Wednesday, October 29, 2003

A Book in the Hand, and New Eyes for its Complexity

A change of perspective gives rise to an intuitive image of dazziling complexity

Think of some piece of sculpture, real or imagined, it doesn't matter. Visualise it, see its shapes, its curves, its lines, its angles. How could the complexity of such a work be expressed? How could it be measured? I have no idea what the answer is, but I'd expect that a major part of it would involve the complexity of its shape. Perhaps mathmatical tools of topology could be used to give some fairly objective measure of it, for example. Hold that thought.

Now imagine you have a book in your hand, again any book will do. We can also think about a book's complexity. This is something which has no doubt been done by various people, and I imagine that issues such as the complexity of the writing, the plot and the characters are common parts of such attempts. The other day it occurred to me a very different conception of a book's complexity. I'm not saying it's a better way, but it is quite different, and I think it is quite provocative.

It's really quite simple, you look at its shape. Turn it around in your hands and you'll see it's a prism. Open it up and look inside and you see what really makes up the book, the words. This change of perspective reveals a dazzlingly complex construction: the intricate shape of a letter, appearing in each of the book's letters, chained one after the other, in an edifice spanning thousands and thousands. This, the combined shape of each and every letter stretching from page one to the end, is the shape of the book.

Friday, October 24, 2003

Book Review: The Curious Cook

The Curious Cook: Taking the Lid Off Kitchen Facts and Fallacies, Harold McGee, 1990.

There's a kind of spectrum of tasks, at one end of which are those things you learn by doing, and the other end those things you learn by studying. We might learn history in books, but our knowledge of cooking is, by and large, firmly placed over the kitchen stove.

It's not hard to see why. Over many generations and millennia, people have figured out what works and what doesn't. In a continuing cycle, most of us learn from a parent or relative or from some cookbook, in a process of imitation, which gives way to an intuitive understanding of what works and what doesn't, and on and on from generation to generation.

Our understanding of cooking is on an intuitive, functional level. We can understand that adding a bit of salt will enhance the flavours, and we know roughly how much salt is required to do this - and how much would make the dish unpalatable. Which seems fair, for if we were to get an understanding at the next level down, we'd be getting into the realms of chemistry. And at this level, things get pretty damn complex!

It would seem that you'd have to know a hell of a lot about the chemistry, the structure of the cooking materials, how they combine and interact in the process of cooking in order to improve upon the millennia of trial and error experience and incremental improvement!

It is in this context that Harold McGee's "The Curious Cook", aimed at the lay reader, successfully brings to bear some understanding of underlying science onto various matters food. The result of this rare combination is interesting, informative and a pleasure to read - and it shows that a deeper understanding can sometimes be very useful. McGee not only knows his science, he also knows how to write in a clear and entertaining manner.

The book is organised as a collection of essays split over three sections. The first covers some kitchen experimentation to uncover the truth behind some kitchen lore, and to troubleshoot some common problems, such as how to keep salads green and fresh. The second looks at matters of health, covering the role of food in heart disease and cancer, and what part aluminum might play in Alzheimer's. The third entails some miscellaneous reflections relating to science and food.

In the first section we learn such things as why searing meat doesn't really seal in the juices, why bleurre blanc is so simple yet other similar sauces are so complex, a simple way of making mayonnaise, and simple formulas for getting the right proportions of ingredients for making various different types of water ices (something that you won't be able to find elsewhere).

As you've probably sensed from that list, the range of topics he covers is fairly eclectic, and, speaking as someone who's more interested in Indian and Asian cooking, it's also oriented towards European food. But even so, it's fascinating stuff, and a lot of it reads more like a mystery novel than a science text or cookbook. He'll have a promising lead, go down the path of exploring it, only to discover it was a red herring, and off he'll go on another lead.

One thing I found quite surprising was how wrong certain kitchen lore was, and how easy it was for him to systematically check it out using a simple kitchen set up (as opposed to an industrial lab and equipment). Like searing meat to seal in the juices - it makes intuitive sense, but if you actually check it out, it doesn't make any difference at all. Yet it's common lore that's been handed down and accepted from generation to generation. The only benefit seems to be that the searing helps to initially brown the meat.

The second section moves from food and cooking to its effects on our health. Specifically, he covers heart disease, cancer, and the potential link between aluminum and Alzheimer's. I'm a bit squeamish about these kinds of things, but I actually found these chapters more interesting than anything else. It's fascinating, and at the same time kinda mellowing, to realise that heart disease and cancer are more the result of imperfections in our body's functioning than anything else - for example, most carcinogens are do not themselves cause cancer, but are turned into things which cause cancer by our own bodies.

Most popular treatments of these issues often talk in very simplified and vague terms, and I thought this section really helped you understand the issues by giving you an account of what's actually going on in the body. In particular, the description of heart disease was very nicely written.

While he attempts to summarise the current scientific understanding of these matters, it's worth noting that the book is close to 15 years now, and I wonder how much things have moved on since then. In any case, I expect the understanding of the fundamentals he presents is unlikely to be much changed.

The third and final section is a scattered collection of three essays related to science and food. The first two are historical in nature, one talking about Brillat Savarin (remembered for a handful of epigrams, such as "Tell me what you eat, and I will tell you what you are"), who was one of the first to foray into the world of food and science, and the other chronicling the history of the substance Osmazome, which people once thought contained the essence of the flavour of meat. The final essay reports recent research findings that shed light on why we find the flavours of cooked food so desirable (while most animals, and indeed our predecessors, don't).

Entertaining, practical, informative and unique - The Curious Cook is well worth a look.

Home Laser Engraver/Etcher



The VersaLaser is basically a PC printer that allows you to engrave and etch designs onto wood, plastics, leather, stone and other materials. It uses a powerful laser, which can also cut through various materials (Article; Pics/Specs; Discussion). Apparently, similar laser-based systems have been around for a while, but not as a PC peripheral. One will set you back about $10 000 (US, I assume). Futher encroaching of the world of information onto the world of atoms.

Today's Comics

Monty - Night Vision Goggles, not bad
(Others: Pearls Before Swine, Dilbert, Get Fuzzy, Herman, Speed Bump, Strange Brew)

Thursday, October 23, 2003

A Nice bit 'a UI Innovation

screenshot

DateLens is a piece of Calendar Software with a very interesting focus + context, zoomable user-interface (Details and flash-based demo).

The basic view is a grid, with each cell representing a day. You can zoom in on a particular day, whereby that cell will enlarge and, to give it more space, the surrounding cells will contract a little. The zooming is smooth and works in a progressive fashion: you can go from a small cell for the day, to a summarised view, to the full details. I haven’t tried it myself, but the interface looks quite natural and flowing.

Tuesday, October 21, 2003

Idea for Web-browser feature: Highlighting Equivalent Links

Here's a simple idea that could provide a minor enhancement to the web-browsing experience. If there's more than one link on the visible portion of a page pointing to the same target, then when the user mouse-overs one of those links, have the browser highlight the rest of them. This way, the user can be alerted to the fact that they all point to the same place. As far as I know, there aren't any browsers that implement this and I'm not aware of anyone having tried it - but then I hardly keep close track on these matters.

It's not uncommon for a webpage to contain multiple equivalent links. Links on the site's navigation side-bar can also appear with in the body of the page. Or the page may simple reference the same document in multiple contexts; sometimes these links may point to different parts of that document. For example, a page on the university site may make repeated references to the university's as policies outlined on a particular page. Or certain links from within the body of the page may be distilled as references at the end of an article.

That all these links point to the same place may not be clear, however. The link text may not be the same on all links, for one thing. (In Internet Explorer, if you open one of the links in a new browser window it gets marked as read but the other equivalent links stay the same until you refresh the page, but this issue would be best remidied by immediately marking them all as read).

Even if the link text is the same for all the links, it won't necessarily be obvious that they're all pointing to the same page. I've had a number of experiences like the following. You might have seen a link in one paragraph and decided not to follow based on its link text and the context. Then, a minute and a few paragraphs later, you see another link and click it, because it this context it seems interesting or useful - only to realise that it's the same one you didn't want to visit earlier.

Because you were like most readers and didn't devote a heap of attention to the target of each link you mouse over, you forgot that the two links were the same. Memory is made more difficult by non-mnemonic URLs that are hard to read let alone remember. If the relationship was shown by the browser, I think this mistake would be harder to make.

If pages were written well, the destinations of links would be clear, and this problem wouldn't help provide justification for such a feature. But often they're not, and tools such as well browsers should be designed to best support the realities of the task they're facilitating.

I think it'd be interesting to try this idea out and see how it flys. It looks like it'd be easy to implement, and aside from it potential usefulness, I think there are other reasons it might fly: it's an incremental, evolutionary addition to the browsing toolkit, and it requires no overhead or change of habits to use. I also don't think it'd be too distracting, nor too confusing to inexperienced users.

Unfortunately I've got too many things I'm pursuing that are higher-priority for me than learning enough about a browser to have a go at implementing this. So if anyone is interested in giving it a go, please let me know how it went.



There's a lot more I'd like to say about this feature, because I think there are a few other situations it'd be useful in, and I'd like to go a little deeper into the underlying reasons why it's useful (or at least seems so to me). There's also some more sophisticated ways this could be implemented (so you can see all the equivalent links on the page, not just those that are currently visible), and I think the technique that could be used to do this would be useful in a number of contexts. There's just not the time at the moment :-), so I'll have to deal with these matters later...

Updating the food pyramid

The well known food pyramid, offering simple prescriptions for a good diet, is now just over ten years old. Scientific American have an article from earlier this year on what we've learnt about dietry health in the intervening years. It's both fortunate and unfortunate that, as the title of the article ("Rebuilding the Food Pyramid") suggest, we've learnt a fair bit more. Unfortunate, because better health will require changes in habits and established lore, but fortunate, because it ought to help bring us closer to that goal.

Songs that've got me right now.

At any time, there's always a few songs that make me want to crank the radio up when they come on.

I really like that new Outkast song, "Hey Ya!". It's got a great upbeat, funky energy to it, and for some reason it sounds to me refreshingly un-run-of-the-mill. When I listen to it, I get the impression that the guy doing the vocals has a real intimate familiarity with how to use his fairly unique voice: the tone, emphasis and timing combine brilliantly and there's a real smoothness and confidence in the vocal delivery. There's some nice lines, too: "ladies.... give me some sugar, I am your neighbour" and "Shake it, shake it, shake it, shake it, shake it like a Poloroid Picture".

The other one that's got me at the moment is "That Great Love Sound" by The Ravonette's (listen to a sample, at Amazon.co.uk). It's well-crafted guitar-rock/pop, and definitely worth checking out. I love the bit when the female vocals echo the "makes me want to scream and shout" (or perhaps 'sing and shout', I can't remember) line.

Monday, October 13, 2003

Friday, October 10, 2003

An interesting future for TV

Director of BBC New Media & Technology on the what the future holds for tv (PaidContent.org). Increased user control, localised content, more user input, ambient tv, programme sharing... and sooner than you might think. The article seems pretty insightful.

Today's Comics

Speed Bump - Author Apparel, not bad
Herman - Afraid of Dogs, not bad
Monty - It's Over, okay
Pearls Before Swine - A Fish Called The Truth, okay - it's a pun
(Others: Dilbert, Get Fuzzy, Strange Brew)

Qualitative and Quantitative Reality

The terms 'Qualitative Reality' and 'Quantitative Reality' occurred to me the other day, and they seem useful.

These terms aren't new, but they seem less common than I think they ought to be. I checked Google, and if you search for 'Objective Reality' you get 72,000 results, and 32,500 results for 'Everyday Reality'. While 'Qualitative Reality' brings up 176 results, and 'Quantiative Reality' only 158 results.

They seem useful because often the underlying, objective reality is not the thing of importance. The thing of importance is often the reality that we perceive and which we can measure.

In the large, the Qualitative and Quantitative Realities are the ones we live in, and I think these terms help to bring this to the fore. They can help us realise that much of our reality is changing, is a function of our technology and understanding.

Wednesday, October 08, 2003

Tuesday, October 07, 2003

Office 2003 gets thumbs up for XML functionality

XML capabilities in Word, Excel, and InfoPath help bridge the gap between desktop documents and databases, and give enterprises a reason to upgrade (Udell)

Sunday, October 05, 2003

Thursday, October 02, 2003

Virtual Hurricanes

Supercomputer simulation of earth's weather models down to level of virtual hurricanes (New Scientist). This is the Earth Simulator in Yokohama, Japan -- the world's fastest supercomputer -- which is aparently proving to be a very useful tool for understanding and predicting the weather.