Archive for October, 2010

Spelling for the new millennium

Tolerance to spelling errors changes as history progresses. For instance, in the middle ages nobody worried about spelling at all; I’ve read many a manuscript from six centuries ago (my wife is a historian researching that period) and the spelling of everything, even names of people and locations, is all over the place. As long as you could guess what is being referred to, nobody cared. The more precise attitudes of the 20th century would not tolerate this, so our spelling has become standardized, enabling us to play Scrabble and hold spelling bees.

But the technology we use dictates our attitude to proper spelling. A single misspelled letter in a name on a flight ticket can doom one to being kicked off an airplane; and search, at least before Google, would be useless unless you spelled your terms right. What’s more, spell checkers depend on an intimate understanding of our tendency to misspell; which is why a spell correction algorithm for typed text – as incorporated in a word processor, for instance – is quite different from one used in handwriting recognition; the mistakes in the one follow completely different patterns from the other.

And now I was made aware of a completely new aspect of spelling intolerance. I was emailed driving directions to a meeting, and the sender made sure to point out how you spell the street name (it was a slightly unexpected version of the name). She also explained why it mattered: in case I wanted to type the street address into a GPS. Using a map, or asking for directions, this would not matter at all; but a GPS would require the accurate spelling.

Good thing they didn’t have GPS in the 15h century!

Keep your hands on the wheel!

The silly, if cheerful, pop song from the fifties, “Seven little girls“,  gives us the chorus:

All together now, one, two, three / Keep your mind on your driving / Keep your hands on the wheel / Keep your snoopy eyes on the road ahead / We’re having fun, sitting in the backseat / Kissing and a hugging with Fred!

A somewhat improbable notion, considering that there were seven girls (plus Fred) in the back seat; but it has an important lesson: the driver should keep his mind on the driving, his eyes on the road, and – most obvious – his hands on the wheel!

And yet this is far from obvious in today’s info-starved, comm-crazed culture. Today’s drivers may not be distracted by amorous exploits in the back seat; after all, most of us drive alone to work. But we have plenty of distraction from our mobile devices.

In the beginning this meant cellphones, and many countries hurried to legislate the mandatory use of a hands-free device to at least keep the driver’s hands on the wheel (pay heed, gentle reader; the use of a cellphone while driving is a serious risk factor for traffic accidents).

But then the problem shifted to a far more dangerous use model: typing and driving. Much of this involves Texting; Prof. Paul Atchley of the University of Kansas spoke about this in the Information Overload Awareness Day event last week and his research shows that 95% of young adults text and drive; and they often text with both hands on the device. This is scary, especially given that  young drivers already have a high accident risk. Many states have enacted laws against this too.

So now I hear of an even scarier culprit: email. I was lecturing on Info Overload when a man in the audience told us that he has a friend that “from the length of his replies I can tell how fast he’s driving”. To our amazement he added that this friend once included in a message an apology for the brevity of his reply, expressly attributing it to the fact that he’s driving.

Don’t. Do. That!

Join us at the Information Overload Awareness Day event!

October 20 is Information Overload Awareness Day, and we’re holding an online event at 11 AM EDT / 3 PM GMT. Attendance is free if you pledge not to multitask during the event!

Basex, who organize this event every year, secured an impressive lineup of academics, analysts and industry practitioners who will speak about IO and what they’re doing about it. Yours truly will speak too, as president of the Information Overload Research Group.

Register to attend at http://bit.ly/dxpWGN (use code IORGGuest).

A good definition of Multitasking

I was lecturing about Information Overload and multitasking recently, and told my audience how the research data shows that trying to multitask makes you less efficient at each of the tasks you try to do in parallel. After the lecture, one attendee came up to me and gave me a lovely definition she had for Multitasking:

Multitasking is a way to screw up a number of different
things at once.

I just had to share this gem with you!


Data Glut: it isn’t only email…

I was reading an article about hi-tech airships in IEEE Spectrum when my eye caught in the sidebar a link to another article titled  The UAV Data Glut. What do you know – we thought Infoglut was a human problem, and now Unmanned Aerial Vehicles bitch about it too?

Naahh… of course, it isn’t the UAVs that complain; it is humans, the only species that can. The problem, according to the article, is that the super sophisticated drone planes generate more data than humans can look at: “In 2009 alone, the U.S. Air Force shot 24 years’ worth of video over Iraq and Afghanistan using spy drones. The trouble is, there aren’t enough human eyes to watch it all.” And it’s getting worse: the next model of Reaper drone will record 10 video feeds at once!

Unmanned Aerial Vehicle

Image: Wikimedia Commons

So it isn’t only email that’s spiraling beyond our ability to read it all; potentially important military information is also doing it. And the analogy doesn’t end there: the two widely different cases both involve nuggets of value – an important email, or the signs of a camouflaged enemy vehicle – buried in tons of irrelevant data. In both cases, technology is enabling the arrival of ever more data without a commensurate growth in human ability to absorb it all…

The solution, the Spectrum article shares, will have to be relegating the analysis to computers (who else?…)  There are efforts underway to develop software that can watch the UAV’s boring data streams and identify those few needles in the haystack. Which makes you think – what about us knowledge workers and our flooded Inboxes? Sure, there are some tools that can help – Google Priority Inbox is one of the latest additions – but I suspect the military will get more powerful stuff going, and we’ll have to patiently wait for the civilian market spinoffs that always come later.

Who knows, maybe in a decade we’ll have software so smart that it will be able to write our outgoing mails, read the incoming mail, screen the junk and file the rest – while we take off to read a good book? :-)

The importance of Desktop Search

A manager recently described to me his system for handling his incoming email, which he viewed as quite inadequate. He would go through his voluminous new mail each day, then move it all to one folder. At least he wasn’t keeping it in the Inbox like the “I’ve got 6,000 messages in my Inbox” crowd; but his problem was that when he’d need to find a message again he often couldn’t.

Some people solve the problem by maintaining a carefully defined folder hierarchy to archive old messages; for others, this just doesn’t match the way they work. But even if you do have a good filing system, after a while there are just too many messages, and you can forget who sent you the one you’re looking for, or where it would fit. The fact is that knowledge workers spend 15% of their work time on searching for information (according to a Basex study made in 2008). Another fact is that they don’t always find the stuff, even if they know with certainty that it’s on their hard disk, within the clutter of documents, files and mail messages they’d stored there themselves.

This is a serious problem. Knowledge work is often an exercise in associative thinking: you vaguely recollect that someone had once sent you a message with an attachment that had a nugget of valuable information for your current task; you find it, and it points you to another source, or suggests another search… whereas if you can’t find the first item you may miss a key point. And nothing is more maddening than knowing the data is in there, and having no idea how to get to it – like the manager I was talking to.

Enter Desktop Search tools. This is the class of software tools that do for your hard disk what Google does for the web. Not surprisingly, Google makes one of them (Google Desktop); there are many others, some new, some old. The first one, though, is long gone: I remember fondly Alta Vista Personal, from the folks that gave us the first really good web search engine. Windows 7 also claims to have such capability, but so far I haven’t found it very effective; I prefer to get a tool from someone who specializes in the tool’s domain.

My favorite for some years now has been X1 Search, a truly powerful tool that combines full text Boolean search with a built-in preview for most any kind of file. This tool can find in one second any appearance of a search expression anywhere on my disk – in files, mails, contacts, attachments, files within zip archives… and while I do keep a rough hierarchy of subject folders, I don’t even bother to use it; searching is faster and more certain. With this tool, I have total mastery over my stored information; without it I’d be totally handicapped in my work. If you’re into any kind of serious knowledge work, you must have this, or a similar, tool; check out what’s available here. It will cut down a chunk of those 15% of wasted time…