Ben's Bits

Christoph Schiller has written a free physics textbook and made it availible online for non-commercial use. I took a look at it and have a few ideas. First, he should make a printed copy availible via the fantastic Lulu publishing company. Secondly, I think he should make it officially availible under the Creative Commons Non-Commercial Share Alike licence. It's essentially that anyway and it would provide a better understanding and probably better legal grounds to the same end.

Finally, this book is overly technical and not user friendly for the introductory college physics student or advanced high school student in the United States. I would love to see a more traditional US physics text be made availible under a free license like this. I found it difficult to locate common US physics units in this book (eg. 2D kinematics, 3D kinematics, springs, waves, light (wave, not particle) and lenses, magenets and electricity). It would be great if someone could put together a book that looked this professional but which was shorter and designed for a more introductory college / advanced high school course. If someone wanted to work collaboratively on a project such as this I would be thrilled to be a part of it.

One example of a free license textbook is "All About Circuits" which was originally a PDF document but has been updated and become an online reference. I do wish they would come out with a PDF formatted book again though, it looks much more professional when printed. This electronics textbook was licensed under the Design Science License which seems to be even less restrictive than the Creative Commons NCSA license.

I'm sure everyone has seen it by now but if you haven't Michael Ossman's site has instructions for building a series of adapters which convert a standard 10 base-T ethernet cable into a crossover, modem, null modem and Cisco console cable. You could also make additional adapters to create a UPS cable or just about any other serial data cable.

For years I've been thinking about a peer to peer wireless mesh technology for automobiles. In the past two weeks I've seen two tradepaper articles about wireless technologies for cars and I have yet to see this hit the mainstream "geek press" eg. Slashdot which has been somewhat surprising.

On November 14 Network World published an article entitled "U.S. pitches wireless highway safety plan" which discussed the US DoT plan called the VII project. The Vehicle Infrastructure Integration (VII) prject aims to reduce highway fatilities and improve congestion problems by transmitting warnings and road condition data to drivers and automobile computer systems via a 5.9 GHz short range (984 foot) wireless connection as you pass "Roadside Units" (RSUs). Data is gathered from your onboard computer and combined with GPS data, the data collection and transmittal is to be anonymous. You can read more about the proposed system on the concept of operations page.

One week later Network World published a second article entitled "GM to roll out intelligent car alternative" discussing the GM V2V plan based on the existing GM OnStar technology in combination with 802.11a/802.11p networking technology. GM already has a demonstration fleet with this technology. One advantage of this technology over the DoT VII project is that these devices are designed to talk directly to one another in addition to talking to roadside units (RSUs) so you can get more data in a more timely manner. One example of this advantage is that the V2V device will let you know if someone is in your blind spot by blinking a small LED on your mirror. This would be unsupported by the VII program which only periodically exchanges data. Some data would also be availible even when you are not on a road equipped with RSUs which is a stong benefit during initial deployment.

Overall I'm impressed with the decision by GM to do direct vehicle to vehicle communications along with communications to RSU. I think there is the potential to get much more valuable data much more quickly using a direct vehicle to vehicle mesh network than solely with RSUs. The RSUs are still an important piece of this though because they can track aggregate data for the roadway and provide statistics about the road itself (eg. temperature, etc) to the vehicles. I applaud the DoT for recognizing that getting widespread adoption will be much easier by committing to keeping data anonymous, this is one of the reasons I'm much happier about these proposals than Mark Gibb's RFID tracking proposal. The one thing I would hope for is that GM would encourage and assist other automakers in creating a standard for vehicle to vehicle communications to encure interoperability of these systems between brands.

A proposal made earlier this month in Westchester County, NY would require all commercial wireless internet access points to have a firewall to "secure and prevent unauthorized access to all private information that such entity may store" and post a sign stating: "You are accessing a network which has been secured with firewall protection. Since such protection does not guarantee the security of your personal information, use discretion." Responding to criticism, Westchester County Executive Andy Spano has written a rebuttal encouraging readers to read the legislation and attempting to clear up some misconceptions about the law. I don't know how far the FCC would let this go, they usually don't take kindly to local government interference. The legislation seems to be aimed at getting businesses to secure their networks but is legislation really an effective tool for getting this done correctly?

I've been thinking it's time to update my VPN solution so I've been poking around looking at current solutions that will work for me. I was excited by the OpenVPN SSL VPN server but I need to support PocketPC clients which they do not and it would be nice if I could avoid installing a client on Windows 2000/XP/2003 machines. Another SSL VPN solution is SSL Explorer.

Jacco has a great page demystifying the setup of Openswan IPSEC VPN servers for use with a variety of clients including PocketPC. Unless something changes or it becomes especially difficult to implement that's probably the route I'll go. Nate Carlson has another page about configuring openswan to work with IPSec and L2TP.