Jeremiah Oeltjen

Piecewise Functions on the HP48

I had to graph this piecewise-defined function using a graphing utility (read graphing calculator) this evening:

       { -X-5 for -4 <= X < 0
f(X) = {
       { 0.2X^2 for 0 <= X <= 5

Of course I knew the HP48G should be able to do this, but I had never done this in the past. According to the user's manual this can be accomplished by entering a program using if-then-else statements. It worked, except I wasn't sure what do do for those parts of the graph that are not in the domain of this function as the calculator still tries to plot them. In the end I used a catchall branch to return 0 if the independent variable was not within the intervals defined in the function. This was not ideal, but I was able to work with it.

Plotting a function this way proved to be quite time consuming to say the least. I think I spend 15 - 20 minutes on getting it right. Once I completed my homework I went on the Internet and found this FAQ and more specifically this entry regarding this very issue. Restrict the domain of an expression by dividing the expression by the domain.

-X-5 for -4 <= X < 0 becomes (-X-5)/(-4\<=X AND X<0)

To do more than one equation at the same time, put them in a list (curly braces). Don't forget to surround each expression in single quotes:

{ '(-X-5)/(-4\<=X AND X<0)' '(0.2*X^2)/(0<X AND X\<=5)' }

Note: The "<=" "\<=" should be replaced with the "less than or equal to" symbols from the character map or alpha-left shift-3 ("less than" signs are alpha-left shift-2).

Classes Finished

As promised, I'm blogging from the CyberCaf� at school. I'm done with my classes for the day and I'm happy I don't have much homework to do. It is not a nice day. Last night it rained along with some thunder and lightning and it has been raining off and on throughout the day as well.

School Starts Tomorrow

Tomorrow morning I go back to school to begin classes. I was on campus a month ago and blogged from the CyberCaf� on the fourth floor mezzanine that overlooks the cafeteria.

I'm really looking forward to being busy with schoolwork again. I also enjoy using the library facilities and free access to several newspapers including The Saint Paul Pioneer Press. (Yes, I know it is not expensive to purchase a newspaper, but it is more convenient to read somebody else's.)

This year's goals: pay closer attention in class and work on organizational skills.

E-mail Virus

Imagine this. Someone's e-mail system receives a message that is infected with a virus such as SoBig or Klez. By its very nature, since the "From" header is usually forged to hide the true origin of the infected message, it most likely did not come from you, but you get a notification about it anyway. This happened to Chiqui:

It's one thing to send back a message saying "hey! you might be infected with a virus".

It's another to send a message saying "Hey! someone is infected with a virus, we KNOW it's not you, but we're sending you a message anyway for no good reason, because you won't be able to find the infected machine, either".

The real problem I see with e-mail virii and worms is the human factor. People send out notification messages of this sort and effectively multiply the damage caused by the virus/worm by wasting even more precious bandwidth across the Internet.

I have seen it on mailing lists before. Somebody gets an e-mail virus to which their anti-virus software alerts them. The message appears to have come from another member of the list or the list itself. Trying to be helpful, person sends a messages with a subject like "VIRUS ALERT" back to the list accusing someone else for sending them a virus and informing everybody to ensure their anti-virus software is up to date and to avoid opening any messages from so-and-so, blah, blah, blah...

This of course is followed by a rather heated thread regarding the issue of worms and virii which is probably ten to twenty messages long before the moderator steps in to declare it off topic. Of course, those ten or twenty messages got sent to a couple of hundred people. Now we're talking about perhaps 2000 messages bearing down on mail servers worldwide. All of this because one person recieved one message conaining a virus that his anti-virus caught for him anyway?

Conflict in SSH Installations

I've used SSH for years. I used to use SSH Communications SSH client and server. As with all software I install that I don't want to re-install every time I upgrade to a newer Linux distribution I put it in /usr/local. Once OpenSSH became a standard feature in Slackware, I completely forgot about the old SSH stuff and it never seemed to have caused any problems.

Until now.

I have been using SCP a lot lately to transfer files between my home computer and the server on which my website is hosted. This has worked well except when copying a file from the server to home computer by executing scp at the server's command line. I was getting this error:

I scratched my head about this for a while. The mention of "SCP1" pointed me to that old installation of the commercial SSH in /usr/local which, I thought, should not have been used by OpenSSH's SSH daemon. Just to be safe, however, I deleted the old binaries, and voil�, the problem was solved.

It appears that OpenSSH was, in fact, attempting to use the old SSH binaries to perform SCP operations. This was possibly caused by /usr/local/bin being listed before /usr/bin in the PATH.

As usual, this was a fun exercise. Not only was I happy to get SCP working properly in both directions, but I also realized how much junk I really have in /usr/local. Perhaps now would be a good time to re-format that partition and re-install the latest versions of the stuff in there I actually still use.

Time Travel

I don't think I will ever use a time travel device.

I remember thinking to myself years ago that if I ever could, I would come back and visit myself when I was young. Since I haven't met myself from the future, I have come to the conclusion that I will never be a time traveler.

On second thought, there is the possibility that I WILL travel back in time, but to will decide to avoid making contact with myself to prevent the occurance of a paradox. Therefore, I cannot conclude that I will never travel through time.

These things keep me awake at night.

Yuck

I have a box of Bertie Bott's Every Flavor Beans sitting on my desk and just ate one. One could only hope for a cherry or a lemon drop. No such luck. It was sardine.

It's a small box of jelly beans, but they certainly last a long time.

According to Google...

Using the new Google Calculator the number one hundred twenty-three to the fourth is two hundred twenty-eight million eight hundred eighty-six thousand six hundred forty-one.

Google News Alerts

I am trying out Google News Alerts. It is quite simple. Enter a topic keyword, choose "once a day" or "as it happens," and enter your e-mail address. Your inbox will be blessed with e-mails full of headlines that relate to the keywords you chose. This is great for tracking news on things in which you are interested. For example, I set up three: Linux, open source, and Minnesota.

rsync Rocks

Others have said it, why shouldn't I?

Today I backed up the files from my website to my local hard drive using rsync. I never used rsync before today, and now I am sorry I waited this long. The simple fact that it can transparently use SSH as its network transport it great. This way the data can be encrypted too.

Fall Semester 2003

I registered for classes this evening. I'm taking two math classes and a programming class. that's 16 12 credits in all. I was considering adding a health class to the roster, but I think this will be enough of a load. I have every morning unil noon off and all day Friday for the whole semester.

Update: 09-Aug-2003 It took me this long to realize 4 * 3 != 16

Back on Google

Searching for the terms jeremiah oeltjen now yields my blog rather than my homepage as I mentioned in a previous post. I am certainly not going to complain about that. The blog is where everything happens anyway.

P.S. Today, searching for jeremiah finds my blog as listing number 783.

HP48 Morse Code Program

Arlen, AA0SG, wrote this wonderful Morse Code generator for the HP48 series of graphing calculators. It is based on the Farnsorth method with three general speeds and configurable audio frequency.

1. Download and transfer the program to you HP48 calculator. This will create a directory call MORSE.
2. In that directory, pressing the CST key will bring up the custom menu.
3. The MESS vafiable stores the string to be sent. Enter a string on the stack and left shift the MESS variable key to store it. Right shifting the MESS key recalls the message. Only uppercase letters are valid.
4. Press the RUN key to play it back.
5. Entering a number and pressing the HZ on the custom menu sets the audio frequency (in Hertz).
6. SLOW, MED, and FAST set the speed.
7. AR, SK, comma, and question mark insert the approriate characters tnto the string.

Download Now

Disposable Digital Cameras for $11

An article in New York Times states that Ritz Camera is selling single use digital cameras for $10.99. My first reaction was that this would be good to use for kite aerial photography, but someone at Stanford using a balloon found one for's $37, which would be a better deal if this is something you do a lot.

Gone From Google?

For the past week or two I've been querying Google with just my first name about once a day to see how much its ranking in the search results fluctuated from day to day. I wasn't keeping track, which would be an interesting exeriment using the Google API, but the index page was holding position anywhere approximately between 405 and 440. Obviously, not as ideal as Jeremy, Kasia, or Chris.

Yesterday I couldn't find it at all. Not in the first 700 anyway. Finally I searched for my first and last name [screenshot]which has had the index page on top for a couple of years now. Instead of my main website I have a couple of TrackBack listings. Clicking on more results from kb0off.sheddwellers.org results in a listing of every TrackBack page on my site, foaf.rdf my resume and a link to my main blog page without a title, trailing slash in the URL. Coincidentally I recently added "/mt" to my robots.txt file because I didn't think that it was necessary to have spiders crawling those pages. Now, thinking back on it, I feel that was unnecessary because those pages contain links back to people that have sent TrackBack pings to my blog which would (should), in turn, help their PageRank. There's also the strange coincidence that this happened at the same time that I began using the new URL's.

There must be a reason for all of this. Restricting robot access from one directory and setting up a few redirects isn't going to get your page unlisted. I'm actually quite suprised that it was unlisted at all. The only other explanation I can conceive is that I am at the receiving end of some foul play.