James Mansfield

I hate it when forms like this Oxfam donation form ask me to re-enter or confirm my email address:

Oxfam donation form - email address re-entry fields

I suspected I wasn’t alone so I tweeted to find out what others do when presented with this situation and I have to say I was overwhelmed with the number of responses I received. It seems most people (well I should say most of my online web savvy friends) hate it as well and that they usually copy and paste the email address from the first field into the second. Ctrl A, Ctrl C, Tab, Ctrl V is less keyboard presses than typing an email address and therefore tends to be the preferred approach.

To me the whole idea of re-entering your email address feels like a heavy handed, ill thought out trend that creates more work for the user, doesn’t solve the problem it attempts to, goes against the websites business goals and causes untold pain and wasted time for many *slight exaggeration there perhaps*.

Where do you draw the re-entry line?

The point of re-entering your email address is to make sure you get it right – error prevention, a usability fundamental. So where does one draw the line on something like re-entering information? If I’m ordering something online that is to be delivered to me at home, should I re-enter my home address to make sure that’s correct? Surely having my expensive online purchase delivered to the right address is more important than getting my email address correct?

It’s rude, it’s assuming the worst of the user

Asking me to enter anything again is to assume I’ve gotten it wrong the first time. I’m comfortable suggesting the majority of people prefer to be treated like they got things right. In the unlikely event they got it wrong it may be difficult and/or embarrassing to rectify the problem but that’s life and I suggest most people prefer it that way.

It doesn’t actually solve the problem

Double entry doesn’t actually solve the problem of mistyping your email address, users can still get it wrong. There’s no doubting it will reduce the number of errors from say 1% to 0.5% but big deal! Why penalise the 99% of users who get it right first time for such a minimal gain? Two fields means two chances of typing it incorrectly and the new possible error that the 2 fields don’t match.

Is my email address really that important?

Using the Oxfam form mentioned, my goal was to sponsor a work colleague. Why is getting my email address correct so important to this process? I can print out a receipt at the end of the process if I need it. I have the charge on my credit card if I need it as well. So if my email address is not critical to the process then there is certainly no reason to ask for it twice. In this case I wonder if it needs even be a required field. My cynical view is that Oxfam want my email address so they can market to me – I don’t appreciate that.

It reduces the chance of conversion

In our time starved, attention deficient society one extra field can make a difference to a user completing their task. This one field, when taken in context of how it makes a user feel and how it can create new errors, may just be enough to push the whole process into the user’s too hard basket.

Disabling copy and paste is just criminal

It’s bad enough asking me to confirm it but not letting me copy and paste the email address from the first field into the second confirmation field, as the Oxfam form in question does, is criminal. Why would you do it? In the unlikely event I’ve entered my email address incorrectly in the first, the fact I’m copying and pasting it means I’m looking at it again so that in itself should be more than enough of a check. Please people stop with this madness!

Autocomplete is reason enough to abandon it

The vast majority of browsers now autocomplete email addresses as you type. This is a wonderful feature and interaction that to me renders the whole double entry idea dead in the water. This reduces the likelihood of errors greatly and I think is the single most compelling reason to abandon email re-entry.

A better solution

As well as the tweets of frustration I got from friends I also got some nice suggestions on better solutions and links to interesting articles on this very issue. My preferred approach to this problem has been to show the email address the user has entered on the confirmation page in large text, with a link that makes it easy for them to change the email address if they did get it wrong. Here is an implementation I recently designed (albeit slightly verbose in hindsight):

A respected colleague and mentor of mine Brett Collinson of Modal Dialog also suggested this as his preferred solution. Another idea that Dan Naumman sent me was this one that suggests repeating the email address at the end of the form before the final submit button. I like the approach and think it, especially in combination with the confirmation page approach, is a much better solution than the double entry. The main reason is because it’s treating the user as though they got it right the first time and not adding any extra work to the process if they did.

I wonder if putting their email address on the button itself, e.g. “Create account for joe.bloggs@domain.com” would be even more effective. Maybe it would create a big an ugly button but then big buttons are very usable.

Thoughts and especially other suggestions welcomed.

Given the encouraging feedback and great suggestions I had after publishing my initial iPhone weather app concept it’s full steam ahead with version two. In this version I’ve added more information to the screen yet I’ve tried to reduce the overall visual noise. I’ve mocked up both a sunny and stormy forecast using data from BOM to get a better sense of how it would work with different forecasts.

Sunny day forecasted

Storms forecast

The specifics of what I’ve done and the rationale include:

• Added background photos that represent the overall forecast for the day. They are to instantly communicate what the day is likely to be like for users.
• Changed the sunrise and sunset text to icons (thanks Brett). Also removed the am and pm from the times, it’s pretty obvious which one is which! Both these changes help to reduce the overall visual noise.
• I increased the size of the current temperature considerably based on the feedback I received. This is the most important piece of information on the screen for users so make it really pop.
• I made the forecast maximum temperature bigger, this is because I consider it to be the second most important piece of information on the screen.
• I aligned the forecast min and max temperatures to the left of the screen with their lines extending full width across. I initially wanted to indicate where the likely peak temperature would be by just having a short line but I feel the projected line as well as the graph are indicators enough. Having them aligned on the left just feels much tidier.
• For Mr Green I added an umbrella icon which indicates whether you need to pack the wet weather gear or not.
• I faded out the beginning and end of the graph line as this feels more elegant. Ditto for the hour and noon lines. This again reduces the overall visual noise of the screen.
• I moved the title bar to the bottom of the screen – the information on it is once off information that once you know becomes insignificant.
• I added the long version of the forecast on the left-hand side of the screen – I feel this detail is important to have on the screen.
• Lastly I added the short version of the forecast below the city name in the title bar. Not necessary but it works there.

Love to hear your thoughts and ideas.

A few thoughts combined today which saw me rapidly put together this initial idea for an iPhone weather app. It’s a hot day here in Melbourne so lots of folks have been checking the weather to see how hot it is and when a change is due. People can get pretty obsessed by the weather, certainly all the users I tested when working on White Pages loved to know what the forecast was and I must admit it’s the first thing I check in the morning so I know what to wear when I hop on the bike and head off to work.

Recently I saw a website that presented the days temperatures on a line graph and I really liked the concept, it’s much nicer to have the days temperature shown on a line graph so you can instantly see how it’s trending. I can’t remember what site I saw it on (I’ll update this post when I find it) but the execution of it wasn’t great. So I thought I’d design something similar for the iPhone (to add to all the weather apps out there) and here is my first draft:

iPhone Weather App Concept

I’ve got a few other ideas I’d like to try incorporating too, like:

1. Compressing the mornings and elongating the daylight hours – how many people care about the wee hours really. This would make the more relevant information prominent.
2. Adding the textual forecast information in one of the corners, e.g “Partly cloudy. Winds north to….”
3. Adding a tiny right-hand side column showing the forecast for the coming few days.

This to me is the ideal home screen for a weather app – all the information I want to know prioritised and presented in an easy to absorb way. What do you think – like it or not? Should I try and develop it? Got any suggestions?

Luke Wroblewski has compiled some very compelling mobile spending statistics showing the amazing growth in the mobile market recently.

This presentation from Steve Krug, author of the wonderful book “Don’t make me think“, is a concise and easy to follow overview of usability testing. Lots of simple, practical things in it for anyone building software/websites who want to create great user experiences. There are some things you would have no doubt heard before but there are a lot of thoughts and approaches that I found were new and valuable for me including:

• Give up on validity – the usability problems will be so obvious that you won’t need numbers to validate them. Once you give up on validity then you don’t need to worry about questionnaires or surveys. This will save considerable time in the sessions.
• Just give me one morning a month – I love this idea the most. I’ve often been in the situation where I just can’t seem to convince the client to do user testing so the idea of asking for one morning a month is great – who can refuse that!
• Only communicate the top 3 problems to stakeholders, if you tell them everything they will be tempted to solve the low hanging fruit and not address the most important usability issues.

The motion graphics in the this music video are amazing! Crafted by Fluorescent Hill. Wow!

“The partnership with LinkedIn affirms Twitter’s role as a network for professional conversation.”

This quote from a New York Times article on the partnership between LinkedIn and Twitter stood out for me.

Twitter is often written off (by those who don’t use it) as a personal tool, a way to socialise with friends and mainly for the young kids. I’m 33, married with 2 kids and I use it predominantly as a professional tool. Sure I use it socially as well (who doesn’t want to know where to get the best coffee or a good cheap meal) but I predominantly use it to stay up-to-date on industry news, to ask my colleagues questions and to hear about what they are working on, struggling with etc.

I’m a huge advocate of user-centered design approaches and techniques and have been ever since I was introduced to them at The Hiser Group. So the recent article Discount Usability by the guru himself Jacob Nielson I think is timely reminder about some of these user centered design approaches and the value they offer. In the article Jacob reflects on the last 20 years in usability and design and how far it has come. Some quotes from it that I find particularly relevant in my day to day work include:

• “Simple user testing with 5 participants, paper prototyping, and heuristic evaluation offer a cheap, fast, and early focus on usability, as well as many rounds of iterative design.”
• “Discount usability often gives better results than deluxe usability because its methods drive an emphasis on early and rapid iteration with frequent usability input.”
• “Most design teams still don’t believe in paper prototyping, preferring instead to spend considerable time creating elaborate design representations before they start collecting user feedback.”
  I can’t emphasise enough how valuable it will be.

I found this short overview article about the Agile development process nicely written and a great summary of why Agile came about and what it’s focus is:

http://www.cooper.com/journal/2009/07/my_vision_of_agile.html

I’m working on a Register, Sign up, Join or whatever you think best to call it process for an online application. Where I’m at now (after deciding to go with “Join” by the way) is what’s the best process for those who have forgotten their password. As I currently see it (from testing some) there are three main ways to do it (please let me know if there are others worth considering) and these are:

1. Emailing the current password to the user,
2. Emailing a new password to the user and asking them to change this when they come back to the site, and
3. Emailing the user a link which they click on and then reset their password on the website.

Now my preference is for the easiest option, number 1, in that it is the easiest and simplest user experience. I do however realise it’s not the most secure but this is where it gets murky for me. I realise that emails are not very secure and can be easily intercepted but my feeling is that if someone can intercept an email then none of those proposed approaches is secure so go with the easiest for the user.

The only thing I can think of is that in options 2 and 3 the password is not actually known by the service – that it is somehow encrypted in the database and can not be pulled out and converted to text as such?

Your thoughts, preferences, views and expertise would be appreciated.