Status-Q

I wrote a few months back about how I was using a GPS logger to keep a record of my movements. Some people think I'm a little eccentric - I think that's the word - for doing so.

But my data-gathering is nothing compared to Stephen Wolfram's. In a splendid Wired article called The Personal Analytics of My Life, he discusses some of the insights he's been able to glean from his own historical records. One inspired idea, which I confess had never occurred to me, is to run a keystroke logger; he's captured everything he's typed for many years. (Now, that's data you wouldn't want to fall into the wrong hands!)

I once thought seriously about capturing, say, once or twice a minute, the image of my screen, which I could then later OCR, search, use to recreate lost documents, etc. But other than helping Sheng Feng Li with a system that did some of this for VNC, I never took it any further. Worth reconsidering, perhaps...

Anyway, many thanks to Richard for pointing me at the Wolfram article, which is worth a read.

I suppose that another way to analyse data about your life is to do the analysis on the fly and record the results there and then. That's called a blog.

An amazingly bright moon tonight but this was still a 50 sec exposure. I used F/22 to get a bit of a starburst effect. Moon pictures are hard 🙂 Got too cold to carry on….

A conversation with friends this week turned to the subject of recipe websites: epicurious.com and suchlike. I realised that I'd never looked at any of these sites. Mmm.

In our household, we believe in specialisation of function, and, despite my offers to help out, Rose has always preferred doing the cooking to surrendering control of the kitchen! We've been married just over twenty years, hence my somewhat embarrassing realisation this week that the reason I'd never seen any recipe websites was actually quite simple...

The last time I looked at a recipe, there was no such thing as a website.

Once, when having dinner at the house of some good friends, I discovered that the other guests, a delightful couple, had a place in New York which they would sometimes rent out to friends and acquaintances. They had a strict rule: they would only let this fine apartment on the Upper West Side to people with whom they had personally had dinner. The rationale for this was simple: they wanted something more than a simple contractual agreement with those who would be occupying their home, and they felt that a certain level of social acquaintance was a good first level of filter, as well as imposing some extra obligations of responsibility on the tenants.

Most people apply similar filters to social networks. At the very least, if you are likely to be reading somebody else's tweets or posts, you don't want to read that which is likely to be tedious or offensive. You're likely to be more forgiving of those who are within your real-life social circle. The concept of a friend, contact or buddy in the online world is open to a wide variety of interpretations, of course, but one network which has traditionally had a clearer definition than others has been LinkedIn.

Linking to someone on LinkedIn has, for me, always implied a little bit more than simple acquaintance. In fact, I think the original site suggested that you should link to people you know and trust, though if that wording is still there, it's much less obvious now. This was, presumably, because others may use the system to ask you for an onward connection to others; a process which is likely to be somewhat awkward if you don't really know them, or don't feel that their acquaintance would be beneficial to your other friends!

So I've tended to have a fairly strict rule that I only link to people with whom I've at least shaken hands, and ideally had some sort of conversation. I've waived the former occasionally for those with whom I've had videoconferences, but in general it's worked well since I joined LinkedIn - gosh! - eight years ago.

But it seems to be going through massive growth recently, and perhaps it's now more of an address book than something that implies any level of recommendation? Is LinkedIn the new Plaxo? At any rate, I'm starting to get more requests for links from people who are just interested in making contact, they're in related fields, they say nice things about stuff I've done, and they seem like people I would like if I did get a chance to shake their hand. So my resolve is slipping. Should I stick to my principles, or am I being very last-millennium to insist on a physical meeting?

Joy Rosen comments on the new NPR Code of Ethics and Practices. Extract:

In my view the most important changes are these passages:

In all our stories, especially matters of controversy, we strive to consider the strongest arguments we can find on all sides, seeking to deliver both nuance and clarity. Our goal is not to please those whom we report on or to produce stories that create the appearance of balance, but to seek the truth.

and....

At all times, we report for our readers and listeners, not our sources. So our primary consideration when presenting the news is that we are fair to the truth. If our sources try to mislead us or put a false spin on the information they give us, we tell our audience. If the balance of evidence in a matter of controversy weighs heavily on one side, we acknowledge it in our reports. We strive to give our audience confidence that all sides have been considered and represented fairly.

With these words, NPR commits itself as an organization to avoid the worst excesses of ""he said, she said"" journalism. It says to itself that a report characterized by false balance is a false report. It introduces a new and potentially powerful concept of fairness: being ""fair to the truth,"" which as we know is not always evenly distributed among the sides in a public dispute.

One of my more interesting acquaintances is Jane Wilson-Howarth: writer, GP, speaker, broadcaster, corporate travel health consultant... Her web site is enjoyable and unusual.

So I'm pleased to hear that she's venturing into the world of blogging. I think this will be entertaining.

Have added it to my Feed My Inbox account...

Twisty Tree by Quentin Stafford-Fraser

Ah - here's an idea, following on from the last post...

I've always wished I could have an email address that cost the sender 1p per message. That would have such a nice effect on spam while not really inconveniencing anyone else, once the system was set up. But here's a refinement of the idea:

Imagine that, when you sent a message, you could choose how much you wanted to pay. And inboxes were sorted, by default, with the most expensive messages at the top. You could override the order, of course, for friends and family, but would this be a good way of prioritising your email? Or am I taking capitalism too far?

It would have different dynamics, of course, depending on whether the money earned went to the recipient or, say, to a charity of their choice. How about that? If you really want to get my attention, it will involve a £1 donation to Oxfam. (Remember, a transatlantic call might have cost you a lot more than that anyway). I'd even be happy to read a lot of your spam at that rate...

The phenomenal success of SMS text messages is a fascinating example of many things - the need for an asynchronous communications mechanism between humans, the surprising adoption of what was originally a test facility for engineers, the merits of enforced brevity in communications, and our voluntary blindness to some costs when they're expressed in a certain way.

Let me explain that last point. Let's say that a text message, once outside your allowance, costs you 7p, and that the average text is maybe 70 characters long, so a character costs you 0.1p. On this basis, a megabyte of data costs £1000. About $1600. (A little more, actually, since SMS characters are only 7-bits). Or, to put it another way: to send a floppy disk's worth of data would probably cost you a lot more than the computer from which you sent it. To send an MP3 track would cost you about the same as a car. And in my case, if I happen to be in the States when I send a text, it costs seven times as much. Seven cars.

Now, you could argue that everybody gets lots of texts in their calling plan, which is of course true, at least in your home country - I never get close to exhausting my allowance - but it's this theoretical underlying price that allows the networks to charge for this as a bonus. Suppose you pay £3 more per month for a plan that gives you 300 texts instead of 100 texts. It looks like a bargain - you're getting those texts for just 1.5p each! That's only £220 or so for a megabyte, or, in music terms, a thousand quid per track.

Now, if you have a smartphone, consider the data portion of your plan. I'm looking at two Vodafone SIM-only schemes here, and the only difference between them is 500MB of data per month. The price difference is £5 - i.e. one penny per megabyte.

This factor of twenty thousand in the two different ways of sending data, over the same network, from the same device, has always amazed me. There are lots of approximations in the above calculations, of course. You could point out that IP-based traffic has lots of overheads, which of course it does, for small amounts of data, but that's mostly because we often see that data wrapped in a web page. I also assumed that people on average only type 10 words per text; if you always used your full 160 characters you'd save a fair amount per byte. So perhaps the true cost factor is more like 5000, or even 1000. But can you think of any other aspects of your life where choosing that alternative wouldn't make you pause for thought?

Now, this only exists at all, of course, because in the past there was no choice. Phones were simple devices without a full IP software stack, they had small keyboards and limited ability to create or display any other kind of media. But once you had phones that could run apps like iMessage or WhatsApp, which could efficiently send messages using protocols of their own, the picture changed.

So it's no surprise that a recent study suggests mobile operators lost $14bn last year because of such apps. It was only a matter of time.