Designed in 2 Minutes?

Tuesday, December 21, 2004

Contactability

One of my friends, the ever-enthusiastic Andrew Oakley of Middlesex University, was recently chatting with me on Skype (http://www.skype.com), the free internet phone service which is taking the world by storm.

We'd been discussing the Skype API which allows third-parties to create software which 'hooks' into Skype in order to provide extra functionality, such as Business Card export or a voicemail program which we'd just discovered and were testing.

Andrew made the following rather amusing observation:

We invent telephones to talk to people.
We invent answer machines to catch em if we're out.
Then we invent mobile phones so we can catch people if they're out.
Then we invent voicemail incase there's no signal.
Then we invent Skype incase they're on the net.
Then we invent Skype voicemail just incase you still can't get them!!!!

It's quite a good description of how we strive to be ever more 'contactable' in our new digital world. Contrast this with some remote parts of Africa where they don't even use timekeeping devices, and certainly no mobile phones, so it's common to be waiting for someone to turn up for a meeting for most of the morning.

But without the pressure we Capitalists suffer of making every second profitable and attempting to be contactable at any given moment of the day, it's not something that matters so much to these people in the grand scheme of things.

Maybe all of us should learn from this, take a break and just disappear somewhere remote, away from all digital technology and the concept of timekeeping, and just enjoy life as it comes?! :-)

Friday, December 10, 2004

"So simple a child can use it..."

A new product from Norway recently hit the shelves and it has impressed a lot of people, including myself.

It focusses on the shortcomings that Microsoft's Windows is certainly not so simple that a child can use it (safely). I've caught my four year old sister attempting to delete the C:\Windows directory before...

From a child's point of view the average computer is far too complex for their needs. My sister only wants to play games, paint pictures, and send emails to me when I'm away at university. She gets baffled by having to load different programs to do different tasks, and having to remember which buttons to press to do something like write an email. She does, however, have a good grasp of the filing systems concept meaning she can reliably locate a file up to 3 or 4 directories deep as long as she remembers where she last put it.

She has no idea how to deal with things like spam or dodgy websites (not that I'm particularly worried about her trying to order a shipment of V!@gr@ from a dodgy Nigerian lawyer) and why should she? The concept of someone trying to sell her something in an email - a text-only medium, not a physical shop - is rather alien to her. But all the same it's best that these things are filtered out before she gets to read them.

Easybits' Magic Desktop is described as "a protective shell which sits on top of Microsoft Windows" and provides a very friendly interface for children to use to interact with the computer. All extraneous data is stripped away leaving a fun desktop with a very simple method of operation.

Programs (or 'tasks' as they appear to the user) are loaded with a single point and click. Double-clicking never made much sense to me anyway, but we've all learned to live with it! If a program is executed multiple times (ie. the user clicking repeatedly on the icon until the program appears) the system realises this and still only loads one copy of the program to save the user from being suddenly inundated with 33 email compose windows.

The email program (which doesn't appear as a separate program to the user, it's all integrated) works on a very visual premise. Click on the "Talking Parrot" to record a message (primarily aimed at under-5s I'd guess, unless my sister is unusual in being able to type, albeit slowly and patiently, at her age) and then click on a photo of the recipient to send it.

Administrators (ie. parents) have the ability to bypass the Magic Desktop shell and configure the environment from Microsoft Windows. Emails are only accepted on a whitelist basis (anything not from an allowed sender is rejected) and websites are similarly whitelisted.

Unnecessary features, such as the ability to blind-carbon-copy the mail to a distribution list whilst requesting return receipts, are stripped out completely as they're likely to be totally meaningless to a child, leaving only the core essential features of the system. Audio compression and attaching to the email are all handled in the background and the user has to make just two button clicks to successfully email grandma or big brother.

There's nothing more irritating to me than typing myprog --help in a terminal and having three pages of meaningless switches and commands scroll past. And don't get me started on Microsoft Word's forays into desktop publishing, web design, email composing, and line drawing (all areas it totally sucks at)... a word processor should surely be a word processor and nothing more.

If only programs for 'grown up' systems could be so well-focussed on a single task with a simple, no-clutter interface!

Digital Divide set to stay

A recent report has shown that currently 50% of Britons are without any form of internet access, and even as far away as 2025 40% of us still won't see the point in getting online at home.

If this is true, it threatens our society in quite a deep way. By 2025 (indeed, long before) the Government wants a lot of our infrastructure to be run via the Internet. (No doubt the telcos are quite keen on this idea too). Education will assume students have Internet access for research and, increasingly, everyday admin such as submitting work from home or keeping in touch with tutors.

The NHS would like people to make use of online systems to get health information rather than taking up valuable GP time.

Employers are more and more often requesting job applications and CVs to be emailed in to save on admin costs, and increasingly work is done from home rather than enduring the stressful, time-consuming, costly commute that so many people currently put up with daily.

Since I got connected at the age of 13 in 1998 I've noticed more and more of my everyday life coming to rely on Internet access. Coursework is submitted by email, course notes are all to be found online, communications with distant family and friends is achieved solely through instant messaging applications and Voice over IP systems. In the 6 short years since I first dialled up at 28kbps the world has changed phenomenally and yet many people are still left back in the 'pre-historic' world before the digital revoolution took off. I would now never consider anything less for my home than always-on 0.5Mbps Internet access as so much of my life depends on being able to communicate digitally with the rest of the world.

But with horror stories every day in the media of worms and viruses, hacker attacks and spyware, spam and the new menace, phishing - it's no surprise that a lot of less technically-confident people would rather steer clear of such a potential minefield.

Perhaps we need to take a step back and ensure our society is inclusive to all regardless of whether or not they have a broadband connection, rather than telling everyone they have to have a computer and Internet access to keep up with 'the modern world'.

Big Brother is Watching You

According to BBC News people are much more likely to allow their private lives to be scrutinised by a computer system if they can associate that system with a real person.

People are naturally a little wary of Orwellian style 24-hour surveillance in their homes (I certainly wouldn't allow it) yet technology makers are keen to have systems which do exactly this.

However instead of reporting back to Big Brother these systems are supposed to analyse details such as when you normally get up to make breakfast, or when you might usually be out of the house. This might allow the system to switch lights and heating on in advance of when you wake up, or cycle lights in the house during weekend breaks to give the impression that the house is occupied.

Sounds pretty cool, but don't forget that your every move is being watched and recorded on camera... I bet you'll think twice before dancing around the kitchen with the mop next time.

One of the researchers into this type of system, Dr Richard Bowden, noted: "When we put the surveillance cameras in our centre, a lot of people were very unhappy about the fact that there was a system watching them".

Richard and his team have created an avatar (virtual face) which sits on a screen where the user can observe it. To add to the personality of the system, they even called it Jeremiah and refer to the system as "him". Jeremiah is interactive to a degree, and can react (by changing his facial patterns) to visual stimuli using camera systems, so he can smile when a person appears to be playing or skipping around, or appear sad if you leave the room, for example. Think Holly from Red Dwarf, only with a significantly smaller IQ and not quite as senile.

"But when Jeremiah's camera went in, nobody minded, because although it's still watching them, they could see what it was watching." said Dr Bowden.

Which just goes to show, if any would-be police state dictators are planning on rising up in the near future, pay close attention to HCI issues with your monitoring systems. It just might make the difference between peaceful world domination and bloody revolution.

Library swipe card readers

In the Main Library, where I work as a Computing Help Assistant, there are barcode readers in the entrance lobby to ensure authorised access by University ID card. Every student or staff member has to swipe in and the systems has been designed to operate as quickly as possible. Here's how it should work:


  • The person enters the building and queues at one of the three entry gates.
  • When reaching the gate they hold their card an inch above the reader panel (to save having to put it physically on the panel then trying to pick it up again).
  • Providing the card is valid and can be read, the gates swiftly open to allow the person to continue to walk through.


If you get it right (and there's no queue) you can approach the gates at walking pace, extend your arm with the card in hand, and be through the other side of the gates without having to slow down. It's that fast. I love it.

However the system falls down at one simple point - a significant number of users don't seem to understand how the reader panel works.

The panel's laser beam is deliberately focussed to a plane about an inch above the glass of the panel. This, as mentioned above, is so that users don't have to place the card on the panel then spend valuable seconds attempting to pick it up off the flat surface using their fingernails.

But most peoples' prior exposure to barcode readers will have been in supermarkets where the checkout operator usually (traditionally) places the object directly on the barcode reader panel. So they assume they should do the same when entering the Library with their ID card.

The card reader is deliberately long-sighted, meaning that users who place their cards on the reader panel then get frustrated that their card appears not to be readable, and they try rubbing the card up and down the panel as if it was a box of cornflakes at the checkout. After a significant time they may either realise their mistake or someone else who's already realised what's going on may point out to them that they need to hold the card above the reader.

So, as usual, Library staff have had to resort to hand-made notices to help users to get around design flaws or limitations in the sytem (witness the Learning Centre front doors which have had no end of problems and have at least three notices explaining how to enter the building!)

A bit more research into user tendencies before deploying new technologies is usually worth the time and money it takes!

Monday, December 06, 2004

Elite

Introduction
Elite is a fantastic retro game which still holds my interest and is very playable, even 20 years after it was first released in the mid-1980s for the BBC Micro. Conversions have been made to many platforms and one of the most highly-regarded conversion was the one to the Acorn Archimedes running RISC OS. The RISC OS version packs a whole load more playability compared to the original BBC version, as well as giving solid full-colour graphics rather than the original wireframe design.

One of the key features of the RISC OS version is symmetry - you are not the centre of the universe (as in the BBC version, with everyone constantly trying to kill you) but rather you can see pirates and police engaging in their own battles, and Bushmasters mining deep into asteroids as you fly past.

My RISC OS-powered Iyonix PC is still able to run the game and it often takes up spare moments between lectures or lengthy periods at night...

Gameplay
The idea is simple, and may even seem tedious in the days of multi-million-dollar multi-mission blockbuster games, until you remember that this game was the first of its kind and introduced a revolutionary new style of gaming which is still being played today.

Put simply, you play the character of a lone trader in a world several centuries away where private spaceflight is available to all. It is up to you what you do with the rest of the game; it's very much open-ended (there is no end to the game!) and you choose what you do from the moment you launch from the space station orbiting around the planet Lave, the Santaari Galaxy's spaceflight training centre.

Trading
You can try to make an honest living by trading in simple goods, ferrying them back and forth between nearby planetary systems, but you will soon need to venture into the more dangerous systems governed in feudal or even anarchy states. In these systems, pirate activity is rife and you need to have a well-equipped ship to defend yourself.

The game is so well thought-out that this simple method of play on its own is more than satisfying. But because of the open-ended design of the game where all actions are up to you, a whole range of career options - none actually specified in the game - is open to you. You can deal in more contraband goods such as slaves, narcotics, Ulganian slug furs, or firearms but be wary of the police who will make it hard for you to approach a planetary system once you have a criminal record.

Mining
You can buy mining lasers and follow the Bushmasters around between asteroid fields, mining the lumps of rock and selling the minerals to the nearest industrial planet.

Bounty-hunting
Or you can play vigilante and become a bounty hunter - going solo or following a police squadron around to find and destroy the pirates that group together at the edges of every system, perhaps even collceting their strewn cargo to sell at the nearest space station.

Piracy
You can also go down the highly exciting route of becoming a pirate yourself - either through dealing in too many contraband goods and obtaining fugitive status, or by ignoring your conscience and blasting apart innocent traders in order to steal their cargo. But beware if you follow this career path - the police (who shoot first and ask questions later), gangs of rival pirates, bounty hunters, and scared traders will all fight back and you'll never be more than 5 minutes from another firefight.

Of course, you can try to clean up your act - but this can be very hard to do with police and traders constantly shooting at you. Killing innocents is seen as Very Naughty and will get you deeper into trouble. The only way to get out of trouble is to kill lots of pirates (this will get you favour with the police) and run away fast when the police themselves arrive so you don't have to fight your way out and get intro deeper trouble again.

Upgrading your ship
Ships can be upgraded with meaningful upgrades - not just bigger and bigger weapons as you may find in more modern games. The principle of Elite is that you attain a kill rating by simply being good enough. Buying 1000 heat seeking cluster missiles hardly shows a great deal of pilot skill, so the game encourages the pilot to develop their flight and combat skills in order to attain the status of Elite simply by surviving and being good at it.

It is this principle which makes the game so endearing as it's all about developing personal skill rather than making lots of money (it's perfectly possible to become Elite with a totally un-upgraded ship, just very difficult!).

Other upgrades are those which enable you to carry more cargo, recharge your shields faster, dock into space stations automatically using a computer, fire lasers from the side and rear parts of your ship, destroy enemy (and friendly) missiles, jump through wormholes to one of seven other galaxies, and collect solar wind streams to convert energy into free fuel (this also enables you to collect floating cargo cannisters and other debris). In some conversions of the game, such as Elite-A, you can also purchase and fly many of the different types of ships mentioned in the game (In Elite you're limited to just your Cobra MkIII which is arguably one of the better ships anyway).

Revolutions
In terms of revolutionary design, the game is one of the most important ever. It was the first to introduce a real-time 3D vector graphics engine to a home computer game, a model which is still followed today. One of the most instantly-recognisable features of Elite is its scanner. For a 3D real-time game its designers had to come up with a new way of describing the precise location of any nearby ship or object as 2D scanner displays used in most games at the time weren't up to the job.

screenshot of the game

The scanner David Braben and Ian Bell designed for their game consisted of an oval representing the space around the player's ship (actually a circle squashed by perspective, as you're viewing it from above and behind) where each object is represented by a green club. Imagine the club sticking up or down from the scanner, beginning at its tail. The very end of the tail of the club shows the x-y location of the object compared to the player, and the head of the club represents the height in the z dimension. Clearly a longer club indicates that the object is further up or down in space compared to the viewer than an object with a short club.

In the screenshot there are six ships besides my own represented on the scanner. As all the clubs have heads below the tails, the ships are all below mine in space. The exception is the 'dot' to the right of the central white dot in the scanner, which is a ship directly to my right, at the same altitude as mine. The blue Krait-class ship disappearing off the bottom of my screen is represented by the stubby blue club to the top-left of my scanner. Again the club head is lower than the tail, indicating that the ship is (just) below me.

Missions
Elite does contain a handful of prescribed missions, but in keeping with the open-ended design of the game they aren't forced on you in any way. One day, when docking after a routine spaceflight, you may receive an urgent message when you enter the space station. These missions do depend on a few factors - you having enough kills, being in the correct galaxy, etc. Missions range from dropping a supply capsule from low altitude onto a planet, to tracking down and hunting a merciless pirate and his gang of support fighters. Missions are definitely not part of the main fabric of the game though. You will have needed to play for many hours before being offered the first mission, and after they're completed there are no more. The main point of the game is still to explore and survive.

Miscellaneous information
Novella
The marketers of the original BBC Micro version of Elite took an unusual and bold strategy for introducing new players to the game. Instead of the usual "you are captain of a fleet..." two-page spiel they commissioned someone to write a short novel which was supplied with the game pack. The novella drops the player straight into a well thought-out universe full of amazing and exotic characters, all of which are superficial to the game, but the end effect is to immerse the player in this fantasy universe and give them some purpose of exploration as they play. It's since inspired many more novellas to be written by fans, and even a musical!

Types of craft
They also provided a full size wall chart of many of the different types of ships that can be spotted, with some elementary statistics and (arbitrary) history of its design. Some ships were deliberately left off this chart as a surprise for the hapless player, and rumours were placed in the Pilot's Handbook of gigantic dredger-ships the size of a city, and generation ships from many decades previously. No further mention is made of these ships and 20 years on no one has conclusively proved that they do or don't exist. My uncle claims to have seen a dredger in the outer reaches of Xeaqu and he's not alone in his claims. I've never seen one of these myself and many people have scoured the game's code in an attempt to find any references to the objects, with no apparent success.

Yet many variations of the original Elite were produced and distributed, and it's possible that there is still a rare variant out there which does contain the craft!

All of this adds such an amazing richness to the game that it's no wonder it has lasted 20 years and still has a strong cult following.

Links:
Simon Challands' Archimedes Elite pages
Ian Bell - one of the original authors
My own ArcElite tribute site

Tuesday, November 30, 2004

In floppies we trust

I work on the computing helpdesk in the Learning Centre out-of-hours and one of the most common, but heartbreaking, pleas for help come from students who have put too much trust in their floppy discs (seriously, some of them leave the helpdesk crying).

"Hi, I'm wondering if you can help me. I've spent the last 72 hours without sleep or food doing my dissertation and saving it only to this flimsy piece of magnetised plastic, and when I took it home with me I accidentally dropped it in a puddle of acid and stamped on it repeatedly before my pet goat chewed it to pieces. Is there any way I can get my work back?".

Well that's not quite how it goes, but the sentiment is real enough. Why oh why do students save all their work on floppy discs and take no backups? Especially considering there's a perfectly good 10MB of securely backed-up, campus-wide online storage under the name 'My Documents'.

Aside from poor education in why floppies started going out of fashion around a decade ago, I think I have some idea of why students so often don't make use of the secure online storage, preferring to save everything to floppy instead (or USB flash memory stick, where the dangers of corruption are lower but risk of loss or theft is higher).

I think it's all a matter of trust. Students can envisage their document being phsyically written to the disc, they can pick it up and hold it, they can carry it around, they can put it under their pillow at night (not a good idea as the dust will soon kill the disc surface). It's tangible. It's real.

Online storage is nameless, faceless, abstract... how can a student understand that there is a rather large computer a few hundred metres away which now stores their files when they can't see it or touch it or take it home to put under their pillow? To them clicking the 'save' button seems like it could mean the last they ever see of their precious work, as there's no comforting flashing light or whirring 'burrr-dzzz' sound as the file is written, and often no concept that 'My Documents' is not local to only that machine, but is available from any computer they're logged into.

People don't trust things they can't see or touch (or kick) which is why there is still such a huge reliance on physical storage media they can carry around with them. Sure, these devices are great for backups. But the trouble is, too many people still use them as primary storage.

Strangely these people have no problem allowing their employer to put numbers representing virtual money into a bank account which they can't see, when there's no physical cash involved... I suppose people have come to learn to trust in the banking system. If only they could learn to trust in online storage too.

Sunday, November 28, 2004

Free wireless access for all. Is it desirable?

The city of Philadelphia in the USA has announced ambitious plans to cover the entire of the city with Wi-Fi wireless network access.

The rationale is that "for a city to succeed in the future, it must be a digital city", according to Dianah Neff, the city's Chief Information Officer. This presents quite a stark contrast to what many people in other big cities would consider as what their city needs to be successful. Surely good economic stability, low crime rates, and a first-class transport system rate higher? Or perhaps the Philadelphians believe that universal network access for all will help to facilitate this utopia.

Whatever the reason, it has to be said that free Wi-Fi access for an entire city is quite high on the 'cool things for a city to have' list, along with giant ferris wheels and enormous multi-million-pound marquees, and I can see that it would bring benefits to business and home users living in the area.

Communications companies in the city, on the other hand, aren't so happy. They're worried about losing potential revenue and are currently threatening to block the entire project from going ahead. The project planners argue that the project will be a benefit to all, and will help to bridge the digital divide as residents can get access to the internet using just a standard computer and Wi-Fi card, without any need to set up costly broadband lines with telcos. Give your great-aunt a laptop and web browser, and she can sit at her desk and keep in touch without having to sign any contracts or plug in any cables.

But my concern is that, as other rich cities (this is a $10m project which will cost a further $1.5m/yr to maintain!) follow suit, poorer cities which can't justify this expenditure will be perceived as under-developed, and a new crack will appear in the digital divide.

Friday, November 26, 2004

Nokia voice dialling

When Nokia first introduced voice-activated dialling on their phones they had clearly thought long and hard about how to make the system as usable as possible. In fact they got it so near to perfect that they've left the interface pretty much unchanged since it was first introduced; the latest models still voice-dial in the same way.

The reason the design is so good is that it provides clear user feedback in all the important areas of exactly what the phone is doing, without being over-intrusive, and which can all be controlled from a handsfree headset with just one button.

Although voice dialling can be initiated by holding down (usually) the Names key of the phone, it can also be initiated by pressing the headset button. Doing this also bypasses the keypad lock if it's currently switched on, as the whole point of a hands-free kit is that you don't have to get the phone out to unlock its keypad first... From here on, the system works in the same way whether the headset is attached or not.

The phone beeps to let you know it's waiting for input (essential if the device is in your pocket as the headset has no other way of displaying anything the phone might be trying to tell you). If you just say nothing within a few seconds, it will beep again in a slightly more depressed tone to let you know it's giving up.

If you do say something after the first beep the phone will wait for approximately half a second of silence to indicate that you've finished speaking. If you speak again (or something else makes a noise) within this half second then it will keep listening...

Once the phone is sure you've finished inputting it will process the input and play back the user-inputted recording which it thinks matches closest to what you just said. This will have been prerecorded by the user and 'attached' to an entry in the phone book.

This step is one of the most important as it provides the user with the clearest possible indication of who the phone is now about to call. It uses the user's own recording (which the user must recognise - it's their own voice!) to confirm the action.

If no match was found then the phone just beeps again in its depressed way and goes back to sleep.


All of this makes for a system with a supremely good concept of cognitive processes, which tells the user unobtrusively when to input data, tells the user in the clearest possible way what the output is, and informs the user of a successful or unsuccessful input.


It's just a shame that the voice dialling feature is limited to (usually) only 10 entries from the phone book :-(

How many ways?!

Now that I'm getting into the swing of ranting about Windows' windows, here's a short post regarding cognitive processes.

As you should be aware, having one clear way of performing an action (such as initiating a phone call from a mobile phone) with possibly a backup shortcut action for 'expert' users (eg. speed-dialling keys) is beneficial to the user as it reduces the amount of thought necessary to complete the action. Rather than thinking about which method will be quickest or most convenient to get the job done, the user can just get on and do it.

So why, then, are there so many ways of doing something as supposedly simple as closing a window in Windows?!

Let's count them:


  1. Click on the window's close icon
  2. Press Alt-F4
    (these two actions on their own should be sufficient!)
  3. Locate the window's icon on the taskbar, right-click, and click on 'Close'
  4. Locate the window's icon on the taskbar, right-click, and press Alt-F4
  5. Locate the window's icon on the taskbar, right-click, and press C
  6. Locate the window's icon on the taskbar, right-click, press the up cursor key, and press Return
  7. Click on the window's program icon (top-left of the window) and click on 'Close'
  8. Click on the window's program icon and press C
    note you can't use Alt-F4 on this menu as Alt closes the menu)
  9. Click on the window's program icon, press the up cursor key, and press Return
  10. Press Ctrl-Alt-Del to fire up the task manager, find the program, select it, and click 'End Program'
  11. Right click on the taskbar and fire up the task manager, find the program, select it, and click 'End Program'


Prizes for anyone who can extend my list to 15+ items - I'm sure I've missed at least some ;-)

[7:30pm - the prize goes to Andrew Oakley of Middlesex University who offered the following:]


  1. Press Alt-Space to open the window's control menu and click on 'Close'
  2. Press Alt-Space to open the window's control menu and press C
  3. Press Alt-Space to open the window's control menu, press the up cursor key, and press Return
  4. Select the File menu and click on Close/Exit
  5. Select the File menu and press C or X (depending on application)
  6. Select the File menu and scroll to Close/Exit using the cursor keys, then press Return
  7. Press Alt-F to open the File menu and click on Close/Exit
  8. Press Alt-F to open the File menu and press C or X (depending on application)
  9. Press Alt-F to open the File menu and scroll to Close/Exit using the cursor keys , then press Return
  10. Press Alt and use the cursor keys (across then down) to open the File menu and click on Close/Exit
  11. Press Alt and use the cursor keys to open the File menu and press C or X (depending on application)
  12. Press Alt and use the cursor keys to open the File menu and scroll to Close/Exit using the cursor keys , then press Return


Surely that's an exhaustive list now?

For your next assignment, try to find all the possible ways of saving a file in Windows :-)

Bad window furniture positioning in Windows

Still on the topic of Windows' windows close icons (is anyone else beginning to see why I dislike the use of the generic term Windows as a name for an operating system?!) I often find myself getting frustrated at how close the potentially destructive close icon is to the relatively harmless maximise icon.

One pixel's grace on the close icon's left hand side (Windows XP 'classic' mode) is not enough, Microsoft designers!

Whilst it could be perceived as useful or neat to have all the window furniture grouped into one part of the window's title area, I favour the arrangment of some alternative OS desktops' window furniture.

On RISC OS there is a close icon and a 'back' icon (which sends the window to the back of the window stack - very useful when you come to think about it) on the left hand side of the title bar. Both of these are potentially destructive actions, as you lose the window you were working on, although obviously sending a window to the back of the stack just hides it from view rather than actually deleting it.

On the opposite side of the window's title bar are the iconise (minimise to an icon on the desktop) and toggle size (full height/user-defined window size) buttons. Both of these are non-destructive, non-intrusive actions as they can easily be reversed by clicking on the icon again as they're toggles.

This organisation allows the user to separate dangerous destructive tasks mentally and spatially from less dangerous tasks so there's less chance of an accident happening if a user was reaching for 'maximise' and hit 'close' instead.

Closing windows quickly in Windows

The 'active area' furniture in Windows' windows actually extends beyond the physical size of the button. This leads to a very useful feature of the way the window works.

When a window is maximised, as it often will be when working under MS Windows, the user is able to close a window simply by moving the mouse swiftly to the extreme top and right edges of the screen and then clicking. The button's active area extends to the most extreme pixel on the top-right of the screen when the window is maximised, therefore the user can use the top and right edges of the screen to decelerate the mouse cursor for them rather than having to hunt around for the button.

Thursday, November 25, 2004

The scissors paradox

A friend of mine recently tried to buy some scissors for the start of his university course, having none in the house at the time.

He soon found a pair in the local shop but very unhelpfully the scissors were fixed to the packaging with a plastic cable and, presumably as a safety measure, kept firmly shut.

As he pointed out to me, how are you supposed to open the packaging and get to your nice new scissors when they're firmly attached to the packaging and require a pair of scissors to be removed! Surely you're only buying scissors because you don't have any in the first place!

Thankfully for him the shop assistant was more than willing to lend him the store scissors before he went home so he could remove the packaging. The mind does boggle though.

GPRS on Nokia phones

I have a Nokia 6100 - a good all-round phone with lots of features such as the ability to run Java programs and connect to the internet via GPRS.

GPRS, for those who don't know, stands for General Packet Radio Service and provides a way for wireless devices such as phones to transfer packets of data to and from a base station network. GPRS is analogous to broadband DSL which a home user might use to connect to the Internet, as it's always-on and works over a digital system rather than the older analogue WAP protocol. It's also quite a bit faster than WAP.

Currently most mobile service providers charge per MB for GPRS data access. Prices are extortionate (in the region of ukp4 per MB) and so I don't usually access the internet from my mobile unless I absolutely have to.

But this presents a problem to phone designers. Sometimes I'll have been trying to accses my saved bookmarks, or check some GPRS settings for a friend, for example, when I'll accidentally select a URL to visit. Before I have the chance to cancel, the page has loaded, appeared on my screen, and cost me on my next bill.

The problem is this: how do you confirm that a user really wants to initiate a data transfer when the connection is always-on? Imagine if every time you clicked on a link in your web browser your computer asked "Are you sure?"!

Perhaps you could have some sort of inactivity timeout so that if the user hasn't made a data transfer in over 30 minutes then the phone assumes the current session has finished and will then ask again the next time they try to access data.

However Nokia seemed to pull a cop-out and just not confirm it at all. Much to my phone bill's dismay...