Update on the open textbook stuff

Well, it is clear that if I want this to succeed, I am going to have to launch a major marketing campaign.

But, as I suspected, the commercial supplementary site seemed to get lots of play (by reading the web logs on my web page server), mostly from international students -- of course it is summer time. Students spend hours searching the web for the solution to their current homework, or if they can't find that, then they seem to look for simplified explanations that makes it trivial to solve their homework.

The content on my supplementary web site is copyrighted, but free to people who register with me. Originally, I just put polite messages in prominent places asking people to drop me a note if they were reading the material. They ignored me. So about a month ago I split the site into a "sample" set of supplementary materials, and a more comprehensive set that people are required to have a login and password to use. I still get many queries on the sample material, and I have a few people who have registered to use the materials, but basically people just won't register to use the material. Clearly I have to rethink the model for delivering the supplementary materials as well as the open textbook.

The Nokia N810

A couple of months ago I mentioned my experience with a Sony UX 380N. Recently I have also been using a Nokia N810. There are many differences between the two machines: The Sony is really a scaled down notebook computer with Vista and a 40GB disk, whereas the Nokia is a PDA on steroids -- it has a modest amount of memory, an external flash memory card, a slower processor than the Sony, and it uses Linux. To me, the two machines serve different purposes: if I don't really want to schlep a notebook computer somewhere, I can take the Sony and still read most things I could on my laptop (but the keyboard is too small to do any significant input work). While the Nokia has an Xterm window and a full keyboard, it could not really pass as a legitimate Linux machine, but it is much smaller than the Sony, has wireless and a means for hooking up your cell phone as a modem (my cell phone doesn't seem to work with it). You can certainly get by reading the web better than on a cell phone (probably about like an iPhone, but I don't have one of these), and reading your IMAPped mail. I take the Nokia with me when I go out for an afternoon, say shopping with my wife, where I expect that I will have 15 minutes to kill here and there -- the Nokia is the perfect toy for this.

The battery capacity on the Sony is not very good, but the battery capacity on the Nokia is excellent -- a charge last several hours. I haven't used it on a long plane ride, but I can see that it would be a great little entertainer for 3 hours or so.

If I were going to buy one or the other, which would it be? Well the Nokia is a lot less expensive, so it has that going for it. But the Sony is a little laptop, so is far more capable. I like them both.

Full Draft of Single-threaded SCC OS booklet

OK, I finally have a new draft of the entire manuscript for my web book(let) on operating sysetms for very small, communicating computers. It is an open book, so anyone can read it. (If you want to modify it or derive work from it, you can do so under a Creative Commons license.) Take a look at the manuscript. Please give me your feedback and suggestions for how to let more people know about this (without spamming them).

As mentioned earlier, I am developing a commercial site to provide supplementary materials for this book, as well as for general OS study. Please feel free to review the content. You need to register to use this site, but it is still free to use at this time.

A Coroutine Simulator

Earlier I mentioned that I was working on an open textbook experiment. One of the pedagogic technologies is a bastardized coroutine idea -- enabling me to talk about single-threaded systems (like tiny embedded systems) that still have ultra lightweight autonomous units of computation &mdash go see Chapter 2 of the draft if you are interested in the details. I mention this, because in November I started writing a Linux simulator for these coroutines. I wanted something that would enable students to write simple programs using coroutines, and also to have simple small computer devices. I ended up using threads to simulate coroutines, with a synchronizing variable to assure that only one thread simulator (coroutine) would run at a time. This worked just fine. Next, I wrote a framework in which an instructor can write another process to simulate a device, so I can write exercises that say something like "write a program with a set of coroutines that use one coroutine to control each device, yada, yada, yada, ...".

An MS student, Tristan Paddock, decided he would like to flesh out my hacked idea, and has started in earnest on the code. If you want to know more about this, or to just chat about it, drop me a line.

Using the Sony UX-380

A month ago I took a trip to Costa Rica. I took my new Sony UX-380 with me, since it was small/portable, WiFi/bluetooth/Cingular, lots of storage, etc. It worked well as a vacation machine for me, but I would not been able to have done any real work on it for the two reasons I have seen in a number of places on the net: (1) The keyboard & screen are too small for serious work, and (2) the battery capacity really is small. I was not that pleased with Vista, because it uses more resources that I want it to use (it is S L O W) — but everyone has the same complaint. But I am not sure I want to mess with it by putting XP on it ...

A friend was using a Nokia Linux machine with wireless. Her machine was all stylus input, but I liked the tiny Sony keyboard better. By the way, I have pretty big hands, but the keyboard is still a little too large to type with your thumbs like you would on a Blackberry or Treo. At the price I paid, I would buy the UX-380 again ... for the usage scenario I sought.

Adam Beguelin Talk

Back in early January I watched Adam Beguelin's talk at Justin.tv/hackertv (the third in a series of such talks). Adam is a very bright, successful guy with a great track record at everything he tries — Ph.D. in computer science, faculty member at CMU, member of two successful startups (Inktomi and Truveo), Distinguished CU Engineering Alum, etc. In the talk, he focuses on startup life, sharing his insight into some strategy and hints. It is worth viewing if you are interested in entrepreneurship (which I am).

The Open Textbook Project

I have been working on a grassroots effort to do open textbooks. I wrote a proposal for this to an NSF program, but they didn't like the idea as much as I do. The idea is that textbooks are very expensive -- it seems like they cost $100 whether they are a simple paperback or a multicolor hardbound book. Students have revolted; in many of my classes, students simply do not buy a textbook.They use the web, borrow a friend's book, or ??? The book(s) will be published under the Creative Commons license. People can read it without restriction, and they can edit it (using a Linux-like model).

A year or two ago I submitted a manuscript for a book about OS for small, communicating computers (this all started several years agowhen I was thinking about the Palm OS, then WinCE/PocketPC/Mobile, embedded Linux,TinyOS, Mantis OS, etc.). The publisher declined to publish the book because there was no course that defined a market for thebook. I have had some of the material (that is not copyrighted elsewhere) up on a site for a year or so. I have not been happy with that, so I am thinking of ways to repackage it so that it is more useful to people. More on this later ...

The catch is that I am not publishing any exercises, lab exercises, or other supplementary materials with the book. It is the content without the pedagogical tools. The other part of the experiment is to see if content were free, but the pedagogical tools had a modest cost, would this be viable? So the second half of the experiment is the commercial supplement site. I'll keep you posted as things evolve ...