Google Reader “Next” Bookmarklet Fail

One of my peeves about using various Google products is that sometimes bugs can sit around unfixed for ages. Like this one: Google Reader provides a bookmarklet that allows you to easily skip directly to the URL of the next unread item in your reading list, bypassing the Reader interface completely. It is nice if you like to read blog entries and stories with full formatting, rather than just the feed content. It is also nice when you subscribe to feeds that only post a summary, and require click-through to get the full content of an article.

However, when using it, sometimes you get this:

Sometimes using Google Reader's "Next" Bookmarklet Causes a "Forbidden" error.

This problem has been reported in Reader-related forums for months. It has something to do with viewing other people’s shared items, and you can work around it by going into Reader and reading the next item there. Still it’s very annoying.

Now, I develop software too, and I recognize that every team has a long backlog of work, and some things fall to the bottom, but it seems like a long-standing, customer- user-facing bug like this would eventually offend some developer enough that he or she would just go and fix it out of pride.

Turtle Graphics In Python? Cool…

I just learned that Python has Turtle graphics built in.  Very cool.  How did I not know about this before?  Now I guess I know what Maggie’s first programming language will be. Inspired by this nice post showing how to draw fractal trees, here’s a quick Sierpinsky triangle algorithm in Python:

import turtle

def striangle(depth,base):
   turtle.down()
   if depth == 0:
      turtle.fill(1)
      for i in 0,1,2:
         turtle.forward(base)
         turtle.left(120)
      turtle.fill(0)
   else:
      for i in 0,1,2:
         striangle(depth-1,base)
         turtle.up()
         turtle.forward(base*2**depth)
         turtle.left(120)
         turtle.down()

turtle.reset()
striangle(6,5)

Depth 5 Sierpinski Triangle

Fun!

Cool Seattle Thing No. 3429: Dev Workshop for Transit Data

Too bad I found out about this after the fact…  Apparently King County Metro recently hosted a Developer Workshop for transit data.  Apparently they’re looking for more people to write innovative transit-related applications like the super cool One Bus Away, which I use every morning and evening on my Android phone.

Having spent years riding Port Authority buses and light rail in Pittsburgh, I still find it astonishing that an event like this can happen at all.  I mean, transit authority officials (including the KC Metro general manager!) hosting a workshop where they encourage third parties to take their data and play with it?

O brave new world that has such county officials in it!

Buy Obama!

I noticed today that Electoral-Vote.com‘s composite of national polling is giving Obama a 338 to 185 lead in electoral votes.  The interesting thing about this is that Intrade‘s state-by-state prediction markets have been giving him roughly the same margin for several days, maybe a week — i.e. the prediction markets have been leading the polls.  Right now I think the only difference between EV’s electoral map and Intrade’s is North Carolina: EV calls NC a tie while Intrade gives the state to McCain.  It will be interesting to see if the polling in NC turns back toward McCain in the next week or so.

Looking at Intrade’s market for the overall outcome of the race, it’s worth noting the technical indicators (as I did in the primary).  Obama’s MACD is strongly positive, indicating upward momentum.

Intrade chart for Obama prediction market

The MACD also shows Obama’s momentum went negative just after the Republican convention, and then reversed again around September 20, just as the shit was hitting the fan on Wall St.

One thing I wonder about the state-by-state prediction markets is how much liquidity they can have?  I mean, how many people are trading in these markets?  Are is there really enough volume in the West Virginia, Indiana, or Missouri markets to get accurate predictions?  Those states polls are all within the statistical margin of error (I think).  Florida and Ohio are also pretty close, but I guess those are more liquid, because they’re known battleground states.

Anyway, it will be fun to track this as the election gets closer.

VideoLectures.net: Good machine learning and AI lectures

A while back I posted about SciVee, a site for posting videos of science presentations. Today my old neural-networks labmate Tal Tversky commented pointing me at VideoLectures, a similar site containing academic lectures.  Although the site doesn’t seem to be explicitly intended for any one topic, the page of “top” lectures is dominated by talks in statistical machine learning.  Skimming the page, I noticed talks by such notable names as Tom Mitchell (chair of CMU’s machine learning department), Usama Fayyad (former VP of Research and “chief data officer” at Yahoo), Michael Jordan (UC Berkeley), and William Cohen (CMU).  Lots more, too.

In addition to ML and AI stuff, there are also talks by Tim Berners-Lee and Umberto Eco on the “Top Lectures” page.

PLASTK version 0.1

In a departure from my recent blog themes, I’d like to get back to AI and machine learning today to announce the first development release of PLASTK: The Python Learning Agent Software Toolkit.

From the PLASTK README.txt file:

PLASTK is a Python class library for building and experimenting with learning agents, i.e. software programs that interact in a closed loop with an environment (agents) and that learn. Especially, but not limited to, reinforcement learning agents.
[...]
PLASTK has an extensible component architecture for defining the components that make up a software agent system, including agents, and environments, and various components needed to build agents, such as function approximators and vector quantizers.

PLASTK started life as an ad-hoc collection of algorithms from the reinforcement learning and self-organization literature that I implemented for my dissertation experiments. While in grad school I managed to clean it up to the point where a couple of other students were able to use it. The current release is the first step in my effort to make it usable outside a close circle of labmates.

PLASTK currently contains implementations of Q-learning and Sarsa agents tabular state and linear feature representations, self-organizing (Kohonen) maps, growing neural gas, linear, affine, and locally weighted regression. It also contains some demo environments including a two dimensional “gridworld” (shown in the figure), and a pendulum. Included examples show how to set up agents to learn in the gridworld and pendulum environments. PLASTK also includes a simple, component-based GUI, shown in the screenshot on the right, for visualizing agent/environment interaction.

PLASTK’s biggest deficit right now is that the agent/environment interaction API is the long-deprecated RLI version 5, from Rich Suttons RLAI group. One of the first to-do-list items is to update the interface to the new RL-Glue standard.

The PLASTK code is redistributable under GPL v.2, and available for download on SourceForge.net. Please feel free to contribute! Patches are welcome.

Somethin’ Happenin: AlphaLab!

If you were looking for something to get excited about in the Pittsburgh start-up economy, here’s someting for you. InnovationWorks has just launched their new software/internet/games start-up incubator, AlphaLab.

Have an idea for a software or internet product that you think you could launch in six months with a little investment and hard work? Submit an application. According to their FAQ, the six funded projects will get:

• $25,000 investment to develop initial versions of your product.
• On-site staff resources guiding you on product strategy and development including market definition, market analysis, and user experience testing.
• Access to advisors, mentors, and industry experts.
• Free office space to develop your product in a collaborative environment with fellow entrepreneurs and technologists.
• Access to the AlphaLab user community to provide valuable feedback regarding your product designs and usability.
• Access to AlphaLab educational programs for practical advice on building your team, securing investment, market entry strategies, and other critcal topics.
• Participation in AlphaLab networking events to introduce you to other technology entrepreneurs in Southwestern Pennsylvania.
• A vibrant environment to launch your company and entrepreneurial career.

AlphaLab seems to be modeled on Paul Graham’s Y-Combinator, though AlphaLab seems to give slightly more start-up money, and provides office space (on the South Side) for the companies to work in. And unlike some other tech startup investors, AlphaLab won’t take control of your intellectual property.

It’s a super idea, and I can’t wait to see the companies that come out of it.

Somethin’ Happenin’ Here…

Between my last few posts here and recent comments elsewhere, I’ve probably given the impression that I’m bearish on Pittsburgh, at least as far as the tech/start-up economy goes. Not so. While it’s sometimes hard for me to keep my mood up in the winter, this has actually been a bullish week for me, Burgh-wise.

Among the things that lifted my mood this week was lunch today with Matt Harbaugh from InnovationWorks. Before today I had lumped IW in with the slew of public/non-profit/consortium/partnership groups around here that purport to help the tech economy. And frankly, it was never very clear to me what any of them did, or whether anyone in the local tech industry would care if they just disappeared.

After talking with Matt, however, a couple of things became much more clear: (1) there is a burgeoning start-up scene here, and (2) a lot of it is in the IW portfolio. (Though by no means all of it!) I don’t think the scene reached the perpetual-motion-machine stage yet, but that just means you gotta keep pedaling. Just knowing that there are people out there taking real, concrete action to move things the right way is a great thing. Talk is cheap, especially in the blogosphere. Action, baby. That’s where it’s at. I expect exciting things from IW.

CCNC/SfN/Epirob/AAAI Conference Blogging

I am currently in San Diego, in the second day of a three-city, four-conference tour during which I’ll be attending the CCN Conference, Society for Neuroscience, Epigenetic Robotics, and the AAAI Fall Symposium on Computational Approaches to Representation Change during Learning and Development.

This is the longest and most complicated conference travel adventure I’ve ever undertaken. Especially since I’m presenting either a poster or a talk at all four conferences. In addition, it is my first time attending SfN, which draws 20-30,000 attendees, making it easily an order of magnitude larger than any conference I’ve ever attended. I have to wonder what the point of such a large conference is, since I’ve been told by several people not to expect to run into anyone I know if we haven’t planned the meeting in advance.

So far, CCNC has been nice — only 200-300 people — with lots of work focused on computational modeling of reinforcement learning and other decision making processes. I’ve run into a couple of people that I knew from elsewhere, as well as a couple of people from CNBC that I’ve met for just the first or second time. I’ve also been slightly disappointed that a couple of people I expected to see are nowhere to be found.

I’m going to try to blog at least once from each location. Newt Gingrich is speaking about research funding priorities for science at SfN on Monday. It’s during the first hour of my poster session, but I may try to attend anyway.

DARPA Urban Challenge has Started

The national qualifying event (NQE) forDARPA’ s Urban Challenge started today in Victorville, CA. The Urban Challenge is the current incarnation of the DARPA Grand Challenge autonomous car competition. Unlike the previous challenge, which was an off-road race, the current challenge takes place on city streets with traffic. The Austin Robot Technology team (ART)  made it to the NQE and is now trying to qualify for the finals, on Nov. 3. ART’s AI is being directed by UT Austin CS professor (and member of my thesis committee) Peter Stone, and AI development is being headed by my good friend and former labmate Pat Beeson. GO ART!

DARPA will be webcasting the finals, but unfortunately, I can’t find any sign of a scoreboard for the NQE which will be going on until Wednesday. Georgia Tech’s Sting Racing team is running a blog, though.