Stumbling at the Door

Update:

I ordered my G1 about a week ago. Google said it would be available on the 10th. There was no mail delivery yesterday; it arrived in the mail today (12th). Good.

I try to connect. It won't connect. After 10 minutes with customer service we discover it's because Google has failed to activate my data plan. Customer service agrees it's pretty dumb to sell a phone that requires a data plan without one. She thinks someone forgot to hit “Save”.

I'm slightly confused because, a day after I placed the order, I got a text message from Google informing me of the plan change. Anyway.

Incidentally, without the data plan working, you cannot use the phone at all. Period.

So, I have the phone in a completely non-functional state.

I ask whether I can connect using my home WiFi. After a few rounds with the rep, it becomes clear she has no idea whatsoever what the question really means; additional questions about this yield increasingly garbled answers.

Along the way, she lets slip that yesterday, apparently no Android phones were working at all. But she reassures me that at least they're all working again today.

I am not reassured.

The G Stands for Goodbye

Over the past few weeks I've been getting progressively worse connection quality from Google Mail (GMail). Then, as of yesterday morning, about 24 hours ago, it stopped working entirely, giving me a 502 error.

Now, this isn't the first time GMail has gone AWOL. About two years ago, I had GMail disappear for one day, then two, then three...and then it came back for a few hours, then disappeared again for about three more days, for a total of nearly a week. I even wrote to some high-ups I know at Google, but to no avail.

But those were presumably growing pains. This one is a bit harder to take. Especially when the error message says to try again in 30 seconds, but their support site says it's expected to be out until about 6pm Pacific time today—that would be a total of eighteen hours of outage.

Maybe you're stretching a bit too thin, Google.

Stopped by the Velour Rope

For maybe the first time in my life I tried to stand in line for a new product release. And I was thoroughly rebuffed.

T-Mobile's Web site has been dreadfully slow all day. Even their usually clued-in and cheerful customer service staff sound frazzled and not entirely together with it. It must be tough to be popular (but, folks, did you ever consider a little extra provisioning?).

But, I have no phone.

I called to find out why I was being charged USD 299 instead of USD 179 for it. I had been given a hint by JJ: it had to be no less than about 18 months since my last phone upgrade. Well, in fact, I've suffered through this awful RAZR for a few years now...

Oh wait. In April 2007, Kathi and I got new phones. I didn't like mine, so I returned it. I was assured this would reset my upgrade clock.

Well, it didn't. That is, their records show that I bought and returned the phone, and they agree that this resets the clock, and they know I'm eligible for the deeper discount...but they can't set the bit in their system. So my agent suggests I wait until October 22, the date it becomes available to the masses, at which point they can give me the appropriate discount.

In practical terms this makes almost no difference, because the phone doesn't ship until October 22 anyway. Plus, their Web site suggests that the data plan you have to buy takes effect immediately—i.e., you pay for the plan without a phone to use it. Clever.

Because I've never actually engaged in technolust before, I was tempted to plonk down the extra USD 120 anyway. I would have, if I felt charitable towards T-Mobile. But I feel especially uncharitable today because I've lost over half an hour to their site design. This is because there's a point at which, before it confirms your order, it asks you to re-confirm your identity by entering either your SSN or your DOB. I didn't notice the “or”, which is in the typical tiny T-Mobile font size, and entered both. The imbicile who implemented the site saw it fit to reject such users. (Why did this cost me half an hour? Because T-Mobile has both my wife's records and mine on the account, and sometimes wants my information and sometimes hers, so I had to run through every combination...and while I was at it, I also tried out every combination with and without the leading zeroes. On a day when their servers were glacial when they weren't timing out.)

Welcome to the big leagues, T-Mobile. Now get your act together.

How to Climb Like a Champ

I have discovered the secret weapon behind Chris Carmichael's success with climbers.

Chris is the coach behind Lance Armstrong's magnificent performances in the mountains of France. Now that we've had two Tours in a row in which other extraordinary climbers have been thrown out for illegal substances, many minds wonder what Lance was on (besides, as his famous Nike ad said, his bike for six hours a day).

Fortunately, for the right sum, Carmichael will tell us. Let's look at his advertisement (Bicycling magazine, July 2008, page 113):

Gives away nothing, does it? Now look at it again.

First, you'll have to ignore the model on the left who has the posture of a squat toad and the expression of someone who has just swallowed one. The one on the right is the one we are all supposed to aspire to be.

Ignore the geological oddity of this place, where each hill seems to be composed of entirely different substances. If the hills you train on don't look like that, well, that also explains why you aren't winning any Tours de France.

Ignore the extraordinary sharpness of the bleached rocks in the middle distance. Why did they use a photograph where rocks, not bicyclists, were in focus? No doubt because the Carmichael-trained cyclists ride so fast, no camera can capture their movement.

Focus, instead, on what's between the “dancing on his pedals” rider's legs. No, no, not like that! Here's the detail:

It's extraordinary. Where you would expect to see the background (of rock and grass in unearthly focus), you see...the fragment of a yellow oval with the letters “MIC” in the upper half, looking exactly the same as Carmichael's logo. And just a bit lower is what appears to be a third wheel for the bicycle, with a tire of clearly different type, hovering in the air, as if ready to drop like landing gear on demand. And if you look further down this montage (not—it is now clear—ever to be confused with a montagne), to the left of Carmichael's corporate logo you'll see a profusion of chains and gears and drivetrains and other instruments of S-and-M. (And the typo in the URL—trrainright.com—is just a bit of icing.)

When Lance won at Limoges, he said he rode with “the strength of two men”. Now, for a small fee, you too can have your second man.

Or, maybe it's possible that the secret to Carmichael's success is something else entirely.

It's Photoshop.

NNTR / NN2R

Mitch Wand pointed me to this article by an email veteran bemoaning what has become of the medium. Obviously, I concur.

One annoying type of message that continues to throw me off my game is those brief “okay” and “thank you” messages, especially from staff. I've been getting increasingly good at anticipating these and affixing NNTR or NN2R to my preceding message. Amazingly, both the former and the latter are indexed in Web dictionaries, which makes me wonder why people don't use them more. [That's “No Need To Reply”, natch, not a Myers-Briggs indicator, though it might as well be.]

Anyway, I do, and I encourage you to as well.

Meanwhile, you ask, what do you do? I'm afraid I still send those replies to staff members, because I think it's just basic etiquette (though many a misplaced good intention has been born there...). I'd like to ask them someday whether they'd be offended if I stopped doing this and instead used that old classic from the days of Usenet: TIA (that's “Thanks in Advance”, for you wee ones).

A Totally New Concept for Learning Devices

The book.

That's the latest “totally new concept for learning devices” from Nicholas Negroponte.

To be fair, it's not just a book. It's a battery-powered book. That is new indeed, because a regular book won't stop working in broad daylight, but an OLPC XO2 can.

The OLPC Kool-Aid-addled salesmen continue to trot out the fuzzy math concept that developing countries spend USD 20/year per student on textbooks. I've heard this before from their staff and demanded documentation, and none was forthcoming. I'm not asking just for the sake of it; I'm asking because I absolutely, totally refuse believe that number, for more reasons than I can count.

Notice that Negroponte no longer mentions India in his list of developing countries (besides, there's no room for two `I's in his BRIC...), now that the country wisely turfed his snake-oil out.

One Lesson Per Century

I'm not usually into the +1 thing, and I've been doing the utmost to hold my tongue on One Laptop Per Child, but Ivan Krstić gets it so right that I can't help but ask you to read his post.

My Slow Email Movement

I've never been a slow-anything person, other than riding slowly up mountains, and that's only because of the weakness of flesh, not any unwillingness of spirit. So it's not often that I embrace a Slow movement.

But on my Web page, I currently say

Every month I get over 10,000 messages. Of these, just over 8000 messages are spam. In this same time I send over 1000 messages. I am, in short, a full time email employee who gets to do a little teaching and research on the side. You know, as recreation.

If any of the deans or assistant deans or vice-deans or sub-deans or deans-in-waiting or deans-in-law at Brown are reading this: I'm kidding! Everyone else: I'm not!

For a day or two, I played with Google Mail on my mobile phone. Then, one day, I was lost during a bike ride, so I pulled out my mobile to find my whereabouts on Google Maps...and found myself checking my email. Soon after the apps ceased to work on my T-Mobile phone, and I was happy to not investigate why.

For the past year or so, I have rarely been checking my email when I travel. That is, I check it once every two to three days. And here's something amazing. If I wait a day, it takes me about an hour to restore my mailbox. If I wait two days, it takes me about an hour-and-a-half. If I wait three days, it still takes me about an hour-and-a-half.

These numbers are slightly misleading. They mask critical tasks that require real attention to detail and will take much longer than a minute to discharge. But those tasks are relatively few: I can be gone for two weeks and find only two or three such tasks lying in wait when I return. Which suggests I'm significantly promoting in importance things I do encounter daily.

There are other knock-on effects. You've played email ping-pong, right? Everyone treats their mailbox as a task-manager, so you get a task, you reply or forward to put the monkey on someone else's back, they do the same to put it back on yours, and suddenly you've lost an hour of the day (because studies haved, shown that these context-switches are extremely expensive, though as computer scientists, we should have known that). And, since you and your correspondent are both on-line, your reply begets their reply, and so forth. Congestion-control through exponential-backoff, anyone? (This is why I enjoy clearing out backlog during times when lots of people are on vacation: significantly fewer replies.)

The backoff strategy shows where our email user interfaces have gotten it wrong. They show us when we received email, but who cares; they should instead tell us when we should be replying to email. And that “when” should be a combination of when we need to (based on message content) and when it would be prudent to (based on correspondent habits).

So I'm making a conscious decision. I'm going to go Slow on E-Mail. I'm going to treat it as an addiction, like drinking too much coffee. There doesn't seem to be a simple, prescriptive or descriptive classification of addiction treatment akin to the seven stages of grief; much of the on-line material about treating addiction is rather disturbing and possibly borderline dangerous (and, mark me, there's a megachurch out there somewhere that is going to make a killing off faith-based treatments for email). So I'll have to figure this one out on my own.

Do feel free to drop me a note with your thoughts.

Just don't expect me to respond.

Resolving Unanticipated What, Again?

My homies Ducasse, Wuyts, Bergel and Nierstrasz have a paper at the upcoming (2007) OOPSLA entitled User-Changeable Visibility: Resolving Unanticipated Name Clashes in Traits. Their abstract, as listed on the OOPSLA Web site:

A trait is a unit of behaviour that can be composed with other traits and used by classes. Traits are an alternative to multiple inheritance. Conflict resolution of traits, while flexible, does not completely handle accidental method name conflicts: if a trait with method m is composed with another trait defining a different method m then resolving the conflict Mayo prove delicate or infeasible in certain cases. In this paper we present freezeable traits that provide an expressive composition mechanism to support unanticipated method composition conflicts. Our solution introduces private trait methods and lets the class composer change method visibility at composition time (from private to public and vice versa), something which is not possible in mainstream languages. Two class composers Mayo use different composition policies for the same traits. [...]

Thanks to Microsoft Word, perhaps? Or the Web-publishing software?

Where's (Candidate) Waldo?

A few enterprising young folks, including (full disclosure) a graduate student at Brown CS, have created a map mashup to track the US election candidates, Map the Candidates. The time-travel feature is quite interesting, and will become more so as the primaries approach.

The site has much more potential. Every press pundit takes pride in predicting who will become a candidate one or two years hence (of course, we only remember their hits, not their misses). By tracking news articles a site like this could perform similar forecasts, and take the bloviators largely out of the mix.