After a little bit of searching, I discovered a site that will allow you to calculate the distance of any saved path in Google Earth. If you highlight the path name in the sidebar and copy it, you can paste it into the box on the site as xml.

The picture above is the run I did last Saturday, before exams destroyed my life. School is on the left and the Cobb’s Hill Reservoir, my destination, is on the far right. The path I made put the distance at 9.3 miles. I’ve made the kmz file available for download.

I suppose I could just measure the path once, but I really would like to be able to keep these paths, for two reasons. It would be nice to be able to change them later on, in case I make a slight variation on a run. But the real reason is that I just love data. I think it would be really neat to make an entire collection of the runs that my team does, and then put them on the internet for anyone to download – new freshmen, other area runners, etc.

I also want to use Google Earth to find some new places to run this summer. I’ve been canvassing the same trails for five or six years, and they are getting worn out. A lot of my runs need to be extended to fit college training distance as well.

When I think about this more, it makes sense. The Macintosh is not the platform of choice for hardware designers, who are probably using UNIX or Linux. The VHDL simulation software that we use in school only runs on Windows, but there are some command-line simulation tools such as GHDL that have been ported to run on Mac OS X.

I’m considering writing my own bundle, which would be a good excuse to buy that new TextMate book that I’ve been wanting. Syntax highlighting and commonly used snippets/declarations would be the main priority, followed in the future by GHDL integration. Of course, I would add the bundle to the Macromates SVN repository once it was in a useful state.

If you got here from Google, what functionality would you like to see in a TextMate VHDL bundle?

Enable Fullscreen: This tip comes from Macworld. The easiest way to get fullscreen in QuickTime without buying the Pro version is with a simple AppleScript. Open up Script Editor, located at /Applications/AppleScript/Script Editor.app, and copy and paste these next few lines into a new script.

tell application "QuickTime Player"
present front movie scale screen
end tell

Save the script as an application (File Format: application), and put it in your applications folder. Now, whenever you are watching a movie in QuickTime, simply run this application to have the movie appear fullscreen.

Fullscreen Preferences: First, find your Quicktime Player preferences file. It is located in the Library/Preferences directory of your home folder, and it is called com.apple.quicktimeplayer.plist. The .plist extension may or may not show up in the Finder. Then, open it using Apple’s Property List Editor. To do this, you must have the developer tools installed. If you do not have the developer tools installed, you can edit this file with a simple text editor, such as TextEdit, but be careful of what you change!

Here we notice, after expanding the root node, that there are a number of preferences with the prefix “Fullscreen.” These preferences are not available through the GUI to users without Quicktime Pro. The preference we are most interested in is FullscreenPreferenceScreenNumber. This may have a very large value, but you should be able to change it to 1 or 2 to enable either your main or auxiliary monitor. Those on a text editor, change the value between the <integer> and </integer> tags under the line <key>FullscreenPreferenceScreenNumber</key>. Save the file, and reopen QuickTime.

This step is important: Drag the movie window to the monitor on which you want it to play in fullscreen. Then run the fullscreen script. If the monitor on which the movie is played in fullscreen is not the right one, try running the script a few more times. If this doesn’t work, you can try experimenting with the value of the preference. Each time you change it, you must reopen QuickTime. If anything does go horribly wrong, just delete the preference file and QuickTime will make a new one for you the next time it is opened.

There are other preferences that can be experimented with:

• FullscreenPreferenceEnableHUD – determines whether the movie controls show up in Fullscreen
• FullscreenPreferenceHUDFadeOutTime – determines how long before the movie controls fade out
• FullscreenPreferenceCaptureAllDisplays – determines whether the background color (usually black) is displayed on every display

Feel free to experiment with the other preferences. Always make sure to restart QuickTime Player after any changes are made, and if anything goes wrong, just delete the preference file. I cannot be held responsible if you manage to irreversibly alter your system (but I can’t imagine an instance in which that would happen).

No human alive is capable of analyzing today’s processors at the transistor level. Digital design these days happens largely at the block level. Simply put, a group of engineers design the individual blocks (arithmetic unit, floating-point unit, registers, data control, bus, etc.) and then another group combines the blocks to create the processor. Specialized software programs, called CAD or Computer Aided Design tools, exist to make this process simpler. They determine how the millions of transistors will be laid out and how they will be wired together.

The 4004, on the other hand, was designed during an era where computers were mostly glorified calculators. Every one of the 2,300 transistors used in the chip was lovingly drawn into the schematic by an engineer. This makes it possible for anyone who has taken at least a semester of logic design to grasp the majority of what’s going on in this chip. These schematics are invaluable for undergraduates who are seeking comprehensive examples beyond those provided by the textbook of complete but simple digital systems. I would have loved to be able to look at these last year and see how my classroom knowledge lined up with a real, commercial product.

The 4004 design is still somewhat relevant when considering today’s Intel microprocessors. The chips that followed the 4004 were the 4040, the 8008 (the first 8-bit processor), and the 8080. Then came the 8086, which was the first of the ubiquitous x86 generation of chips which are at the heart of most computers today. The x86 generation is defined by, among other things, its backwards compatibility. Every new x86 chip must be able to run programs written for previous x86 chips without recompiling. The instruction set can only be added to in successive generations. The backwards compatibility requirement gives x86 processors some inherent shortcomings. For example, all x86 chips are CISC (Complex Instruction Set Computers). However, RISC (Reduced Instruction Set Computer) chips are more efficient, because they implement every instruction in hardware. CISC offers a larger instruction set, but at a tradeoff; some instructions must be simplified into three or four component instructions before they can be executed in hardware. The Pentium is actually a RISC core interfacing with a CISC translator – it could be so much faster without the CISC layer! The PowerPC chips found in Apple’s computers (until a year ago) are RISC chips. This is one reason why some chips cannot be compared using clock speed alone.

To understand the 4004 and the tradeoffs made during its design helps to gain insight into the processors made today. Some of the tradeoffs made when designing the 4004 propagated through the generations and vestiges of them can still be seen in Intel’s chips today. Besides that, it is empowering to know that I can at least partially understand what was state of the art so recently. Of course, the reason at the bottom of all this is that I am a nerd and I love computer history and doing things like reading schematics and learning about computer architecture in my spare time. But regardless of what I specify as the reason, I plan to spend a little bit of time in the near future with the schematics and my Digital Logic book and see exactly how the first ever microprocessor works. There is also a simulator coded in Java on the 4004 site that I plan to check out.

Side Note: The chip in the picture is the C4004 version – the rarest kind to find. They have sold for around $1000 on eBay in the past.

There are two cool hacks that I came up with for this script. Both involve changing ~/.bash_profile (or the startup script for whatever shell that you use). If you don’t have a file named .bash_profile in your home directory, then create one. AppleScripts can be invoked from the command line by using the osascript command, and we can use this to our advantage in two ways.

1. Every time we start up a new terminal, randomize the color. Simply add the line

   osascript ~/Library/Scripts/Applications/
Terminal/RandomColorTerminal.scpt

   somewhere in your .bash_profile (all on one line).

2. Set an alias to the above command so that we can change the color very quickly without needing to navigate to the script menu with the mouse. To do this, add the line

   alias newcol='osascript ~/Library/Scripts/Applications/
Terminal/RandomColorTerminal.scpt'

   to .bash_profile (all on one line). Now, whenever you type newcol at the command line, you will get a new color scheme.

Update: Halloween is over, and TextMate has reverted back to the normal theme. Apparently some people didn’t like the implications of the picture. They probably still live with their parents. Luckily the icon is still available.

For those who care to know, DNS (Domain Name System) resolution is a service usually provided by your ISP (Internet Service Provider). When you type in a site name in your browser, the first thing it does is to query your ISP’s DNS server to find out the IP address of the computer that hosts that site. Once it has done this, it can load the page. OpenDNS claims to shorten the time that this IP address lookup takes by having a huge cache, meaning that they usually can return a result to you without having to ask another nameserver, which would take more time. In addition, they have servers at the main crossroads of the internet. Whenever you see “Looking up (example).com” in the status bar of your browser is when DNS resolution is occurring.

So maybe give OpenDNS a try, if you have access to your network settings. You may be pleasantly surprised at how fast your internet goes. They also claim to offer phishing protection and will block known phishing sites. The service corrects spelling errors, too, so if you type in something like “google.co” instead of “google.com,” you will be instantly redirected. Pretty cool!

I also decided to go with a new theme for Wordpress, the much raved about K2, with all of its AJAX goodness (live search, rolling archives). I’ve been rolling my own with the colors. I originally had a light-on-dark style, but it was too depressing, so I modified the CSS a little and I like what I see. The combination of dark edges with the white page is really pleasing, and I really like how the red titles and blue links stand out. The color scheme seems much more unified than previously. My favorite part is the picture on the top, which one of my parents took, and I edited. Enjoy!

There is a way to switch which key that Mac OS X recognizes as meta by using an X11 keyboard mapping preference file. First, uncheck the “Use the system keyboard layout” option in X11 preferences. You may also have to disable the “Enable keyboard shortcuts under X11″ option as well. Then, create (or add to) the file ~/.Xmodmap and insert the following six lines:

clear Mod1
clear Mod2
keycode 63 = Mode_switch
keycode 66 = Meta_L
add Mod1 = Meta_L
add Mod2 = Mode_switch

It clears the keymappings for alt and escape and then reassigns them to the reverse of what they were before. Restart X11 and it should pick up the changes.

If this was (or wasn’t) helpful, please let me know in the comments!