The Pebble and the Avalanche

"Speech in the User Interface" Published

Mon, 08 Feb 2010 08:20:00 -0500

Bill Meisel informs me that Speech in the User Iterface: Lessons from Experience has been published and will appear on Amazon shortly. The book consists of a compilation of essays from industry experts on, well, speech in the user interface.

I have an essay in this book, and I'm looking forward to reading everyone else's.

Of Quail Eggs and Twitter

Wed, 03 Feb 2010 08:46:00 -0500

Does everyone in the world need to know that I had quail eggs for lunch on Monday? I think so, so I'll broadcast that on Twitter in a few minutes. But sometimes the most surprising people read Twitter.

Early Tuesday morning I mentioned on Twitter that I was not going to vote in the Illinois primary election. Although I almost always vote in the primaries (and I never miss a general election), this year I was too thoroughly disgusted with the choices on tap and decided to protest by staying at home.

And then I got a telephone call from a reporter at The Chicago Tribune, who interviewed me about the reasons for my refusal to vote. The reporter had noticed my comment on Twitter and was curious enough to ask me.

I do admit that I'm curious about what tool he uses to scan Twitter so effectively, but that's not the point. The point is that the most innocuous short remark can provide information to someone, perhaps not very efficiently but in real time. Twitter has accomplished something profound and I think we don't quite understand the implications just yet.

The Next Blog Iteration: Can I Have My Data, Please?

Tue, 02 Feb 2010 08:59:00 -0500

Comments are turned off on this blog because I had trouble with the software that runs this blog; the base software is rather old and quirky. I hope to have time this week to repair it.

I am giving due consideration to new software, but there's a spot of bother. A modern blog is really a database: the web page you see is constructed from a lot of data in a database. In fact many web sites are buil that way &mdsah; they're really databases and the web pages you see are constructed from that database. (Trust me: Amazon doesn't create a new web page from scratch for each of the millions of books it has on file, it creates web pages automatically from a database of all its books.) That's actually quite nice in lots of ways, but not so nice in another, namely that if you don't understand how to extract your files from the database you're at the mercy of the web site software. Migrating to a new database system is no easy task.

My current web site is fairly straightforward and stores all its input in ordinary files, and the blog software simply does formatting for me. I don't feel at the mercy of the software and my inforamtion is always available. But I'm afraid it's time to overcome my fears and place all the data into a real database — and make certain I can pull the data out one day by using an open-source version of the software.

"Is this still you?"

Fri, 29 Jan 2010 09:23:00 -0500

A friend just sent an email message to my email address to ask "Is this still you?" I added in a few more details about how to find me:

A microblogging service
Two different IM networks
Two social networking sites (the ones I use most often)
Three of my blogs (again, the ones I update most often)

The question "is this you?" is more complicated than ever before.

Cloud Communications Summit at ITExpo

Thu, 14 Jan 2010 08:33:00 -0500

I see that I've managed to neglect something important: Cloud Communications Summit at ITExpo. I'll be speaking at 10 AM about "Communications Enabled Business Process Fundamentals." And since the Summit is in Miami Beach, Florida, you'll get to look out the window at the warm sunshine while shivering in the air conditioning.

The Cloud Communications Summit is the brainchild of Thomas Howe, who knows this stuff backwards and forwards. There will be an eBook of short essays on the topic published when the conference is over; I'll have an essay (tentatively) entitled "Consistent Communications: Lessons from Russian Military Intelligence."

Toyota's and Western Union's User Interfaces

Tue, 12 Jan 2010 06:24:00 -0500

I had to wire money on behalf of a friend the other day, which gave me the opportunity to try Toyota Motor Credit Corporation's user interfaces as well as that of Western Union. For Toyota, the less said the better; about half the time their telephone system refused to accept calls at all and instead served up an announcement about technical difficulties. When I did get through, they were unable to clearly explain how to send money via Western Union's online service.

Western Union was more interesting. I tried to send money in person, but my debit card would not work at my local currency exchange — I can only theorize why, since I've never tried to use it before. The paper-based, in-person version of sending money seemed to be straightforward enough and cost $13.

I next tried to send money via the online version of Western Union. This was a bit more confusing because Western Union's web site uses different names for the same service in different places on the web site. In the end, I managed to find the correct method to send money — and even so the web site claimed, in the end, that I needed to check the Toyota account number. The price of the service was $44, which ordinarily would make no sense (since when is Internet-based service more expensive than human service?) but probably reflects the increased risk of fraud over the Internet.

Finally, I tried Western Union's telephone service in the hopes that I'd be able to straighten things out with human intervention. The automated telephone service makes every user interface mistake in the book: the system asks me to confirm credit card numbers instead of automating the check of the number's validity (which the web site does with ease). The system uses very poor speech recognition — after using it, I can finally understand why some people tell me they hate speech recognition. When a human came on the phone to complete the transaction, he cheerfully accepted the same account number that the online service had rejected. The cost of the service was $49, the most expensive of all; again, I attribute this to a premium based on fraud experience.

The entire experience left me with a bad taste in my mouth. Western Union, of course, cannot be held responsible for Toyota's inability to explain how to use Western Union, but I have to wonder if Toyota should be held responsible for Western Union's inability to figure out Toyota's account numbers. These disaggregated services should work together; otherwise they're rather useless.

Patent Troll

Tue, 22 Dec 2009 15:51:00 -0500

A patent troll continues to prevail in its lawsuit against Microsoft. Given that the underlying patent at issue is absurd and that the troll company has (in my opinion) done nothing to advance technology, this lawsuit illustrates once again how far the patent system has spun out of control.

One possible benefit: maybe Microsoft will support a rationalization of our patent laws.

Ford and the Car Computer

Mon, 21 Dec 2009 08:26:00 -0500

Some time ago — in the pre-iPhone era, no less — I was asked on two separate occasions for an idea for an automotive product. In both cases my response was the same: an iPhone-style computer for cars, one that can accept clever applications from anywhere. Neither company took my advice.

I see that Ford has implemented this idea with an in-car computer. I can't quite determine if the computer is in production from the news articles, and certainly it does not seem to have the world on fire just yet. (In a remarkably annoying blunder that also keeps search engines away, Ford's home page is completely dependent on Flash, which means I won't bother accessing the Ford web site; I'm left to wonder how software developers will be enticed into working with Ford.) I personally don't hold out high hopes for this computer because it depends on a Microsoft operating system: to my mind this means lack of security, bloated software, bugs, and annoying applications.

Still, I admit that it's nice to be vindicated, and I look forward to the next release of this style of computer, perhaps something that runs on Linus or the Android operating system.

More on Compensation

Thu, 10 Dec 2009 09:10:00 -0500

The US government's "pay czar," Kenneth Feinberg, will impose pay caps on hundreds of financial-sector employees, according to recent newspaper reports.

Mr. Feinberg made no effort to conceal the explicity political motives behind this initiative. For example, the numbers he uses are plucked out of thin air. Why a $500,000 cap? Why not a $400,000, or $600,000, or $502,409.95? Why does he want to expand the caps to more employees — was the previous round of pay caps somehow ineffective, and if so, how? Not only has Mr. Feinberg failed to answer these questions, he's failed to ask these questions, because these pay limits are a political exercise.

The strength of the US financial system stems from its independence from government interference. The Obama administration continues to chip away at the disaggregation between government policy and business activities, which argues that the Obama administration does not know or does not care about the consequential inevitable damage to business innnovation.

Pop Quiz in the Grocery Store

Mon, 07 Dec 2009 10:02:00 -0500

I've just read a nice quote from Craig Herbert, the CEO of Supervalu, a conglomeration of grocery store chains. In an interview in the Wall Street Journal, he said something I absolutely agree with:

"Look at three-for-$5 deals. Even if you're good at math, you have to stop and think for at least a few minutes and say, 'Oh, wait a minute, how much is that per?' We don't want you to have to bring a calculator to our stores."

I've said the same thing many times myself: "A phone call is no time for a pop quiz." You can see some of my comments on proper user interface design in this video of my latest talk at SpeechTek in New York.