in by swiping your finger at the password prompt, and more usefully in my opinion, swipe your finger instead of entering the root password at terminal and graphical password prompts. This is one of those little things that, once you get used to it, is hard to ever live without. Check it out:


By the way, while there may be valid security concerns with fingerprint readers, don’t listen to the critics who say you can just breathe on it to get a swipe. 2D fingerprint scanners may work this way, but laptop fingerprint readers take a reading in both space and time. Try using tf-tool –verify and finding out for yourself; you can blow and breathe on your fingerprint reader all day without getting it to even recognize a scan, let alone a failed one.

Yes, I’m blogging from the toilet, but there’s no need for your mind to be in the gutter. I just moved in to a new apartment and haven’t set up internet yet, and as it turns out the bathroom is the only place that gets an unencrypted wireless signal.

Overall I accomplished quite a bit today. I worked a 10-hour day at work, came home and made myself a nice dinner, and did a complete move from start to finish, including all packing and cleaning at the first place, and unpacking into the other. This whole moving process took only 2.5 hours as well, so that’s got to be _some_ kind of world record. Granted, the two places are a few hundred feet apart, but I still feel accomplished.

In the Ubuntu world, I just noticed in Jaunty that when you switch backgrounds it does a rather sexy fade transition from the old background to the one. If you haven’t seen it yet I would definitely recommend checking it out in a VM with Jaunty Alpha 4. Now all we need is more than two wallpapers that ship with Ubuntu! Install gnome-backgrounds by default, anyone? If you don’t know about those, please install that package and check them out too, and subscribe to the linked bug report to get it more attention!

On a sadder note, recent Intrepid updates broke my fingerprint reader <--> PAM integration; did this happen to anyone else? There was one particular update asking me what I wanted to do about pam settings and I click the recommended suggestion, to use the new ones, but now pam doesn’t understand that I have a fingerprint reader to use to login/[gk]sudo, et cetera. Granted, it is only a one-liner to re-configure this, but it seems quite suboptimal!

Okay, good night/morning/afternoon everyone :)

As a developer on a team, when (if ever) is it reasonable to check in something which breaks one or more tests and causes the build to fail?
However there seem to be disadvantages to this type of workflow. First of all, if a build is broken, it can’t break. What am I talking about? If I check in my expected-to-fail test or my half-finished code and the build is failing, the next time someone unexpectedly breaks something, it isn’t nearly as obvious. A passing test suite changing to a failing suite, in a good environment, should be blindingly visible. But what about a failing test suite continuing to fail, albeit in more ways? That’s more subtle. From the second that happens you’re accumulating debt and the faster you find it, the easier it will be to fix. But if you can’t check in broken code, how do you easily collaborate with someone else on it over time?

Another problem that can arise is the accumulation of potential debt by replacing functional code with incomplete code. If you aren’t able to make the timeline or the feature gets dropped for release, you now have a reverse merge on your hands. This could be particularly time consuming if others have been working in the same files that your work touches. On the other hand if you had been implementing it side-by-side and waiting to actually replace it until it worked, it would be no problem to ship the code in that state at any point if need be. Your deployment is more flexible and less error prone.

So how can you balance these tensions in an agile environment? I’d love any feedback that anyone has to provide. What are the reasons for prioritizing a passing build that I missed, and what other drawbacks exist?

Python is a great language, but if you have ever tried to do anything web-related beyond a basic page fetch, it gets complicated quickly. What are single actions in your mind become multiple operations in Python.

Take for example POSTing some variables to a page. You are going to have to import both urllib and urllib2, and know what is in each of these. Use urllib.urlencode to encode your post variables, then pass them into urllib2.urlopen to get a connection object, then read that. Yikes! Oh, does the site require cookies? That’s another import and three lines of code; I hope you like reading up on CookiePolicy’s!

Attempting to accomplish this task with built-in modules will likely result in something similar to:

import urllib, urllib2
from cookielib import CookieJar, DefaultCookiePolicy

cj = CookieJar( DefaultCookiePolicy(rfc2965=True) )
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cj))
urllib2.install_opener(opener)
postVars = urllib.urlencode({“username”: x, “password”: y})
conn = urllib2.urlopen(“http://example.com/login.php”, postVars)
htmlResult = conn.read()

Compared to Java or C#, this is probably a terse solution. We are using Python however (for a reason), and that block of code sucks; that’s not how anyone thinks. It is hard to remember, leads to copy-and-paste code, and isn’t particularly readable. It also requires you to work with things you probably don’t care about such as cookie policies, openers, and url encoding. You just want to send a page a message!

After forgetting between projects and having to re-discover how to implement this functionality a few times over already, I finally decided to write something to remember it for me. Suddenly we can write:

import web

web.enablecookies()
htmlResult = web.post(“http://example.com/login.php”, {“username”: x, “password”: y})

The web module is quite short and not even remotely impressive (you could write what I’ve exposed here in 5 or 6 lines), but it takes something I found tedious and verbose and turns it into something simple. It adapts the broken-down functionality of these libraries to the more abstract level that I think on. Everyone thinks (and works) differently, and surely for some people it WOULD make sense (and be necessary) to open connections and read from them (if at all) byte by byte.

My interest in posting this has less to do with this specific example, and more to do with finding out what other “thought adapters” people have written to make something easier, more readable, or more pleasant. I have a few of these and pull them as I need them for various projects. What about you?

Avant Window Navigator has released version 0.3.2 today. This includes the release of the core dock, “awn”, and all the applets and plugins, “awn-extras”. There was a combination of about 130 bug fixes and feature requests closed in this release, including a few entirely new applets! One of my favorite new applets, moderately pointless I admit, is the Animal Farm applet which displays a cute animal who gives you a fortune on a click, thereupon changing to a different random animal. Below is a shot of 10 of them running :)


Other fun applets include a new customizable notification tray applet which supports transparency (with GTK 2.15+ in Jaunty), a flexible web comics applet, a new themeable clock, a simple to-do list, as well as plugins for Remember The Milk and Tomboy. Don’t forget that great applets like Pandora, weather, calendar, and shinyswitcher (a desktop switcher) already exist and have been improved as well.


Awn-manager has also gotten a lot of love since the last release; managing themes and launchers should provide a much better user experience. Tons of bugs have been squashed in awn-manager and most changes will be reflected immediately…no need to restart AWN!

For more detailed information please check out Mark Lee’s blog post, one of the main developers. To get it, check out the PPA, and don’t forget to Digg it! I’ll leave you with some more screenshots, all of which, including the ones above, are licensed under the WTFPL.

Awhile ago I promised to explain how to run applications in a different locale. This is quite useful as a developer, so that you can more robustly test your localization code. It can also be useful to translators who are translating to a locale which is not their default. Maybe you are learning a new language and want some extra practice in specific applications. Or, you may just want to see what your favorite application looks like in Russian or Hindi :)

There are basically only two steps to this rather simple process.

1. Install the desired languages. Go to System -> Administration -> Language Support. Scroll down the list, checking the “Support” box on the right for any language you want on your system. Once you are done click Apply and the necessary files will be downloaded for you. A logout and login is recommended by the application after this, and while not necessary, I recommend it as well (I’ve had a rare instance or so of segfaults when trying to use the locales before restarting the sesssion). If you want to follow along, ensure Russian is one of your choices.



2. Run the application under a different locale. You need to figure out the locale code that you want to run. In a terminal, run “locale -a” without the quotes, and you will see a list of all locales available on your system. If I want Russian, the one I am looking for in this list is “ru_RU.utf8”. It is usually fairly obvious which one you want. Now, again in a terminal, just add “LC_ALL=ru_RU.utf8” before the application you want to run. If we want a Russian calculator for example, we would execute “LC_ALL=ru_RU.utf8 gcalctool”. Ta-da!




This is a great way as a developer to make sure your applications are correctly detecting locales. I’d love to hear what you think and if there are any other reasons I missed that you may want to do this!

A lot of people criticized Windows Vista when it first came out, and I was one of them. Another large group of people also don’t believe there’s money in open-source. However for me, the existence (and negatives) of Vista and the awesomeness of Ubuntu landed me a sweet job. How?

Around the time that Vista came out, I was using a laptop with a 1.7GHz Pentiun M processor and 1GB of RAM. I got Vista (Business edition) for free through school so I threw it on. I quickly grew to love the new start menu and many of the improved usability features it had. Unfortunately it ran like CRAP. It was so slow and unusable due to my machine specs that it was unbearable. I couldn’t validate purchasing a new laptop at that time since Windows XP ran perfectly fine and really an OS SHOULD be able to run fine on those specs. But I also didn’t want to go back to XP and lose the features that I liked from Vista.

Around that same time I was also browsing Digg and noticed a release announcement for this thing called Ubuntu, Feisty Fawn to be precise. Everyone seemed to be raving about it and I thought since it was free, I might as well give it a try and see if IT can run decently on my machine and allow me to do everything I wanted. As it turns out, it ran quite well and either supported everything I wanted out of the box, or was flexible enough to allow me to do it myself! Even better, it shipped and supported the applications I used on Windows but previously had to install and keep up-to-date myself, like Firefox, Thunderbird, and Pidgin.

Over time I started getting more into contributing to Ubuntu, first by finding bug reports matching problems I had and adding more information, then more general bug triaging help via BugSquad. Eventually I joined the BugControl team and also started contributing to projects like Avant Window Navigator (AWN). At one point the bug bot that announces new bugs in #ubuntu-bugs-announce went down so I wrote a new one (EeeBotu) which lives on to this day happily (I presume) announcing bugs. When I heard that community members could be sponsored to the next UDS (Ubuntu Developers Summit) in California, I excitedly applied and even more excitedly was accepted to attend courtesy of Canonical.

Around THIS period of time I had been applying to various jobs, one such job at a fun startup in California, Genius.com. They had recruited at my college, the Rochester Institute of Technology in Rochester, NY, and had picked some candidates including byself to be flown out for second interviews. Then the downturn in the economy came however, and they decided not to fly anyone out and re-evaluate new hires at a later time.

This was understandable, but I thought that maybe since I was going to be out in California for UDS anyway, they might want to take me up on a free interview. As it turns out they did. I was offered a job, accepted it, and after about a month I can say that it is a pretty sweet job!

So because Vista sucked (at least initially), I gave Ubuntu a try which I heard about on Digg, got involved and was sponsored to attend UDS in California, where I was able to interview at the company I currently work at.

So just remember, there’s a positive side to every negative (thanks Microsoft!), and there IS money in open-source, at least indirectly (a huge thanks to Canonical!). Has Digg found a new business model?

In other news, I’ve joined a pact with 9 other friends to write a blog post a day for a month starting today, so if all goes well you will be hearing many (hopefully) interesting or fun things from me!