Contemplating netbooks

Having played around a bit with Tristan’s EEE PC, I am considering getting a netbook computer myself. My old iBook is not very portable and, while the big screen and keyboard it offers have advantages, there is considerable appeal in a machine that could be a default content of my backpack.

Do any readers have experience with particular netbooks? I would be looking for something that is tolerable for writing emails and blog posts on, and good for surfing the web. Long battery life would be an advantage, as would an operating system that does the most common tasks well.

I won’t be buying anything for a couple of months, at least, due to a bit of a financial crunch, so machines that look promising and have not yet been released are also worth mentioning.

The Code-Breakers

For those with a serious interest in the history and practice of cryptography, David Khan’s The Code-Breakers: The Comprehensive History of Secret Communication from Ancient Times to the Internet provides an enormous mass of knowledge. The scope of the 1200 page book is vast: covering everything from the earliest ciphers to the origins of public key cryptography in detail. It is probably fair to say that the period best covered is that between the Middle Ages and the Second World War, though the sections covering the decoding of Egyptian hieroglyphics and Linear B are also detailed and skilfully written. Those interested primarily in the contemporary practice of cryptography – or those seeking a more concise text – would be well advised to consider the books by Simon Singh and Bruce Schneier on the topic.

Khan’s book excels in actually describing how various cryptographic systems work, as well as how they were broken. For the most part, his analysis is factual and dispassionate. The sole exception is in the period covering the Cold War, in which his ire against the Soviet Union and those Americans who turned into traitors for it are acute. At times, the book gets into an excessive amount of detail about the bureaucratic organization of different cipher bureaus: including lengthy sections about how various wartime bodies were reorganized. In most cases, the book does not provide much biography on the men and women involved, though exceptions exist in the case of some of the most eminent or interesting cryptographers. The book does provide an interesting discussion of the history of writing on cryptography, including the impact that major publications had on the development of the field and its comprehension within society at large. Kahn also does a good job of debunking some of the many spurious claims that have been made about ‘revolutionary’ and ‘unbreakable’ cryptosystems that people have invented: stressing how the making of cryptographic systems is a realm of abstract mathematics, while the breaking of such systems is a gritty and practical exercise.

In addition to covering the techniques of cryptography and cryptanalysis themselves, the book covers many related security issues: including physical security, invisible inks, elements of spycraft, decisions about how to use information gleaned through cryptanalysis, and the use of broken cryptographic systems to transmit fake or confusing information. The book also covers the relationships between cryptographic work and the activities it is supporting. An especially intriguing section details the efforts of the American navy to combat rum smuggling during the prohibition era. Ships with floating cryptoanalytical laboratories provided vital intelligence to interception vessels, just as other cryptanalysis had helped re-direct U-boats away from German submarines during the Second World War. The book covers an enormous variety of code systems, ranging in use and sophistication. These include diplomatic and commercial systems, high level military systems used between major installations, systems for vehicles, trench codes for those on the front lines, and more. The most abstract section of the book contemplates communication between human beings and extraterrestrials, covering questions about how we could recognize alien communication, as well as mathematical steps through which a comprehensible discourse could potentially be established.

For those interested in actually breaking codes and ciphers themselves, the book provides detailed information on techniques including frequency analysis, factorization attacks of the kind used against polyalphabetic substitution ciphers, and the index of coincidence. It also provides a lot of information on the weak ways in which cryptography is often used and the kinds of errors that have allowed for key breaks into previously unreadable cryptosystems. While it would not be especially useful for attacking modern computerized cryptographic systems, it would provide some guidance for those seeking to break into amateur or puzzle-type cryptographic challenges.

The Code-Breakers may well be the most comprehensive cryptographic history available, though it is far less detailed in its description of post-Cold War cryptosystems than some of its more concise recent contemporaries. For those wishing to gain an appreciation for how cryptography emerged, the role it played for most of human history, and the techniques that have been employed to guard and attack messages, this is an ideal place to turn.

Distributed tremor detection

Jesse Lawrence is an Assistant Professor at Stanford University, primarily interested in earthquake seismology and distributed computing. One idea has now merged the two fields: using the accelerometers increasingly commonly built into laptops and phones to make a distributed system for earthquake measurement. By having lots of sensors, it is possible to distinguish earthquakes from other forms of motion. The distributed approach also has advantages: it can provide more detailed information about extreme vibrations than delicate seismometers. It can also provide data collected at many more points, increasing understanding of the earthquake as an effect across a large area. Apparently, with appropriate signal processing, it would be possible to use the system to warn people in surrounding areas not yet affected by the quake, since the data could move more quickly than the seismic waves themselves.

Those wishing to join the Quake-Catcher Network can get the Mac or PC software online. Presumably, people in California are especially encouraged to enroll.

Keeping track of discussion threads

One of the major reasons for which I keep writing here is because of how it forces me to engage with and clarify my own thinking on important issues. One of the most important mechanisms through which that occurs is the discussions that often accompany posts.

I realize that it is awkward to keep re-visiting the same post over and over, looking to see if anyone has responded to your comment. To make it easier, there are two alternative options for seeing new comments:

  1. You can subscribe to an RSS feed of the comments. If you don’t know what that means, this guide provides an introduction.
  2. You can sign up to receive the comments daily by email.

Either way, you can keep track of discussions (as well as links to news items relating to posts) more easily.

Attacking encrypted bitmaps

Just because your photos are encrypted, it doesn’t seem that you can count on them to be totally unreadable to someone without the key. The attack only seems to work against bitmap images, so those secret JPGs, PNGs, and GIFs should be safe for now. This is because most types of files contain significantly more entropy than bitmaps. That is to say, there is a lot more redundant information in a BMP file than there is in something compressed. Even in the case of the vulnerable images, the technique can only produce “the outline of a high-contrast image.”

Once again, it proves the statement that ‘you can’t hide secrets from the future with math.’ Cryptographic attacks – and the resources available to attackers – will only keep increasing over time.

Global preferences regarding US presidential candidates

Given the degree to which the American president influences events all around the world, there is a certain degree of sense in polling the rest of the world to see which of the two current presidential candidates they prefer. This page on The Economist‘s website is doing just that. It is set up to mimic the American electoral college system, with each country getting three votes by default plus additional ones by population. In total, there are 9,875 votes.

At the time of writing, Canada’s 49 electoral college votes are going to Obama, who is preferred to McCain by 87% to 13%. The 432 American votes are also going to Obama, reflecting a 79% to 21% preference. The only countries that are toss ups or leaning towards McCain seem to be Macedonia (5 votes) and Andorra (3 votes). As such, Obama is leading by 8,360 to 8.

Of course, there are huge methodological problems with this type of survey. It is amalgamating the preferences of those who have volunteered to take it, and who therefore presumably have some knowledge of The Economist. It is neither a random nor a representative sample. Even so, the results are pretty striking.

An idea for reducing problems with dead links

As mentioned before, one of the most frustrating things about the internet is the likelihood that following a link will lead to a page that has moved, changed, or vanished since the link was posted. Given the massive increases in bandwidth and storage space that have taken place, I had an idea for combating this. Basically, it would be an automated system that saves a cached copy of any linked page, then allows anyone viewing the linking page to view the saved version of the linked page, in the event the latter became unavailable. That means a blog post linking to a news story or other blog entry would be able to provide access to either of the latter, even at a point in time when they are no longer available in their initial contexts.

It would work a lot like Google’s cache: saving text and formatting, with links to any images and video in their original locations. As such, the maximum amount of content would be retained without using too much disk space. To begin with, the system could be integrated into content management systems like WordPress. Eventually, it may be sensible for every link created to express this behaviour by default – at least on websites that choose to enable it.

Like so many useful things, the system would but up against copyright restrictions. That being said, Google has thus far been successful in defending the legality of their own caching practices. Perhaps the courts would be willing to consider the kind of enhanced links I described as a fair use of potentially copyrighted material.

A few Apple complaints

Last night, after the Bluetooth connection failed for the hundredth unexplained time, I switched back from my Apple wireless Mighty Mouse to my old Microsoft optical scrollmouse. I must say, the change is for the best. The old mouse is lighter, smaller, and more comfortable. It is possible to press both buttons at once, and press the middle button without accidentally scrolling. Most importantly, the scroll wheel itself is much less finicky – it may not be able to scroll horizontally, and it lacks the Might’s Mouse’s useless ‘squeeze’ buttons – but it seems the superior device overall, despite the need for it to be tethered to my computer.

In general, I think Apple does a magnificent job of making computer gear and software. If I had to make two complaints, the first would be about the way they sometimes privilege form over functionality. Alongside the Mighty Mouse (and the infamous prior hockey puck mouse), there is the interface of Time Machine, which is pretty but probably less useful than it could be. My other complaint is their willingness to change things after the fact in ways that cannot be reversed and that people might not like. For example, there was when they locked iTunes so that only three people per boot session could access your library over the network (a real pain in university residence), or when they limited the volume on my iPod Shuffle through a software update.

A Wikipedia paradox

The site is most useful when you know either absolutely nothing about something, or a great deal about it. It permits those utterly unaware of a topic to get some essential facts – probably true – very quickly and easily. It also allows real experts to track down something they once knew, can remember, but had forgotten very quickly.

Wikipedia is least useful for those in the middle zone. These are people who know more than the minimum, but not enough to really judge the credibility of complex arguments in the subject area.

Nonetheless, it is a wonderful resource. I use it at least twenty times a day.