Two recent but fairly typical interactions lately serve to remind me just how stupid computers really still are.

Every time I upload pictures to Facebook, it can’t tell the difference between a photograph of a person that I know and an image from advertising, artwork, street art, or otherwise, and asks me to indicate which of my friends are in these pictures. Face recognition is gee-whiz and has some utility in this context, but if they can’t make it work properly (e.g,. no false positives) why is it being used? Why is it okay that I have to work around the computer’s mistakes?

LinkedIn runs over with suggestions for other people I might to connect to. Yesterday it encouraged me to connect with someone who recently passed away (and whose passing was widely noted). As with the previous example, our immediate reaction is “Well, how should the computer know that?” And that’s my point: we’ve become so used to working within the constraints of technology that we just shrug off this annoying, potentially offensive interactions.

I’m interested both in the limits of the technologies we use and in our enabling of those limits.

• [from steve_portigal] Did My Brother Invent E-Mail With Tom Van Vleck? [NYtimes.com] – [Errol Morris' series is a fascinating. personal history of computer technology.] Batch processing…was like taking your clothes to the laundromat. You’d take your job in, and leave it in the input bins. The staff people would prerecord it onto these magnetic tapes, they would be run by the computer. The output would be printed. This cycle would take at best, several hours. It was maddening, because when you’re working on a complicated program, you can make a trivial slip-up ­ you left out a comma or something ­ and the program would crash. You would try very hard to be careful, but you didn’t always make it…A process that could take a week, weeks, months. People began to advocate a different tactic called time-sharing. Have people at typewriter-like terminals. It certainly seemed feasible. But no manufacturer knew how to do it. And the vendors were not terribly interested, because it was like suggesting to an automobile manufacturer that they go into the airplane business.

• Rest in Peas: The Unrecognized Death of Speech Recognition [robertfortner] – Progress in conversational speech recognition accuracy has clearly halted and we have abandoned further frontal assaults. The research arm of the Pentagon, DARPA, declared victory and withdrew. Many decades ago, DARPA funded the basic research behind both the Internet and today’s mouse-and-menus computer interface. More recently, the agency financed investigations into conversational speech recognition but shifted priorities and money after accuracy plateaued. Microsoft Research persisted longer in its pursuit of a seeing, talking computer. But that vision became increasingly spectral, and today none of the Speech Technology group’s projects aspire to push speech recognition to human levels.
  [Speech recognition comes up all the time in user research. It represents some idealized version of "easy-to-use" though people typically recognize the demanding social norms that talking-to-tech evokes and reject the ideal they moments ago requested /SP]
  (via kicker)
• As Seen on TV – a tribute to doing it wrong [YouTube] – A collection of supposed Epic Fail moments from our daily lives recreated by TV commercials as a precursor to the solution being offered /SP

• Medieval helpdesk – How do I use this thing called a "book"? This video has been around for a while, but it's some good Friday fun. (Thanks, Julie!)

Although the principal conceit of Apple’s latest Mac vs. PC ad is, as always, “PCs suck,” the ad does a nice job pointing to the difference between innovative thinking and the mere creation of features.

While the cupholder suit that appears at the ad’s end is presented as a joke, many companies do have an unfortunate habit of burdening their products with clunky, grafted-on features as they try to push their ideas into new territory.

Compare the cupholder suit to Apple’s breakaway MagSafe cord, which the ad references. While there’s some debate over how well the Magsafe cord actually does what it’s supposed to, it at least intends to address a real issue that computer manufacturers had previously ignored (people’s cords get tripped on, yanked out).

Discovering that aspect of the user experience – however Apple may have done this – and recognizing it as one worthy of design intervention is the real innovation here.

• PC Makers Abandoning a Sales Pitch Built on Complex Specs – Ludicrous article claiming that all of a sudden, computers will be sold based on what they can accomplish, with a focus on aesthetics and design, and with models aimed at specific customer segments. How is any of this at all new? Much of this has been going on – to varying degrees of success – for well over a decade. And there's no indication that these "new" efforts will be any more well executed than they have in the past.

30 must-have PC skills

Here’s the top 10, just to make the point

1. Move and copy files
2. Navigate using keyboard shortcuts
3. Use shortcuts in Word
4. Install and remove new hardware
5. Send image files as attachments
6. Search your hard disk
7. Hard disk maintenance (including disk cleanup and defragmenter)
8. System restore and backup
9. Update software online
10. Create desktop shortcuts

How many can you do? To me, this list suggests how daunting a lot of this stuff is. I know several PC users that don’t understand file systems – where files are stored when you download them or save them. To ask them to attach something, say, by finding it somewhere on the vast structure of files and folders, yikes. They’ve managed to use apps without building that level of knowledge. So doing these tasks, I dunno, it’s pretty challenging without getting a sense of the underlying model of the system that these tasks all require.

I’d opt for things like “cleaning up the desktop” “knowing what the start menu is” “choosing a printer” – some real fundamental stuff that you can build on to get to these skills.

But anyway, the fact that the 30 things they picked are actually pretty hard just confirms my sense that computers are still too hard to understand/use.

Update: A reasonable-looking intro/tutorial on Windows XP can be found here.

story

Ticketmaster Canada confirmed Friday it is shutting its only unionized call centre in Toronto, along with centres in Vancouver, Calgary and Red Deer, Alta.

I worked at Ticketmaster’s Toronto office (not sure if it’s the same actual office – probably not) as a Night Run Operator back as an undergrad, I think it was 1987 or so.

I would go in on Friday nights and sit there with the computers in an empty office doing homework or listening to the radio. I would wait until all the box offices had closed – around midnight, and then run a bunch of “scripts” and get printouts and move these enormous cake-box sized disc packs from one VAX drive (about the size of a small dishwasher) to another. I had no idea how any of it worked so when something broke in the scripts I had to start paging the regular IT staff who were always drunk and wouldn’t call back forever.

I don’t know why I took this job; it seemed cool. I knew other people that worked there, both doing what I did, or even answering the phones. I got second-dibs on tickets, before they went on sale, and I think that’s how I got 4th row Rush and 12th row Yes or something like that – at Maple Leaf Gardens.

The drives looked something like this
DELETED
and the discs themselves were
DELETED

You had to rotate that handle a number of times and then pull it straight up. Sealed inside were many platters of magnetic media that gave the whole assembly it’s capacity. They had two systems running that recorded each purchase transaction, and I would run a backup from A->B and something else, every night.

One guy I worked with had figured out how to purchase General Admission tickets – for free – in between one of the backup stages and then would overwrite any record of it during the backup stages. Since there were no seat assignments, it wouldn’t be very detectable – yeah the overall ticket count sold versus number of arriving guests would be off, but it was GA, so who’d notice? I was horrified at the dishonesty, actually.

And I think we had, in the computer room, these DECwriter terminals.
DELETED
They weren’t for regular operations but if something went wrong, that was what you had to use to interact with them – but mostly they spat out many pages of transactions – you could see, on paper, every ticket purchased that day. If you worked on a day when a big show sold out (i.e., Floyd) then you would be there a long time while it generated output for each purchase. Name, address, credit card info, etc. And it could take a long time for the credit transaction to be verified – it didn’t happen in real time, so the transaction could go through but you wouldn’t get your tickets sent out in the mail because it would turn out that your credit was no good. I think they had someone who would call you up and talk to you in that case.