There are a whole slew of unpleasant things that can happen to you from using computers:  neck problems, back pain, carpal tunnel syndrome, repetitive strain injuries, etc.  These things are too mundane to be really frightening, but occur with disturbing frequency.  I don’t want to sound alarmist here:  fortunately, the ergonomics of computer use have been well studied over the decades.  If you follow the published advice in setting up your work area and take on healthy habits (posture, typing technique, etc.) then you can dramatically lower your risk.

At a desktop, I think it’s much easier to pay minimal attention to adjusting your working environment and still get it mostly right: the screen will be in generally the right place, and its easy to find some combination of chair height and keyboard position that suits.

With laptops, however, it is nearly impossible to work ergonomically using the laptop on its own.  In order to have the keyboard at a comfortable position for your hands, you have to have the screen at a height that is much too low (unless, of course, you are a tyrannosaurus Rex) for your neck.  Simply put, if you work primarily with a laptop computer, and you don’t setup your primary work area with either a separate screen or a separate keyboard, then you are asking for trouble.  The stress we put on our bodies working like this may seem slight, but consider the amount of work you do at a computer day in and day out.  Most of our careers will be dependent on using a computer on a least a regular basis.  Some of us will have careers that we spend almost entirely in front of a computer.  Don’t risk it!

This problem has been so well studied, and the guidelines so easy to find (a google search for “computer ergonomics” will land you dozens of helpful sites.  I’d recommend the CDC, or DoHS) that you’d be foolish not to make a little bit of effort.  A separate keyboard you can plug in while you elevate your laptop to the proper viewing height is a cheap and effective solution if you’re freelancing.  If your laptop is issued by your employer and you live in an industrialized nation, then there will be appropriate health & safety statues that should make them only too eager to give you the equipment you need.

I already know two people (my age!) who have been diagnosed with various problems caused by their computer use.  It was a wakeup call for me, and I’m a desktop man.

The repercussions of the switch to a laptop-style interface remain to be seen in terms of long term health problems for computer users, but its hard to imagine them being anything other than, on the whole, bad news bears.  I don’t mean to sound anti-laptop or anti-computer here (nothing could be further from my sentiment!), but there’s a safe(r) way of computing and a dangerous(er)* one.  I’m optimistic that a little education will go a long way in preventing injuries.

Of course, the laptop is not the only interface paradigm that’s on the rise.  Apple’s iPad seems to be leading the charge into large(ish) scale touchscreen devices entering widespread use, and people typing on various cramped mobile phone keyboards have been common for several years now.  These interfaces need some scientific scrutiny as well, but the typical usage pattern (short activity with plenty of breaks) probably means there’s not too much to worry about.

*yeah, I know.

I used this method on an iPhone 3G S, but I’m confident it also applies to the iPhone 3G, and perhaps even the original iPhone.  The problem was that the earbuds that came with my phone stopped working as they should: I could still hear everything through them—calls, music, games, etc., but the inline microphone and clickable controls (volume, play/pause, next/prev track) stopped responding.  No one could hear me on calls through the buds, and I could no longer control music or get the phone to answer calls.

At first I wondered whether the earbuds had gone bad.  It’s certainly conceivable that the inline module could fail without taking the headphones with it.  I also tried resetting the iPhone in various ways to no avail.  By far the most curious feature of this issue, though, is that when pairing the iPhone with a Bluetooth stereo headset (Jabra HALO), the play/pause button on the headset wouldn’t work either.

What tipped me off to the problem was that the 3.5mm stereo connector for the earbuds no longer seated properly in its jack on the top of the iPhone: there was about half a mm gap between the connecter base and the tallest extent of the iPhone case.  When seated properly it should be flush.  Now, seeing the bottom of the jack is no easy thing, even with some bright lights overhead, but when I looked I was pretty sure that there was some lint crammed into the bottom of the jack.  That’s right, pocket lint was the culprit. How pocket lint could stop a Bluetooth headset from operating correctly is interesting, and probably says interesting things about the design of the iPhone: high-level functions like play/pause being tied to their low-level inputs (headset jack, Bluetooth radio) in some non-trivial way, but I digress.  Removing the lint not only fixed the plug-in earbuds, but also fixed the Bluetooth headset.

Now, a word of caution: sticking things inside electronic connectors is generally not a good idea, and could void your warranty.  But, if you’re like me and the idea of relying on your own fine motor control and senses sounds better than taking a walk (or drive or subway journey) to your nearest Apple Store, then read on.  You need something small and rigid enough to get inside pick out the lint.  My feeling is that those little interdental brushes would be ideal, but I’m a floss man myself and a paperclip was what I had on hand.  It will take a bit of doing, since it’s been compacted against the bottom of the jack by your earbud connector.  A little scraping and blowing out with compressed air and I was in business again.

Now, if only Apple would allow Bluetooth headsets to do next/prev track functions by implementing the full AVRCP profile.

1. I don’t want it to impact the bandwidth I have when I’m using my connection.
2. I don’t want them to use it in such a way that it triggers my ISP to throttle my connection.
3. I don’t want to be held responsible for what people do on my connection.

Basically, what I’m saying is that I’d like to be able to share without it negatively impacting me.  I feel like a lot of people would also share their own connections if it were easy to do and didn’t have negative consequences.  I also feel that it should be entirely possible to do this!

In (1), what I really mean is that my use of the connection should get priority over whatever random strangers I’m sharing with.  Maybe it’s my neighbors, who don’t have a connection of their own, or maybe it’s just someone passing through, looking for a few minutes of WiFi to check their email.  There’s plenty of time during the day when I’m not using my connection at all, and plenty of time when I’m just doing some light surfing.

What I want is a WiFi router that supports this kind of prioritizing.  Set up and broadcast two different SSIDs: one for priority traffic, one which just gets whatever bandwidth is left over.  The priority channel gets normal encryption and access control, the shared channel is unencrypted.  Maybe I’d call the shared channel “Spike’s Free WiFi.”  The quality of that connection would fluctuate wildly, depending on whether or not I’m filling up the priority channel, but hey, it’d be free.

For (2), I’ll start off by saying that net connections in the UK have advantages and disadvantages over those in the US.  I think they typically come out slightly cheaper per MBit of connection speed, but most ISPs have annoying “network management policies.”  These include throttling connections if you use push or pull too much data during peak hours, and, I suspect, if you use particular protocols, like those that run P2P services (I’ve noticed severe slowdowns when I put up a bittorrent client or the BBC iPlayer Desktop application, even before I’ve moved enough data to trigger the limits detailed below).  For the most part, ISPs are relatively transparent about their traffic management with respect to peak hour use (see for example Virgin Media’s), but I would like more detail.  Like, the kind of detail that would allow configuration of a router to keep strangers using the connection from getting my connection throttled.  In particular, a clear statement about how different protocols trigger throttling, if any.  If UK ISPs are going to go down the road of advertising “Unlimited” plans, but actually enforcing limits by throttling connections, then they should publish enough detail to allow customers to configure tools to avoid those limits.

(3) is one which needs a legal solution, rather than a technical one.  Basically, what is needed is either legislation or legal precedent that establishes that simply providing internet access doesn’t make a person liable (in both criminal and civil senses) for what is done by others over that connection.  ISPs enjoy this legal protection, as do other providers of communication services—you can’t sue the postal service if they deliver a harassing letter sent to you by someone else.  I know of no cases where a person operating an unencrypted access point has been successfully sued or prosecuted for what others have done on the connection, but there are particularly worrying measures being considered.   At the top of the list are the so called “3 strikes” proposals where being accused of sharing of copyrighted works over a network connection 3 times can get your connection terminated.  I think this is a bad idea for a number of reasons, but it would effectively kill the idea of sharing your network connection with strangers—which would be a real shame.

If you walk down a city block or residential suburb with a WiFi sniffer these days you’ll find dozens of operating access points.  Imagine how much coverage could be achieved if everyone got into sharing the bandwidth they had going spare.  Coverage would be patchy, and speed unreliable, but it would be free and leverage a piece of tech most people will be upgrading over the next few years anyway.  Crowd-sourced municipal WiFi!

Text messages are limited to just 160 characters, which can be encoded into just 140 bytes.  To give you an idea of just how little that is, I compared it to the size of the web-version of the NY Times story linked above.  Just the basic HTML (i.e. no images) is 89,986 bytes (or about 88 KB).  That’s about 642 text messages, which is more than my monthly allowance.  Including images, this figure jumps to 858,333 bytes, or 5,360 text messages, more than I send in several years.  On my iPhone, my monthly plan includes unlimited Internet data, and I can download the NY Times article (over the relatively poky GPRS connection) in about 20 seconds.  Yet my plan only includes 500 text messages: an amount of data that could be transmitted in a second over standard connections.

A space scientist at the University of Leicester calculated that sending text messages cost more per byte than data from the Hubble Space Telescope.  It’s all to create the illusion of scarcity so the carriers can keep charging their exorbitant fees.  I remember seeing signs in India for text messaging at 0.08 Rupees per text, about 2 tenths of a cent.  This means US carriers charge 100 times more for their text messages.

The NY Times article goes into a little more technical detail to explain that text messages are actually packed into what’s called the control channel, used to send instructions back and forth from handset to cell towers.  These channels get used whether there are text messages or not, so an increase in volume adds little to operating costs.  The messages don’t appear on the high-bandwidth channels used to transmit voice, further supporting the conclusion that text message pricing has nothing to do with the actual costs of carrying the data.

Fortunately, as the NY Times article explains, Herb Kohl, the chairman of the Senate antitrust subcommittee has taken the first steps in attempting to get the carriers to account for their behavior, and several lawsuits have been filed accusing the companies of price fixing.  All I have to say is, “about time.”

This has two main consequences:

Graphics programmers will start to explore alternate rendering paradigms like ray-tracing or voxel rendering.  Ars Technica has a great interview with Tim Sweeney of Epic Games on this subject.  Sweeney’s point is that with really programmable and really powerful GPUs, the rendering APIs like Direct3D and OpenGL become a crutch and limit innovation and creativity in creating graphics.  Developers will just write their own renderers using other established or hybrid paradigms.  GPUs become little more than extra processing power.

Secondly, GPU hardware gets general enough to get repurposed for other processor-intensive tasks.  I remember my computer architecture teacher telling me in 2005 about biology researchers using GPUs to accelerate floating-point intensive simulations.  Now I’ve just read, in another Ars Technica piece, about a commercial software package that uses GPUs to accelerate password cracking.

We’ve known since Alan Turing’s time that all information processing hardware is equivalent up to some small overhead.  Building specialized hardware for specialized tasks can yield performance gains, but only for large tasks where this overhead is non-trivial.  The so called “physics accelerators” which are starting to appear on high end gaming machines seem to be doomed to failure for this reason.  Hardware accelerated audio was just a blip on most gamers’ radar 8 years ago.  Why create a dedicated processor for physics/audio/graphics when you could instead just upgrade to a CPU with more cores…

After dealing with the pain that is Vodafone (my previous carrier)’s repair service, I was reluctant to go down that dark path, and just dealt with my limp iPhone for a couple weeks.  Remembering that my 1 year warranty would expire in November, I finally got around to making the call to get some reparations.

Fortunately, as I discovered, Apple’s customer service and repair arm is a class act, even in a country where the bar has been set pretty low.  The first phone call to Apple went well: short wait to talk to a rep with fluent language skills, and a competent sounding manner.  He issued me a repair ID and ordered a returns kit sent to my house.

Now, the story hits a few bumps in getting the returns kit delivered.  Of this I hold no malice against the rep I spoke to; it was a simple misunderstanding.  When he read the address they had on file, I assumed he left out the house number for brevity, but as it turns out, it just wasn’t there.  So UPS received an address with the correct street, but no house number.  As helpful as I have found the order tracking website in the past, once there was a problem the cryptic “delivery exception” messages were of little use in deciphering what the problem was.  Fast forward about a week, several phone calls to UPS, and several mornings wasted waiting up for the deliveryman who seemed to need a signature to deliver me what is essentially an empty box; I called Apple to have them deliver to my lab instead.  I had my kit the next day.  This time didn’t bother me too much, since the iPhone still worked OK, but if I had a dead phone, I’d have been pissed at UPS (and probably Apple for hiring them).

Opening up the returns box was almost like unboxing a new gadget from Apple.  Carefully laid out were all the things I’d need to ship my iPhone to Apple, right down to an included paper clip to open the SIM card tray on the iPhone.  Also included was a pre-paid envelope and address label to ship the iPhone to Apple.

Once it shipped, I could use my repair ID to track the status, but it turns out I needn’t have worried. Apple sent me an email once my iPhone arrived, and, 3 hours later, after they shipped a replacement.  That impressed me—3 hours after my gimped iPhone arrived at their repair center a replacement left destined for me. Unfortunately, this was on Friday afternoon, and the replacement didn’t arrive until Monday.  Since they picked up the entire tab including shipping both ways, I couldn’t begrudge them for not splurging on Saturday delivery.

The difference between this and Vodafone’s service, which they charge £7 per month for, is like night and day.  I knew where my iPhone was at any time via the web, even during transit using UPS tracking.  When Vodafone repaired my phone I gave it to a man in the store and just waited until the predetermined pickup day.  Several times (yes, I had several repairs) I returned on the appointed day to be simply told that it hadn’t returned yet and that I should come back tomorrow.  Apple took 3 business days (5 including the weekend) to return my iPhone, while Vodafone typically took a week.  Apple proactively informed me about the progress, while Vodafone didn’t even let me know when there was a delay.

Good customer service includes the tenet that a customer’s problem is your problem until it’s resolved, and it includes keeping the customer in the loop.  Good customer service is something Apple UK has, and Vodafone doesn’t.

“Unboxing” Photos of the Return Kit below:

OUCS announced today that iPhone will be an officially supported platform, at least for getting WiFi access, with the possibility of them building some ‘local applications’ (which I interpret to mean native apps).  The site claims that they’ll release instructions for accessing eduroam WiFi access points shortly, which will add iPhone to the list of mobile platforms that already includes Windows Mobile 5 as well as the Nokia N95, 770 and N800.

However, eduroam isn’t as widely available as the older OWL-VPN system, which is available in most libraries and commom rooms.  In order to use this service, you had to install a Cisco VPN client which authenticated your connection and allowed you to access the internet.  Unfortunately, mobile users (not just iPhone) were left out in the cold because there was no Cisco VPN client for mobile platforms, hence the switch to eduroam, which doesn’t require the VPN.  However, the new iPhone 2.0 software includes support for Cisco VPNs, meaning that OWL-VPN is now available for iPhone!  You can use the instructions below to configure your iPhone to access the OWL-VPN for places where eduroam is not yet available.

1. Register for a remote access account.  If you’re accessing OWL-VPN from a laptop, you’re already good to go.
2. View this configuration document (you’ll need to sign into WebAuth to view it), and make note of the IPSec secret.
3. On your iPhone, connect to an OWL-VPN wireless access point.  This will appear without the ‘lock’ symbol, but when you connect to the VPN, you’ll be secure.
4. Next, on your iPhone, select Settings -> General -> Network -> VPN -> Add VPN Configuration.
5. Choose IPSec at the top, and enter the following:
   Description:  <whatever you want it to appear as, I used ‘OUCS’>
   Server: vpn.ox.ac.uk
   Account: <your remote access username, i.e. abcd3456, from step 1>
   Password: <your password from step 1>
   Use Certificate: <should be grayed out, leave it alone>
   Group Name: oxford
   Secret: <the secret from the document in step 2>
6. Tap Save
7. On the VPN control at the top, tap to turn it ON.  If all goes well you should see a ‘VPN’ symbol next to the WiFi signal indicator at the top of your iPhone.  Surf away!

I haven’t tried to get onto an eduroam access point yet, but as soon as I do, I’ll post an update with any special instructions.

Update: I’ve been able to connect to eduroam (outside the Earth Sciences building, if anyone cares).  Oxford types will need to get a Remote Access account, then connect to the ‘eduroam’ access point from their iPhone.  It should ask you for the username and password (your remote access password, not your WebAuth/Email password).  You’ll be prompted to accept a certificate, which should be issued by GTE Cybertrust Global Root.  It came up ‘not verified’ on my iPhone, but I connected anyway and was able to start surfing away.

Update 2: After my triumphant first day of success with iPhone and OWL-VPN, I’ve been subsequently unable to surf successfully.  It connects to the WiFi and VPN without complaint, but seems unable to get any data.  Anyone else have this trouble?  I’ll have to investigate this further when I’m back in Oxford.

After a bit of a struggle, I pulled down and installed the iPhone 2.0 software last night and added some apps.  There are some good ones, which I’m sure you’ll hear described all over the intertoobs if you’re so inclined.  My favorite so far is called Midomi.  You fire it up, then sing or hum at your iPhone and it tries to figure out what song you’re after, then gives you links to the iTunes store to buy it.  What floored me is how incredibly good it seems to be at figuring out what you’re singing.  I did a 12 second off-key rendidtion of the opening line of “Don’t Stop Believing,” and in seconds it came back with the correct result.  This’ll be great for those moments when you hear a song you recognize on the radio but can’t for the life of you remember who it’s by.

There should have been many more settings, not just kill and stun.  Kill, stun, limp: that’s the next one down, isn’t it?  …or maybe on “bit of a cough” setting, even lower than that.

Some devices already have a remotely enabled kill switch, such as corporate Blackberries with remote wipe cabability (intended to protect sensitive company data should it be lost or stolen), and others will soon follow, like reports that OnStar is adding the ability to remotely stop the engine of a connected car (again, marketed as an anti-theft system).

Microsoft, however, is looking to set its phasers on stun, limp, or even “bit of a cough.”  They’ve filed a patent application for something they call Device Manners Policies (DMP), another Minitrue-style name and acronym, which, like Digital Rights Management is less about manners (or rights) and more about restrictions.  Schneier calls it Selective Device Jamming.  Essentially, under this scheme, locations will be outfitted with hardware to broadcast to your devices the rules of the land, such as “vibrate only” for cell phones, or “no photography” for cameras.  Hospitals or airplanes where critical equipment can be subject to interference from wireless devces would be able to force your devices into sleep mode until you leave the area (how will such wireless transmissions be guaranteed not to cause interference themselves?).

Microsoft wants to draw analogies with the societal guidelines we call “manners,” i.e. that it’s considered rude to talk on your cell phone in the movie theatre.  However, this is a false analogy since manners are guidelines, not rules.  DMP wants to disable functionality in your electronics (albeit temporarily) without your consent, or force them into sleep mode: limp and stun settings.

No, an actual manners technology is only a short step away from the “location-based services” stuff that all the cool kids were talking about 2 years ago–some of which are already out.  See, once your devices know where they are, you can do digital manners all client-side, without having to contact the Borg Cube to get your orders.  You have a couple different profiles, such as “theatre” which might mean switching to silent, “office” which sets ring volume to low, and “street” which sets it to high so you can hear it above the sounds of the city.  Simple, no external restrictions, and the user still stays in control.  Each person is free to choose to obey social guidelines or not: just like real manners.

Photo by Ted Sali
Creative Commons Licensed

Strong Authentication, or multi-factor authentication is considered by researchers to be significantly more secure than using a single factor.  A factor in this case is something that identifies a person, and factors are usually classified into 1) things a person knows, like a password or PIN, 2) things a person has such as a bankcard or keyfob built for this purpose, and 3) things a person is or does, like a retinal scan or fingerprint.  So online banking, which only required the user to enter in username/password combo relied on a single factor, whereas the ATM uses strong authentication since the user is required to have their bankcard and know their pin.

So NatWest (and I guess other RBS banks?) are sending that ATM-style authentication home to users by sending each a small calculator-like card reader for use with their bankcard.  It works pretty much exactly like the card readers in the grocery store, except that they give you a code online to enter in reader, and then the reader gives you a code to enter online.  I, for one am pleasantly geeked-out to use it, and glad to see that NatWest is taking the security of online banking seriously by putting so much money and effort into getting it out to users.

I don’t think they’ve done a particularly good job so far of allaying people’s concerns.  Lots of comments on blog posts are bemoaning the fact that they’ll have to carry the damn thing around with them—no, you wont.  You’ll only need it to make a payment to someone online if you’ve never made a payment to them before.  The readers are also entirely identical, meaning you can borrow your cubemate’s reader if he has his at work and you keep yours at home.  But I’ve heard rumors that the big reason NatWest is beefing up security is because they’ll be cutting down the delay between making a payment and the recipient getting credited.  It’s now about 3 working days, and apparently the plan is to make it happen in seconds.  If true, that’s a really important new feature that NatWest could use as a way to introduce the readers:  “We’re working faster to process your payments, but also means that we need to increase security.”

Finally, since the reader is just a standard thing (even readers from other banks will apparently work), I’m hoping either they’ll release software that works with laptop smartcard readers, or someone hacks it together.  The security is in the microchip on the card, so putting the reader as widely available as possible shouldn’t undermine the system.