How alarming!

It's confession time. I'm an Apple Fanboy. I've got an iPhone. I've got an iPod Nano. I've got my own personal MacBook. And i've also inherited a MacBook Pro for work. I am an Apple Fanboy through and through. If I could justify the expense of an iPad, then I would.

So, given my love for their products, it takes a quite serious violation for me to be annoyed by any of my Apple products. It's usually my fault things go wrong, not theirs. But this morning, my iPhone did not awaken me at 7am with We Are Scientists' After Hours. This is the first time, ever, that my iPhone has not awoken me when expected. Slightly unusual as I have a repeated alarm set, for 7am every day Monday to Friday. So what went wrong?

Well, it starts with the fact that yesterday I took a trip to Newcastle on the train, with walking to the station and a huge delay I listened to music for almost 5 hours yesterday. I was quite bored on the train too, so I was played Bejeweled too. Fun fun fun. Anyways, I was quite heavy on my battery use yesterday. By the time I went to bed, I only had 3% of juice left. My late-night facebooking drained the last of the juice in my battery to nothing. The iPhone turned off. So, I plugged it into charge, turned over and went to sleep expecting that my repeating alarm would wake me up the next morning.

I already know that the phone has to be turned on for the alarm to work, but I expected that plugging my juice-empty phone into a power supply would power it back up and the alarm would go off as expected. My mental model of the situation was that plugging the phone in would turn it back on. I don't think that's unreasonable. But evidently it was wrong. My mental model was not supported, and my alarm didn't go off.

Now, actually, it was quite nice to have a lie in. So, in the words of a loyal twitter follower, my iPhone was actually sufficiently intuitive to know that I needed a lie in and gave me that option. Ah sweet iPhone, how could I ever doubt you?

Disclaimer: I wrote this post a long time ago, but for reasons that currently elude me I didn't post it.

HCI Fail #2 (part 2): Further Web Form Woes

So, following my post last night when I got frustrated with Barclays I am attempting to log in with my online banking from home - but this time, as I am at home, I have my trusty PINsentry card reader (which I didn't have on me yesterday) so I am hopeful that I might be able to get in using a different route.

I wasn't going to post about Barclays again, but this is just such a huge HCI fail that I couldn't resist. I entered my passcode without issue. Then, I was asked to select the 3rd character of my memorable word. Unfortunately, their fancy drop down box only has the letters a-f. This box is not scrollable. Neither can I type the letter I need.

Dropdown box failure.

Now, I suspect the reason this is failing is because I am using Safari on OSX Lion which has been known to cause scrolling issues. However, in my humble opinion a bank as big as Barclays should be able to develop sites that work on all modern browsers.

The good news, is that there is one final option for logging in that does not require my memorable word. So, I have now successfully logged into my account. Woohoo!

However, it has taken 3 attempts and much annoyance to get to this point. I am unimpressed.

I know this is a recent redesign for Barclays, but didn't they think about testing it with real users before it went live? Or perhaps employing a usability consultant to test it?

Barclays - you've had a few freebies from me, but if you would like me to test your site further, I am an HCI expert and my rates are pretty reasonable. Just saying.

HCI Fail #2: Unusual Characters

Those who know me, will know that my surname is double barrelled, and that I use a hyphen to join the two parts of my surname. This is my official, legal name. It's not particularly unusual to have a hyphenate surname, yet still many web forms consider it to be. So much so that their forms do no accept my name as a valid input.

Now, in the dawning age of the internet it was possibly forgivable if web forms could not handle "unusual" characters such as the hyphen (-) or the apostrophe ('), but now... it is not. And I get very frustrated when I receive input validation errors due to my hyphenated surname. So, I thought I would reinvigorate my blog by posting about input validation errors.

Let's name and shame those companies who should know better.

So, let's start with Barclays. I have banked with Barclays for a long time and have successfully used their internet banking in the past. Today however, I tried to log in but they have updated their pages and they need me to reenter some details that I had forgotten. So, I filled in their "forgotten details form".

It asked for my surname, I entered my surname and received this error message:

Error message. My surname isn't in the correct format.

So, that's pretty annoying. Thanks Barclays. Why is my legal surname not in the correct format?

It matches the name you have for me on my debit card. It also matches the name on my banks statements, and the name I used to enter to get me into your online banking.

So, although mildly frustrated, I tried again as I really needed to access my account. However, this time I entered my surname without the hypen, and I received the following:

Error message: The details I entered didn't match their records.

Well, yes. The details I've entered don't match your records, because when I entered details that do match their records they complained about the format of my name.

This is a fatal error of their system. It's very, very frustrating. I cannot complete my task. This means, I can no longer use their internet banking and will have to telephone them, at a cost to me. I'm not going to do this now. So I shall just have to go to the pub instead.

This actually happens to me on a semi-regular basis. So, now whenever it occurs I will name and shame the culprits. I encourage you to join me :-)

UX Myths

Just a quick post to prove i'm still alive and thinking...

I've just come across UX Myths The site provids a demonstrates why some common user experience myths are actually misconceptions. Using the site's own words they "debunk user experience misconceptions" The content is quite brief and probably aimed at non-HCI boffins but it's a nice starting point.

In particular, the "People always use your product the way you imagined" post relates nicely to my previous post about secondary use.

So, check out UX Myths and learn some things.

P.S. I will do a proper update soon... promise :-)

Secondary Uses

I'm going to tell you a little story (poetic license has been used in the story - and for those HCI boffins, yes this is a "scenario of use" and yes, the "persona" Alison is based on my experiences...).

The other week Alison walked to uni in the middle of a thunderstorm. It came on all of a sudden and she was caught out a bit. She did not have a hat or an umbrella and ended up getting very wet. Alison's hair was soaked through and her jeans were soaked too. Alison arrived at uni feeling fairly unhappy and miserable.

Alison wanted to dry off before she taught her 9.15 class, in the old department building she used to dry herself by going to the bathroom and standing under the hand dryers for a few minutes. So, Alison went to the bathroom in the new building and realised that the new design of hand dryers does not easily allow her to dry her hair or trousers (the latter she would only be able to dry if she removed the item, which did not seem appropriate). Alison could not dry herself, and went to teach her class in a horrid mood and set lots of nasty homework for the students. Eventually Alison dried off and felt a bit better, but by the end of the day Alison had started to develop a nasty cold, possibly as a result of standing around wet for a couple of hours.

As stories go I appreciate this wasn't thrilling... but it illustrates the point (as a scenario should!). Although a hand dryer's primary objective (and the user's primary goal) is to dry hands (well.. d'uh) and the new fancy hand dryers do this more effectively, efficiently and ecologically than the old ones (or so I'm told - that's why they were installed!) the secondary objective of allowing the user to dry other things cannot be achieved with the new design.

Is this a problem? Well, for Alison, it was very annoying as she had to wander around with wet trousers and wet hair until it dried naturally. The end of the scenario tells us that Alison developed a nasty cold. Are these two things related? Well, that can't be proved, but I don't think it helped... However, Alison's mood in the class was certainly affected be the product's inability to meet her secondary goal.

But, do the ecological benefits of the new design outweigh her irritation at not being able to dry herself? Quite possibly. It might be the case that the designers of the new system were aware that their new design no longer allowed users to use the product in all the ways that they traditionally did. The designers may well have decided that, on balance, the ecological benefits were sufficiently superior that it didn't matter that people will, occasionally, wander around with wet hair and wet trousers. Or perhaps they didn't actually fully consider the requirements/objectives of the existing system and overlooked this element of functionality.

Are secondary goals important? Well, that's entirely context driven and depends upon what the user expects. As it happens, Alison did not pay for the hand dryers (well not directly) and was not involved in the design. Alison will not stop using the hand dryers to dry her hands because of it, and the ecological benefits (probably leading to economic benefits) to the company that paid for the hand dryers will probably outweigh the fact that Alison got annoyed, so I suspect this secondary goal (and the fact that the new design does not fulfil it) will not affect sales of the product in the future.

BUT, some products will have unintended secondary uses that do impact on sales volume if the user cannot see a way of achieving this functionality. When designing products, with a user focus (as all products should be designed!) it is imperative to think about the user's secondary or alternative goals, particularly if the product will be used as a replacement to an existing technology, otherwise not supporting this activity could result in the product being somewhat useless to a large number of potential customers.

Please comment if you can think of any other products that have secondary goals/uses that are no longer supported. I will put up a photo of the new dryers when i get a mo...

HCI Fail #1: Am I a human? Apparently not...

A regular slot in this blog will be for those annoying HCI failures; the things that should work, but just don't!

So, I have been asked to sign up for google wave... I tried... however, the "verification code" check to see that I was a human and not a machine (or computer-oid!) kept failing.

Now, I know that I am not particularly good at getting those stupid, annoying visualisation codes (otherwise known as captchas) correct so I tried typing in the different verification codes at least 5 times before I realised that it wasn't working. I really hate these captchas.. they are so annoying! In the example shown "gonfi" did not work, has anyone got any suggestions on what it should be?

(As an aside: it's also rrrrrrrrrrrrrreally annoying when systems refresh because you have input some information that it wants to check - i.e. the captcha, and then tells you that you have not entered all the required information. You are correct google wave, I did leave the First Name and Last Name fields blank... this was because I had not got that far before you decided I was a computer-oid!)

Even if we ignoring the fact that even though I entered the code correctly the site didn't accept that I really am a human, I still take issue with how some of these verification processes work. It is not always easy to read these codes, even real humans get them wrong and it is VERY irritating to be repeatedly asked to input codes (which change!) because you get it almost right... but not quite. (I am finding it very hard to avoid referring to Pinocchio and saying "but I am a real boy!" in this post...)

The same visual verification code process is used in many different web applications. The best one I have found is on Ticketmaster. Congrats go out to them. HCI winners! I have purchased a number of tickets from Ticketmaster recently and their method of checking verification codes is vastly superior to Google's or any other that I have come across. 

Humanoid attempt at entering code: Close but incorrect. Attempt accepted.

So, on Ticketmaster they use captchas too, but they only require you to be close to getting it correct - not absolutely right. This is clever because the verification codes are actually pretty hard to read! (See the example beside where I think the code wants me to write "comerany chifving" but I intentionally put "camerany chifling". Ticketmaster accepted that I am a human trying to legitimately purchase gig tickets, not a machine trying to automatically purchase on behalf of a ticket tout or something).

To verify that Ticketmaster's check does sometimes work and "correctly" identify those pescy computer-oids I tried pretending to be a computer and input a stupid verification code. The code was vastly dissimilar to the required version. The attempt failed. I received a nice error message explaining what I had done wrong. Not useful if the attempt really was made my a computer-oid, but as I am a real boy I did somewhat appreciate it.

Humanoid pretending to be a computer-oid attempt at entering code. Completely incorrect. Attempt failed.

Whilst I have not conducted any vigorous research into this process, I can say that I have purchased a fair few tickets recently using Ticketmaster and I am yet to be accused of being a computer-oid (apart from on the occasion above where it was intentional!).

Congratulations Ticketmaster on being sensible. But Google Wave? I'm afraid you get to be known as my first HCI failure!

Numerical Nuisance

For anyone who might be interested, a fellow HCI researcher, friend and all-round cool dude, Sarah Wiseman, has started blogging about her research: Numerical Nuisance

Sarah's research focuses on number entry errors and why these can be very, very bad! This is particularly true in the medical domain where HCI fails can (and have) actually be responsible for people dying. I don't expect her blog to be full of doom and gloom though, so take a read... it's pretty fascinating stuff!