XKCD on Death

Google Wave is in Beta

Yes, really:

Design lesson: outages are an opportunity to make users smile, rather than irritated.

Using Gmail as an OpenID Provider

 

• Login to gmail
• At the Relying Party login page (where you’re trying to login) enter https://www.google.com/accounts/o8/id

What were they smoking when they thought this up?

(Perhaps marketing people saw it as a means to get Google logos alongside other logos that tend to appear beside OpenID? Most un-Google.)

(Thanks)

The date command and ISO 8601 date+time representations

8601 provides a variety of interesting format variations. I routinely wish to use, but can never remember, the following:

$ TZ=UTC date +%Y-%m-%dT%H:%M:%SZ
2009-10-29T04:10:09Z
$ date +%Y-%m-%dT%H:%M:%S%Z
2009-10-29T12:10:43SGT
$

Recorded here for easy reference next time.

The need to “create jobs”

It just occurred to me that Paul Graham and Seth Godin were both covering similar ground, about a year apart: An Alternative Theory of Unions and Henry Ford and the source of our fear

A long-lost link and Google’s allegedly making us stupid

It has bothered me for some time that Google is frequently portrayed as making us stupid (this article for example, but there are hundreds more) not just because I happen to like being able to search a large fraction of the sum total of [written] human knowledge at will, but because the complaint just doesn’t ring true.

Sometime in the last week I was dreaming about this:

I don’t recall anything happening, just a recollection that it exists, that it has some relevance to distributed computation, that a shallow copy of it was on the first web-page that I ever created and that I wanted to find it, along with some sadness that I wasn’t so thorough in my archiving then as I am now so it would be hard to find. I promptly forgot all about it until a familiar passage of words appeared in the references section of one of dozens of papers that Google yielded earlier today while I was searching for something else entirely; 30 seconds later the graphic above was on screen in front of me for the first time in a decade and a half.

(What it is and how I found it aren’t actually the focus of this post so I’ve moved them to the notes at the bottom.)

What this incident brought to mind for me was the immense power that Google brings to my finger-tips. The ability to find that long-lost document, indeed the means to do so serendipitously while pursuing something else entirely, is something that would simply not have been available to me, nor almost any other human being, just over a decade ago. This is an entirely new capability for human beings to have readily at their disposal in much the same sense that cheap electronic pocket calculators were a few decades ago or – perhaps more relevant – the way that writing was millennia ago.

In fact I’ll go out on a limb: This casual power that is now so readily available to us was unimaginable just half a century ago.

People had “imagined” all-knowing machines of course, but amongst those who gave the slightest thought to how such a thing might be implemented, none came close. The nearest counter-example that I’m aware of is Ted Nelson’s seemingly off-the-wall ideas about hypertext in the early 1960s, most of which leapt abruptly into being within a few years of the invention of the web. Another decade out and even his visions seem rather modest.

“But wait” the nay-sayers wail “people today aren’t as good at X as they were in my day”. This is probably true:

• No doubt there was concern expressed that a generation of engineers didn’t really understand multiplication and exponentiation as intimately as their forebears did when pocket calculators became widespread.
• When literacy became widespread in Europe, I’d hazard a guess that the troubadour’s ability to recite lengthy sagas diminished sharply, yet few would consider an illiterate person today to be more able than a literate one.
• Events in Sumer are so far lost to us that it’s hard to know how the dependency upon written records was viewed but, in a historical sense, surpluses arising from permanent agriculture combined with record-keeping to make possible the formation of human civilisation. It’s difficult to view this as as any sort of disability.

My guess is that, 1000 years hence, Google and its ilk will appear to be inventions as important as the invention of writing itself. The ability to recall things at will that one knew months or years earlier presaged an enormous increase in human ability; the ability to “recall” things one never knew, but which were known earlier by any of millions of others, can scarcely have less impact.

In some sense those concerned are correct: we probably are losing certain abilities to retain trivial information unaided, but the skill of doing so – and the need to – belong to an earlier era, in the same sense that the ability to use a slide rule, recite hours-long stories or live from birth-to-death absent human civilisation were.

Further thoughts:

• What I dreamed about was actually the graphic for Luca Cardelli’s lexically-scoped untyped interpreted language that supports distributed object-oriented computation named Obliq. I’m pretty sure that I didn’t understand it when I read it 14 years (!) ago. I may now struggle to make the time to read it again. As I now know that the author’s name now means considerably more to me than it did at the time, my motivation to make that time is strengthened.
• Finding things on Google when all you have to go by is the dim recollection of a graphic is tricky, even for Google images. I stumbled across my long-lost link while researching a blog post for Lost in Reception. I wanted to make a comment about the fact that failure modes in distributed systems were different in character (rather than merely in degree) from those of simpler systems and to refer for a similar example to Wollrath et al’s A Note on Distributed Computing which, itself, I had trouble finding as I couldn’t recall the title, author or even approximate date of publication, only that it was referred to as inspiration for part of the design of Java’s RMI (forcing programmers to deal with the different failure modes that distributed systems introduce instead of pretending that they don’t exist). While searching the references sections of likely looking papers, I actually spotted the reference to Obliq a few minutes before I found Wollrath’s paper.
• Having a good graphic (logo, cartoon, whatever, …) for an important project may in fact be rather important. What I first recalled was the cartoon, I was not able to recall the rest of the details at all, however the recollection did predispose me to recognising the description when I happened to see it and knowing that I had found what I was seeking as soon as I saw the cartoon. Brand-obsessed marketers will no doubt enjoy the support for their position.
• In the late 1980s I sat the UNSW School Mathematics Competition. The rule then, as now, was “open-book, any resource”. I note that they’ve now explicitly excluded “computers with internet connections” but not computers without internet connections. At the time, the use of computers in the classroom was pretty minimal, but after delivering many thousands of newspapers I had managed to purchase a portable computer and printer and, with no particular plan, took them into the exam room. As it happened, there was a question that required calculus – which I had not studied at that time – so I devised a solution using progressive approximation (which I also had not studied) which made use of the computer’s ability to make thousands of attempts around a promising-looking bound in a couple of minutes. I enclosed a print-out of the source code and output in my answer book. My result suggests that the examiners were suitably impressed. My point in adding this particular anecdote is that how we view human ability is a matter of perspective. In one sense I went into the competition ignorant of calculus, in another I went in with a particular computer-dependent/enabled problem solving ability that I was able to bring to bear in a completely unfamiliar realm and still solve the problem that was posed. Fortunately this exam was of the [rare] type where problem-solving ability is what was being assessed, rather than recall and application of what a teacher had taught. I suggest that a similar approach is relevant when assessing Google’s impact on human ability; from the standpoint of educational norms just a decade ago, the [alleged] loss of recall ability represents some sort of loss of ability; from the standpoint of human ability, the changes that are taking place represent a nett increase in ability.
• The concept of “exam” requires a lack of communication with others, which leaves the UNSW competition having to shut out the very capability that I’m extolling. This remains a difficult problem that I’ll leave for another day other than to mention that UNSW Computer Science has long sought to avoid exactly that situation by structuring their students’ assessments in a very different way.
• The ready availability of so much information does represent a challenge to focus, that now more than ever it requires a conscious choice to focus on one thing for a period of time (and increasingly, to shut out electronically-delivered interruptions while doing so). This, again, looks to me like simply part of a change, not a “good” or “bad” thing in its own right.
• In some sense, this concern is an echo of George Bernard Shaw’s “he … thinks that the customs of his tribe … are the laws of nature”. It’s an easy mistake to make.

NSW parliament lockout

Sadly, I rarely find time to pay much attention to politics back home, but every now and then I read a a few articles from the Sydney Morning Herald’s RSS feed and find off-the-wall stuff like this.

I half-understand why Tony Kelly did what he did, but it seems a little like the behaviour of a sore loser. Was there really _no_ worthwhile business with bi-partisan support for the House to conduct?

Interesting numbers on carbon reserves

Monbiot offers numbers that point out that much of the current carbon-related politicking is problematic not because it’s based on incorrect data, but because it’s irrelevant (shuffling deckchairs).

I don’t agree with his conclusion (even if the UK did cut its reductions by 10% next year – and thereby gain moral authority to tell others that they should do likewise – it seems hopelessly naive to assume that China would be willing to do so, or that India would be able to do so), but the numbers paint a pretty bleak picture.

Stephen Shapiro on Edison’s 700 non-failures

Stephen Shapiro makes the point that the tendency by R&D organisations to puppet Edison’s famous claim are essentially justifying their own mistakes.

This is not to say that R&D organisations should be expected to get everything right in the first place – this is plainly absurd – but that developers who get to useful solutions with fewer mistakes will often do better than those who make more. There is, of course, no way to measure this.

Stephen has been banging the drum about crowd-sourcing and, in this post, makes the point that crowd-sourcing will save an organisation a large part of this cost. There is, of course, more to it. I added this comment to his post:

You’ve missed another important consequence of having large numbers of people [able to] look at a problem (other than just “how much it costs you”).

The odds that the each of the people best able to fix all of the different problems that come up in a project (soonest, least effort, most correct result, most efficient result, ….) are present in a single organisation are vanishingly small. This is the basis of Torvalds’ “with enough eyeballs, all bugs are shallow” mentality and the immense lead that open-source projects have over most closed-source ones on robustness and security.

There are, of course, other qualitative aspects of what gets delivered that matter (performance/efficiency, fit-and-finish) that are often served very poorly by this approach.

I suspect that similar benefits and limits apply to crowd-sourcing in other domains.

The “other” solution to ethical problems

Genetic Ethics