Exponential Decay of History -
A common, naive approach to history is to keep a simple linear ring-buffer. When the buffer fills, we simply overwrite the element from some time ago. An advantage of this technique is constant space overhead. Unfortunately, a linear history too easily loses important contextual information about the deep past due to minute updates in the recent past.
The further into the past we look, the less information we need, but we don’t want to have any particular cutoff where we lose the history. What we need is a decay model with predictable computational properties. What we need is exponential decay or something similar – i.e. information with a half-life.
If you see a UI walkthrough, they blew it -
Unfortunately a UI walkthrough is quite an inelegant way to explain the core functionality of an app. It can be a frustrating obstacle before you can dive into an app, and you have to remember all of those new ways of using it once you get in. […]
Arguably a less intrusive way compared to a walkthrough is to guide the user in the situation with UI hints. This can be done through slight visual cues and animations. A hint should not be a popup (it’s probably even more disruptive than a tutorial).
Half the Facts You Know Are Probably Wrong -
A 2010 study in the journal Scientometrics, looking at data between 1907 and 2007, concurred: The “overall growth rate for science still has been at least 4.7 percent per year.”
Since knowledge is still growing at an impressively rapid pace, it should not be surprising that many facts people learned in school have been overturned and are now out of date. But at what rate do former facts disappear? […]
Applying the concept of half-life to facts, Arbesman cites research that looked into the decay in the truth of clinical knowledge about cirrhosis and hepatitis. “The half-life of truth was 45 years,” he found.
In other words, half of what physicians thought they knew about liver diseases was wrong or obsolete 45 years later. Similarly, ordinary people’s brains are cluttered with outdated lists of things, such as the 10 biggest cities in the United States.
Beating the Averages -
As long as our hypothetical Blub programmer is looking down the power continuum, he knows he’s looking down. Languages less powerful than Blub are obviously less powerful, because they’re missing some feature he’s used to. But when our hypothetical Blub programmer looks in the other direction, up the power continuum, he doesn’t realize he’s looking up. What he sees are merely weird languages. He probably considers them about equivalent in power to Blub, but with all this other hairy stuff thrown in as well. Blub is good enough for him, because he thinks in Blub.
Paul Graham on choosing a programming language
‘Kurt Vonnegut’ by Mat Pringle. Click to read!
Portfolios on the Web -
Web designer Daniel Eden on the issues with presenting a portfolio:
We don’t design pictures. We design experiences.
An experience is something that is impossible to display in a static format like a picture.
What Programmers Want -
Michael O. Church on what motivates a software engineer (and what does not work)
Zen Coding is now Emmet — the essential toolkit for Web developers -
Emmet takes the snippets idea to a whole new level: you can type CSS-like expressions that can be dynamically parsed, and produce output depending on what you type in the abbreviation. Emmet is developed and optimised for web-developers whose workflow depends on HTML/XML and CSS, but can be used with programming languages too.