The worlds twin brother

Customer relationships ...

2010-06-03T15:38:00.002+01:00

... are very important, as any successful business knows ( and many not-so-successful business know to their cost ). It is important to engage with, relate to, and work with your clients, whoever they are ( and, I would say, it is important to know who your clients really are - they are not necessarily only the people paying ), to hear their problems, to resolve them, and to spend time with them to have clear lines of communication.

Having said that, the customer is not always right. If I am providing a professional service to someone, then within the areas of that professional remit, I am probably right. Establishing that is also part of the customer service relationship, meaning that they come to believe they are getting something worthwhile from me. If all I do is whatever they say, then what am I bringing to the table?

But this also applies within application interface development. Managing the relationship between the application and the user is critically important. It really doesn't matter what goes on below the surface, as long as the interaction and engagement with the user is working correctly. It means the application has to do what it says it is going to do, when it says it. It means that the application has to take responsibility for failures, to an extent, and not just send them back to the user. It has to take the responsibility to know what needs to be done within its remit - the computer - and handle that. If it needs to tell me what it is doing, that is fine, but I should be able to trust it to save stuff where I can find it, to handle problems that occur without my intervention, to just get on with the computer side of things, because that is what I have bought them to do.

Customer relationships are critical to both businesses and computer applications, for many of the same reasons.

Persona Driven Development

2010-04-27T21:54:00.002+01:00

I am a great believer that this should be the Next Big Thing. Personas, to those who do not know, are stereotypes of groups of users, reflected in a single person with a name, a profile and full context. They are used in the process of design, to identify types of users and what they are wanting to do with an application.

A lot of code work is done using TDD, test driven development, an approach that verifies code by writing tests for it, and verifying that the tests all work. This is fine, to a degree, and I do not have a problem with it, however, there is a danger that the bigger picture - the question of "does this application work" can get missed in the minutiae of lower level tests.

The approach that persona driven development takes is of acting like each of the personas, each of the people, and making sure that their progress through the system is as expected. Fine I hear you say, but what about edge conditions - what about ensuring that the application handles unexpected situations? Well there should be personas for these too. If you have a public web site, then one of your personas should be a hacker, and the behaviour that they exhibit will be to try to attack the site.

Similarly, if you are concerned about dubious input, then you might want a persona who is a new or inexperienced user, whose input may well be at the extremes of what is acceptable.

The importance of personas, and the reason that this is a different approach, is that each persona will have a level or rank, so that the personas representing the most common users will have the highest rank, the personas representing the least common will have the lowest rank. So, while you will cover the whole range of requirements, you can focus attention on the processes for the most common users, and not focus equally on all types or approaches. So the people who user the software the most will get the most attention.

Which seems right to me.

Another year, another experiment.....

2010-02-19T12:13:00.001Z

Well I have finally got to the point where I am setting up an experiment, doing some testing work, seeing what happens to people when they are asked to do tasks on web sites. And I expect some explosions!

Have I vanished?

2009-02-07T11:29:00.002Z

Not that anyone really cares, but no I haven't. I have submitted my first draft of introduction to my tutor, and recieved no response (doh). And I am going to be published! In ACM journals - very prestigious and remarkable. Which makes me feel much better about life.

Which is good, because work - as in any part of IT retail business - is a challenge at the moment.

Cloud computing

2008-11-23T13:03:00.002Z

Hey -there's a new buzz-word! It involves using remote servers to process your requirements on. Of course, it is not a new idea - mainframes used to do this when I started in the business. But even on PCs, the idea is not completely new - the world community grid has been exploiting a similar principle ( although very different technological solution ) for many years.

Which makes me wonder what would happen if a WCG agent were to access a cloud network? Or a similar sort of agent that was programmed to utilise a good proportion of the computing power available to it? I presume that it would suddenly become very productive! OK, probably not, but if your available computing resource is ( effectivly ) unlimited, there are applications that will take advantage of this. And not all are made by Microsoft.

The point ( as there is always a point )? Firstly, new ideas are rarely new, just newly developed and skinned. Which I don't mind as long as they are not sold as radical new concepts in computing. We have not had any of these for a very long time.

Secondly, introducing new ideas, new means of doing things sounds very good, but they will always have an impact on, and be impacted by, existing technologies and approaches. Nothing that is done is in a vacuum - it all has to fit into the context of the existing paradigm. Even if, as does occasionally happen, it ends up changing the paradigm.

Station announcements

2008-11-22T17:01:00.003Z

Why are they so bad? Why do they not give you the information that you need? I need to know which train will get me to my destination quickest. Instead of telling me this they give a whole lot of irrelevant information ( the timetabled departure, the final destination ). And out of this confusing set of information about various trains, I have to calculate which train I think is the best for me, as does everyone else.

And using the timetabled departure time and final destination as the key reference sort of works, until there are two trains with the same information. Then, to announce that the 08:16 to Moorgate is going to be late is meaningless, as there are two trains that might be late. So rather than clarifying the situation, it just makes things more obscure.

Now with all of the improvements that they have made to station announcements, why do they till not just tell me what I want to know?

The importance of chatter

2008-11-08T10:49:00.003Z

Having read a lot from chicks-sent-me-high about the importance of flow and the ability to focus on a task without being disturbed, I was interested this week to read in Donald Normans book "Things that make us smart" about the importance on chatter - the background discussion that often happens in an office, most of which is seen as unwanted noise that distracts from the focus you are looking for. According to Norman, this chatter - much of the details of which are irrelevant to us and we filter out - is important for giving us the overview, the scope the context in which things are happening. So, for example, pilots in communication with the airport control tower perform better when they can hear all of the controllers discussions, because they get something of an image of the skies around them.

So having, as we currently do, quiet times in the office is undoubtedly productive in many ways, but it is also important to appreciate the significance to productivity and working of the chatter - be it verbal or email - that is a normal part of the office environment.

Why men shouldn't be ordained

2008-10-11T12:58:00.004+01:00

Unashamedly pinched from various other places:

10. A man’s place is in the army.

9. For men who have children, their duties might distract them from the responsibilities of being a parent.

8. Their physical build indicates that men are more suited to tasks such as chopping down trees and wrestling mountain lions. It would be “unnatural” for them to do other forms of work.

7. Man was created before woman. It is therefore obvious that man was a prototype. Thus, they represent an experiment, rather than the crowning achievement of creation.

6. Men are too emotional to be priests or pastors. This is easily demonstrated by their conduct at football games and watching basketball tournaments.

5. Some men are handsome; they will distract women worshipers.

4. To be ordained pastor is to nurture the congregation. But this is not a traditional male role. Rather, throughout history, women have been considered to be not only more skilled than men at nurturing, but also more frequently attracted to it. This makes them the obvious choice for ordination.

3. Men are overly prone to violence. No really manly man wants to settle disputes by any means other than by fighting about it. Thus, they would be poor role models, as well as being dangerously unstable in positions of leadership.

2. Men can still be involved in church activities, even without being ordained. They can sweep paths, repair the church roof, and maybe even lead the singing on Father’s Day. By confining themselves to such traditional male roles, they can still be vitally important in the life of the Church.

1. In the New Testament account, the person who betrayed Jesus was a man. Thus, his lack of faith and ensuing punishment stands as a symbol of the subordinated position that all men should take.

The point? Just because you can argue for something doesn't make it right.

Can you help

2008-09-30T19:35:00.002+01:00

If anyone can help me with my research, it would be much appreciated.

I am now into the focus group section, and need participants. Please check out this, and let me know if you can assist. Thank you.

David Blaine - get a job

2008-09-26T18:51:00.002+01:00

The latest "stunt" by David Blaine has finished this week. He was supposed to hang upside down for 60 hours.

He didn't of course - he came down regularly for checks and toilet. He said "You don't expect me to pee on myself, do you" - well no, there are plenty of other people who will do that. But he didn't do what he said he was going to, and what he did do was tedious. And yet people pay to watch him? He makes money from being that boring?

So I have a suggestion for his next stunt. He works in an office for 45-50 hours in a week. He has people demanding things from him, he has a workload that he can't expect to get through. And he has an hours commute morning and evening - except for the days when the trains run late, when it is longer. Just a week, though, as any longer would be cruel, and I realise that he would need a few weeks to prepare and several weeks afterwards to recover.

Of course, this is what some of us do regularly. Its called having a job. And it is a whole lot more challenging than most of Mr Blains other "stunts".

Just how wide is my area of study?

2008-09-13T19:08:00.002+01:00

I have just looked up another book I need, in the UCL library. I thought it might be somewhere slightly unusual, as it is an Edward Tufte, and really should be in "design" or suchlike.

My first set of books came from Engineering, Geography and Computer Science, which at least were all in the Science library. Three different floors, but then I needed the workout. So where is the Tufte book? Archeology. Ok, it is called Beautiful Evidence, because it deals with the visual presentation of data, but why archeology? What train of events led them to getting this book, and didn't drive the science department in any form to get it?

I am so glad that the catalogue makes it simple!

Sometimes people get it right.

2008-09-11T20:17:00.002+01:00

I have joined the library at UCL, to provide me with some technical books. Boring, I know, but it has to be done, and the science library is walking distance from work - I can get there, get books out and get back in a lunchtime.

Last week, I looked up some books from the online catalogue, wrote down the location details, went to the library and found the books quickly and easily. Obvious, simple, exactly as I should be able to do - finding books in a library is an obvious thing to do. But I was pleasantly surprised, and I thought it was worthy of note. It is sad that such things are so rare that they are worth noting. But that is how things should be. I think.

The Motor Show

2008-08-04T20:17:00.003+01:00

Last Saturday, I went to the British Motor Show, which was a great day out, a fun and enjoyable trip, concluding with the Meatloaf concert.

I saw an F40 being driven, which is my favorite car ever. Even though it is 20 years old, I still believe that it is the peak of automotive engineering. The combination of performance and looks has not, in my opinion, ever been matched. Beautiful and brutal, it was a joy to watch.

Which does not do much for my green credentials.

Which is why the most interesting car I saw was the Lightning. I love the idea of electric cars, they have a lot of potential to revolutionise driving and our use of oil. However, most of them look and drive like milk floats, and so will never catch on with people who enjoy driving powerful cars, who enjoy the thrill of speed and performance. And whether you approve or not, until electric cars can start to satisfy the petrolheads, they will not make a difference our driving habits. Which is where the Lightning comes in - it is an electric driven sports car, with performance that would keep most people happy. Maybe not yet the F40 drivers, but it is a big step along the way, and I hope that the technology will both develop to make them even better, and also trickle down to give perfomant, reasonably prices cars. That are not milk floats.

It is very easy to tell peopel what they should have and shouldn't have. But if we want to win peoples hearts - and HCI work is very much about winning people to your applications - we have to acknowledge what people want ( not preaching about what they should have - there is plenty of that from everywhere else ), but do it in a good and positive way. Give people what they are asking for, doing it in the way that is should be done. Give them a Lightning.

Appointment systems

2008-07-30T21:14:00.003+01:00

A couple of weeks ago, I had to book an appointment with my doctor. The surgery has a new phone booking system, which I used, because it has some advantages, enabling me to book out of normal hours. However, in trying to use the system, I came across two problems, which I want to reflect on.

The first problem was that my normal doctor was on maternity leave. That is reasonable, but the booking system had not been updated, meaning that when I tried to book an appointment with her, the system hung. It didn't tell me that she was unavailable, it didn't decline gracefully. I had to ring the surgery in normal hours to get this information. By this time, I had to wait until the following week to get an appointment.

The second problem was that when I did book an appointment, the system double booked the doctor. Fortunately, there was a gap in her schedule, otherwise I would have had to take another day working from home some other time.

I have no idea why the first problem happened, whether it was human error or a software problem that meant this could not be done. For me this was the biggest problem, but I suspect that it will not be addressed. The second problem may be addressed, which is important, but is really an issue of the staff monitoring - there are ways of getting around this.

The problem is that what could be ( and probably will be ) a really good system is let down by some problems that should have been resolved or considered at an early stage. It is a pity, because even though I will use the system again, other could be - probably will be - put off it.

Silly error messages

2008-07-16T15:06:00.003+01:00

So, if no error occurred, why didn't you just do what you were asked to? And if ( as I suspect ) an error occurred, why not tell me?

So, hows your research going then?

2008-07-08T14:06:00.004+01:00

I was asked this the other day. It is an interesting question to try to answer. Unlike with a taught course, where there are clearly defined milestones, doing research is very foggy. It might be that in the last few days I have read something that will prove to be vital to my research - there again, it might be that none of it is. So am I making progress? I think so, mainly because it is so slow and agonising at the moment. If pain is a good indicator, I am doing very well.

The problem is that at the moment I am trying to produce some papers for publication, broadly around why I am doing the study. I know why I am doing it, and the range of ideas that make it a particular and distinct approach are clear to me, but writing it down in a consistent, interesting and word-limit-defined way is not so easy. I think - and hope - that this process is helping me to clarify what I am thinking, and will help me when it comes to writing this up in my thesis. But at the moment, it is giving me a headache. I want to get on with practical work, stuff that can give me data to work on, that can make me think I am doing something. But, of course, I need to know what I am doing first.

Season ticket - part two

2008-07-01T21:48:00.003+01:00

So, today my last season ticket ran out. I ordered my new one a week ago, but it hasn't arrived. To add to this, todays journey home was a nightmare - slow, infrequent and packed.

There are two lessons from this. Firstly, some things are time-critical, and they have to be done by the time required. And I have had no contact or help - apparently nobody cares about this.

Secondly, communication. At the station tonight the communication was non-existent. I wanted to know a) how often the trains were coming b) when they were coming and c) where they were stopping. And I got none of that without asking, which was frustrating ( for me and many others ).

So the train company has failed me twice, mainly by failing to communicate with me, or show any real concern for me. That is poor service, for which I pay a lot of money. Grrrr!

That time of year again

2008-06-27T18:35:00.004+01:00

When I have to buy my new annual travelcard to get into London with. I could have a good rant about paying a large portion of my months salary to the train company to provide me with a service that leaves a lot to be desired, but I might do that another time. Quite how the government expects people to use public transport when it is as expensive as that I don't know.

What I do find odd is that the ticket I get is made of cardboard. My current ticket is worn out - you cannot read anything on it without a lot of difficulty, and that ticket is a replacement for my previous one that also wore out.

I us it for travel to work, in and out of stations at each end of the journey, and no more. I don't generally travel at weekends, and because I do a station-to-station trip, it is not being used multiple times a day. It gets, if anything, the minimum usage that I would expect of it, and yet it wears out and cannot cope with even that.

Well, you will say, it just means that they have to replace it occasionally, which is no real aggro for them, so why invest more money in robust tickets. Well the inconvenience of my having to replace my ticket is bad customer relationships. The trouble the staff have in trying to read the ticket for replacement is time consuming, and makes them look bad. And the worst is that when I had it replaced last time, it was because it wouldn't go through the automatic barriers any more, so I had to show it - which was no use, as there was nothing to read on it. Which is time consuming when I want to be quick.

It may seem like a minor thing, probably because it is. But if you are paying a lot of money for a service, then minor things as well as major ones should be right. I pay fr an annual ticket because of the convenience, but the poor quality of the materials used makes it inconvenient - maybe I should revert to monthly tickets, save myself some ( a little ) money, and not have them wear out.

The magicians principle

2008-06-06T16:33:00.003+01:00

I have heard it said that, for a magician doing card tricks, if he can get you to select a card in 10 ways and produce it later in 1 way, he has 1 trick. If he can get you to select it in 1 way and produce it in 10 ways he has 10 tricks.

I think there is something very important about this approach, in terms of application design. If we can gather data in 10 ways, and show it in one, we have one application. If we can gather it in one and show it in 10, we have 10 applications. So if you can write your application once and pull data from a number of places, you can sell it to a number of different clients, each pulling data from a different place - it is one application that you can sell multiple times, which is far better than an application written once for a particular client.

But if you can front your data flexibly, you can sell multiple applications to existing clients. So you don't even have to go through the hard work of getting new clients. Gathering statistics once, and then providing different solutions to display or manage them is a prime example of where one piece of work can be re-utilised multiple times for single clients, thereby making it very cost-effective.

But is it right to sell software to people multiple times? Assuming that the costs are not ludicrous, and that there is some value adding taking place too ( so you sell them in-house monitoring, for example, using a variant of a tool they use themselves ), then I think this is OK. And it is hugely better than writing something that is so complex and configurable by the client that they never really learn how to use it themselves. 10 simple applications that are sold for £100 each are actually better value than one highly complex one sold for £1000 that people won't actually get full use out of.

Boris

2008-05-05T21:29:00.002+01:00

One of the problems with the London Mayor is that it is a high-profile position, that is therefore going to attract people who want some publicity for themselves. Ken was just that when he was first elected, although I think he improved as time went on, because he had a genuine passion for the city and the people.

And now Boris. Who I once thought was just an upper class idiot, then I wondered whether he was actually quite clever, and just pretending to be an idiot, and have finally concluded that he is, in fact, just an idiot. I suspect he will be a disaster, and I hope that the other assembly members will keep him in order just enough to avoid him causing too much havoc before he is thrown out in 4 years time.

The triumph of substance over reality? Something we know - sadly - too well in the software business.

The Apprentice

2008-04-20T15:45:00.002+01:00

I am sure that I am not the only one who is watching this series of the apprentice but last weeks episode was of particular interest. Both teams actually suffered from the same problems, that they were selling stuff before they had a clue whether they were able to fulfill it. At one point they had promised a turn around in an hour, when they had no idea as to whether they could do it at all, still less in a hour.

And this is where HCI stretches beyond simple user interface work. For a computer application ( as much as any other work process ) to work, it need both a clear interaction with the user, and functioning ( and reliable ) back end processes. Selling without being able to fulfill is just as bad, overall, as have the processes set up, but no orders. Providing a good user experience is about doing both, doing the whole package, seamlessly.

Oh, and this years bunch of apprentices are probably the biggest group of losers I have ever seen. The only talent most of them have is backstabbing. From the perspective of software applications, co-operation is the key, not blame or criticism.

Ubiquitous processing?

2008-04-12T10:24:00.002+01:00

In my early days of computing, we were very excited by getting a disk drive ( yes, it was a long time ago ). And what is more, in drive controller had a processor in it, that could be programmed. Cool. The concept of having a dedicated processor for specific functions developed into the graphics card, which will have its own processing power and memory, and has made a big difference, although mainly for the gamers.

But why did we stop there? Why do we not put a processor and memory in the CD and DVD drives, to read the entire disk in one go ( at top speed, so fairly quickly ), and then present the data as requested from memory? The reason, I suspect, is cost, but it would seem like a reasonable use of computing resources. I am sure there are situations where this would be viable.

But what about user interaction? Why not a processor dedicated to taking your input, dealing with it sensibly, and providing responses appropriately? What about managing all of the user interaction separately from the actual processing? A sub-system that could render web pages or application content, receive input, buffer it and ensure that it went to the right place could make the interaction experience vastly better. When we request functionality that requires the processing, we can hive that off, let it get on with it, and return to user interaction.

And one of the reasons we don't is that we don't trust the computer. When we ask it to save a document, we need to know that it is saved, not that the process of saving has been started, because we don't want to lose anything. But if we could get reliable hardware, reliable software, and be sure that requests we issue are carried out ( even if there is a power cut between them being issued and executed ) we could dedicate processing power to the interactions that are what we need ot be really interested in.

"Doesn't expect visitors"

2008-03-26T18:50:00.002Z

Over the Easter weekend, we went away. On Sunday, we went to church in the morning, to the local Anglican church. It was, overall, a good service. However, there were a few little points that made me realise that, as a church, it doesn't expect visitors. And if non-regulars do come, they are probably members of another church. I could have a big rant about churches who don't expect visitors, but I won't.

Afterwards, we went out for a meal, to a local carvery - not a particularly posh place, but a very nice meal. They had there a maitre d' - someone whose job was to look out for the customers, to ensure they were dealt with, order taken, seated appropriately whatever. I didn't actually need to know anything about the mechanics of getting my meal to me - they are important, but not relevant to me. And that was his job - to ensure that I didn't have to know about the mechanics, just to enjoy my meal.

There is, I think, a lesson for many churches here - that someone whose focus is on the "customer" is very important. But there is also a lesson for software developers. Yes honest. It is that most users don't need to know the mechanics of how stuff works, but if your website or application sends out the message "doesn't expect visitors", then there is a good chance that it won't have many. To provide a consistent interaction, irrespective of how things happen, is important. To provide some help and assistance to those who are visitors ( which is everyone at some point ) is important. Thinking of users PRIMARILY, not as a secondary idea, is crucial ot good software design.

Mussolini at least made the trains run on time

2008-03-05T18:55:00.002Z

( Actually, I gather he didn't, but that isn't necessarily relevant! ).

Yesterday, my trains were messed up. I should point out that this is actually quite unusual - I think only the second time in 6 months or more. I can accept that sometimes there are mechanical problems that cause the trains problems. I wish there weren't, but I don't have a basic gripe with that.

What I do have a gripe with is the customer service issues that arose out of this. In this situation, where some two thirds of the trains are canceled, I don't need the indicator boards telling me that about canceled trains. I do need them to be telling me about trains that are actually running, where they are stopping and when they will arrive.

The train I squeezed onto was, according to the station announcements and train indicators, a stopping train - all stations into London. I was slightly disgruntled by this, as I would rather it was fast to my destination, but at least I was on a train. But when it got underway, it appeared that no-one had told the driver, because it was fast, much to the irritation of people who needed intervening stops. And this just increased the frustration with the delays and problems that were occurring.

And this again is a question of customer service. delays and problems are not the issue. The issue is failing to communicate with the people who need to know what is happening. This applies to trains, software and computer service companies. Avoiding problems where possible is good, but accepting that there will be problems and being open and talking about them makes a huge difference.

The opposite of people bigging you up...

2008-03-05T18:47:00.003Z

It is very easy to assume that getting your customers to say nice things about you is an added extra. It is good if they do, but if not, then that is just expected.

I suppose it comes from the assumption that the opposite of being bigged up is not being talked about at all. However, the opposite of customers selling you and telling other people how great you are is that they will tell people how lousy you are, how rubbish your customer service is, and how they will never work with you again.

One example : We used to have phone services provided by a cable company. We will never return to them, no matter how good their packages are, and how cheap they claim it is, because their customer service was a nightmare. They did not do what they said they would, they could not be contacted, and their service provision was erratic. As well as never returning to them, I will tell anyone else not to use them. So their reputation is shot, something they could have resolved by either doing what they said they would, or making some reasonable financial offer. They did neither, because they don't really care about their customers.

Usability starts at home...

2008-02-23T17:59:00.003Z

One of the authors I am trying to read has done some significant work on "flow" - the state of being "into" a task, totally focused and thereby being very productive at the task. One of the applications of this idea is to computer tasks, which is why I am studying him.

However, you would have thought someone whose interest is in making things easier would have changed his name from "Mihaly Csikszentmihalyi". The surname is pronounced, apparently, Chicks-Sent-Me-High. But how you can get into any sort of flow having to grapple with a name like that, I don't know.

Be careful what you say

2008-02-14T18:51:00.003Z

Because sometimes it can be misinterpreted, and sometimes it is a Freudian slip that you may not notice. One friend ( yes, VZ, it is you ) seems to always read comments as meaning "I want a man to have sex with". Roughly.

However, sometimes it is just too good to miss. I listen to Christian O'Connell on Virgin radio in the mornings. This week, he has been running a competition to send someone to Chicago to see Bon Jovi - cool, and a good prize. However yesterday morning, he made a slip up. I think he meant to say something like "the winner will get a chance to see bj ( Bon Jovi ) in Chicago". What he actually said was "the winner will get a bj in Chicago". Which might have attracted a whole different set of entrants.

Quick Wins

2008-02-13T17:46:00.003Z

Rarely are. The problem with a "Quick Win" is that it is rarely quick, and usually not a win for everyone. If you are going to do something, it is worth doing properly. That may cost more, it may take longer, it may be more difficult to justify, but in the longer term, it will be a better result.

UI design is one area that is often skipped over, because it is seen as being a lot of money/time for very little. However, a bad UI will cause users problems and irritations ( often that they will not be able to explain ), and is likely to cost the client far more than would have been spent getting it right in the first place.

So no quick wins. You can have it quick, or you can have a winner. You can't have both.

Service

2008-01-26T11:03:00.000Z

I commute to London - into Farringdon station. Last night, when I arrived at the station, there were big queues, and most of the gates were closed. Why? Metropolitan line services were delayed, and the platforms were too crowded. So they stopped everyone going into the station.

I could actually see over to my platform - for the overland train. It was no more crowded than usual, but I couldn't get there because of a problem that was nothing to do with me. The officials took the panic option, and tried to close and evacuate the station.

Now I do have a degree of sympathy with station workers. However, the problems had been going on all afternoon, and there were only 2 people at the station - they were under-resourced to deal with the hundreds of people who wanted to use the station, including those like me who could have ( and did, in fact ) continued their journey without hindrance. There were problems, but the controllers went into panic mode, and caused ( probably more problems ).

Computer related? Sort of. How often have you been working on something, and then had problems because of an unrelated issue, or an issue that could have ( and should have ) been contained, without disrupting what you were doing? Too often the software goes into panic mode, rather than isolating the problem, and letting other things work around it.

Multitasking

2008-01-09T21:22:00.000Z

Two of the major advantages of computers are, it would seem, speed of processing things, and multitasking - doing a lot of things at once. Unless, of course, you are installing software, where it seems that neither of these seem to apply.

I have built a computer today, and it takes an age to do, and you can only install one thing at once. Why? Well I know the practical reasons - that is how things run, and they have to write to to the disk, and they often are looking at the same files. But why? I know what I want to do - install half a dozen applications. I don't want to care about doing one at once, the order that they have to be done. I want them to install, and I want to know when it is done, but I don't honestly care about the interim process.

Once again, I am having to get involved in a process that I care nothing about, and spend time ( all of my working day today! ) working through. Big Red Buttons please!

Gender specific design!

2007-12-23T18:50:00.000Z

I found this recently and thought that there is a point here, that ( largely speaking ) men and women behave and respond differently. Which is why variety and choice in interface design ( not just computer software interface - it applies to anything you use ) is so important. To provide something that interacts well with all sorts of people, it needs to be able to adjust responsiveness and response means very subtly.

Of course, one aspect of this is that there should be more women involded in interface design work.

Christmas

2007-12-17T18:37:00.000Z

It is interesting that the song consistently voted the best Christmas song is A Fairytale of New York, which has to be the most cynical Christmas song ever. I, of course, adore it.

And yet we are so often regaled with happy, cheery Christmas songs, because that is what Christmas is about. Isn't it? But maybe we should listen more to the fact that this jollity and fake happiness is not what most people experience of Christmas. In fact, for many people, it is a miserable and depressing time - a fact that is reflected in this song.

So, if you want an enduring hit song at Christmas, forget the Mistletoe and Wine, go for the Stubbed toe and Whine. I may be an old cynic ( I'd put money on it ), but it is far closer to the real experience of Christmas. And it doesn't give people a false sense of expectancy that it might, just once, be as it "should" be. Sorry, but it won't.

And what of the "true" meaning of Christmas - the coming of Jesus? A baby born to unmarried parents, in a borrowed place, who went on the run immediately after. A child who grew up to tell people that there was far more to life than materialism and fake happiness ( because once you have "stuff", you are bound to be happy, aren't you? ). And a baby who only lived to 33, before being killed by a mob. They are the stories of Christmas. We may not like it, but the truth is that life stinks a whole lot of the time.

Revolution in naming

2007-12-06T14:39:00.000Z

A revolution in software naming - or not. It is really not a lot of news to know that the next version of IE will be called IE8. That is easy to guess. What we really want to know is what it will contain, what issues will be fixed, what standards it will support etc.

And Dean has been slated for this post, not because there is anything wrong with it as such, but because it has a smug, self-satisfied sense of having given people information, while actually saying nothing.

Sorry, but communicating by saying nothing will not do. I presume the reason is that there is nothing to say, as IE8 has nothing other than a name. Fine - I don't have that many problems with that. But pretending that this is news, or that this indicates progress is deceitful and does nothing to boost the image of Microsoft.

Recruitment

2007-12-04T18:50:00.001Z

Well, recruitment agents in particular. The task of getting another job is a very difficult, time consuming and emotionally challenging one. And it strikes me, why do we do things that way? Where we register with a dozen agencies ( or more ), go to interviews where they ask you the same questions ( nearly ) and you have to answer as that particular job wants. And, in the end, it is still something of a lottery.

Why not do it a completely different way. Why not interview a candidate ( maybe even at a weekend? Or an evening? ), and then identify the roles that would actually suit them? This would give a degree of independence - so jobs that are totally unsuitable, however attractive, would be ignored. Also, the agent could have some discussion with potential employers, and so may be able to identify roles that would fit the candidate. The candidate would then go for interview to identify whether the person fits in personality-wise, not whether they are technically capable of it - that would have been sorted out.

Radical? Apparently. But it would mean that employers would see far fewer people, but all of them would be good fits for the job. And candidates would know that they have a really good chance of getting the position they go for. Why I am writing about this? Because doing things in a different way, in a way that works better with the candidates, is what I am all about! It is changing the approach, changing the focus from where it is to who it needs to be.

Or do you disagree?

Banksy and Tox

2007-11-25T16:42:00.001Z

Two of the best known graffiti artists, particularly in London. Firstly, one of the most prolific is Tox, whose activity stopped briefly while he was in prison, but without really interrupting his flow. I see his work on my journey into London each day.

Secondly, there is Banksy - this picture is one near to where I work. I like it - funny, artistic, clever. But are they so different? One is arrested and jailed, and the other is celebrated, with some of his art in galleries. But they are both painting where they are not supposed to, both grafitti-ing the walls of London ( and elsewhere ). Is it just that Banksy is more aesthetically pleasing? Is aesthetics the only difference between jail and art galleries?

If so, whose aesthetic decides? Public opinion - which is notoriously fickle? "Art experts" - then on what basis are they considered experts? Is there really that much difference between artwork that is acceptable and artwork that is not acceptable?

Big Red Button

2007-11-21T12:06:00.000Z

Sometimes, what we need is a big red button approach to interface design. This means, rather than the subtle approach so common, we put a big red button that says "Press me", which does the next thing we need to do - start the application, run some processing, it doesn't matter ( and shouldn't matter ) to me what is behind this. This applies to hardware interface too - I don't want to know that to start this up I need to turn on the power, plug in the computer, turn on the computer ... I want a button that will start things going, and tell me that I don't have a computer connected, so it won't work, or that the computer is not switched on or whatever.

Choice is wonderful - when it is real choice. But real choice is not about having to know what is going on in the background. When I start my car, I don't really want to have to turn the ignition to position 1 ( Radio on ), Position 2 ( Electrics on ) then start the engine and release the key. I want to start the car. Which is why the button push ignitions are such a good idea - I push a button to start the car, I push the same button ( marked, Start/Stop of course, not just "Start"! ). The fact that it has to run through a process underneath this is irrelevant to me. All I want is to start the car and get moving.

So applications having some clue about what I might want to do next may remove "choice", but it makes the experience of using them so much nicer. Which is why Amazon's one-click ordering is so successful. That is why applications that start up for themselves and I never have to interact with are great. Choice is great when it is genuine choice. But for such a lot of stuff, the BRB approach is so much better.

Gormengast

2007-11-18T18:58:00.000Z

I have recently finished reading this, and it did strike me as having something relevant to software applications. Gormenghast the castle is a massive edifice, with corridors, even whole wings, that are no longer used and the ways around them are forgotten to most. It also runs not for the benefit of the residents, but for the maintenance of the rules and the rituals. Titus Groan, the young heir, hates it and want to escape.

So how is this relevant to software? I have worked with various software that is rather like the castle, big and overgrown, extra bits being added as felt necessary, and parts left abandoned because they are no longer relevant or used, but they are forgotten about, and who knows what might be lurking there! And they are build and devised for the benefit of the software writer and organisation, not the end user. They are bound by the rites and rituals of software use - these may be good as the original rituals probably were, but they no longer serve the needs of the residents or users. And user like Titus feel a desperate need to escape while struggling with the fear of the unknown.

Sex

2007-11-13T13:56:00.000Z

Well, I have to get some hits to my blog somehow, don't I? Anyway, there is a degree of relevance here. Sex works best between two people where they are compatible. In many respects, it doesn't matter what their particular style is, as long as they both enjoy it . The problems often come where one partner wants flowers and romance, and the other wants whips and chains - that is incompatibility.

In user interface design too, the issue of compatibility is important. Different people want different things - some like a very in-your-face approach, an aggressive interface that shouts at you. Other people would rather have the B&O approach, where you have to guess where the buttons are. I am not suggesting that either is right or wrong - I know usability studies tend not to like either style. The important thing is compatibility, rather than any one particular style being "right" for everyone.

Oh so true

2007-11-12T09:13:00.000Z

It is one of those strange truths that wherever I have worked, Dilbert has been relevant, and could have been written in my office.

The art of consistency

2007-11-09T21:57:00.000Z

Archery - I am an amateur archer, not particularly good yet, but I enjoy it. How does archery relate to HCI and software interface design?

Archery is all about being consistent. There is no point in getting a 10 if the other 5 arrows are 2s and 3s. What is needed is to get all the arrows in the same area - consistently. So they should be scoring 9s and 10s. And maybe 8s from wild shots. I should point out my scores are not quite there yet!

Interface design is also about being consistent. There is no point in producing something very clever at one point of a site or application, and making the rest mediocre. If it is all consistent, even if none of it is hyper-flash, then it will be a better application overall. If everything works in the same way, it will be a better system. Consistency - ideally consistently brilliant - is far better than variability in terms of actually using things.

The Gulf War did not take place

2007-11-04T17:58:00.001Z

I have just read Baudrilland's book by this name. While being rather hard to read ( I suspect this is something to do with it being translated - very well, I gather, but lacking, maybe, the flow of the original ).

How does it relate to HCI? You will have to wait for my thesis* (!). but the idea that the control of the media - the interface - is of vital importance to controlling the perception of truth is an interesting one, and one that does have an impact on how we present information.

*i.e. I don't know the answer yet!

Managing my blog

2007-11-02T21:58:00.000Z

This is not meant to be a creep, but a reflection on my experience of blogging here. Setting up the blog in the first place, and maintaining it ( adding my comments, modifying it, making those little tweaks that are needed ) has been very simple.

What is saddest is that this comes as a pleasant surprise. It is an indication of the poor state of software interface design that finding a good product - something that is easy to use, and ( of crucial importance ) simple to get going on - is such a noteworthy thing.

Marx, Nietzsche and computers

2007-10-27T18:13:00.000+01:00

They really don't seem to have much in common, but after a training day today, I think that they do, and quite importantly. They are both social theorists, who look at the society we live in and interpret it in a certain way. ( Marx in terms of capitalist society and Nietzsche in terms of our lack of God and power struggles ). The oppressive situation Marx was writing into is no longer present as such, but, for many people, the slavery to computer systems is very similar. The HCI angle is that people being slaves to computer systems implies that the systems have not been well designed - people should be able to do a job, and utilise computers to help them, not be forced to use them to facilitate other needs within the company.

And Nietzsche? The power is very much concentrated in computer departments, and, as computers impact more and more of our lives, people feel powerless. Having destroyed god, we then have to use the god-like powers of computers to manage and control people. We - the computer developers - are the new high priests and priestesses.

The social theories may not seem relevant to HCI approaches, but a study of these can help understand not how to make things better, but why people struggle with computer applications.

Rome and HCI

2007-10-26T21:42:00.000+01:00

We have just had a nice few days in Rome, and some relevant thoughts struck me while there. OK, a little tenuous, but that is some of what this blog is about - working out some of the more tenuous links, and seeing if they have any value.

Anyway, Italians do things differently to the English. They don't do things wrong. They just behave in different ways. Lets look at their driving - especially in Rome. They are crazy drivers, they don't drive in lanes, they cut each other up, they don't stop for others. And as pedestrians, the only way to cross the road is to out-stare the drivers, and walk across, with cars all around you. Unnerving if you are not used to it. However, it all works, because everyone follows the same set of rules. And, most of the time, it works.

How does this relate to HCI? The interface from computer systems needs to understand that different people work in different ways, and as long as everyone follows the same rules, it can work. So often we write out interfaces in the English way - we wait, we look for gaps, we let people enter information when we are ready, we process in an orderly fashion. But I wonder if an Italian approach might also work - not clearly structured, but more random, flowing, taking whatever information the user wants to supply at time, and handling it. Yes it is more complex, but as long as everyone knows what is happening, it could work.

Maybe the idea is silly, but the principle that different cultures work in different ways, and we should devise our interfaces to work in ways that suit the users - matching their thought patterns - is at the core of HCI work surely.

Software and patents

2007-10-20T10:32:00.000+01:00

I was reading in the latest edition of CommACM which arrived last week, about software and patents. One comment that struck me was that a patent is awarded for something "non-obvious".

It takes me back, oddly, to the question of software documentation. As a developer, I am very poor at doing this, but when I do have to, my first question is "who is it aimed at?" If it is aimed at the complete novice developer, then I am being expected to write a C# training manual, and my code is probably not the best example - there would be more comment than code!

However, if it is aimed at the other members of the team, assuming they are moderately intelligent, then I struggle because most of what I write is, with hindsight, intuitive and obvious. Those aspects that are not need some explanation, but most of it is clear. So shouldn't need documenting.

In the HCI world, most of what we do is ( at least with 30 seconds hindsight ) very obvious - in many ways that is the point, that we try to design and develop application interfaces that are clear and obvious in their use. And if we apply the principles not just to the user-level interface, but also the code we write, then documentation should be not required, or at least very minimal.

However, if application code and user interfaces should be clear and obvious to other people with appropriate experience then it suggests that patents should never be granted, because we never do anything very innovative. Which, of course, is wrong, because writing good quality, clear code and interface design is, in itself, hugely innovative.

Whats all this about a twin?

2007-10-17T20:48:00.000+01:00

I have called this the worlds twin brother, because that is what I am, at least sometimes. It is because I can show empathy, like a twin, but with anyone - if I could actually do anything with it, that would be nice!

It does mean that all of those people who have complained to me about computer applications, I feel with them. And I get angry and frustrated when companied can spend billions on making applications fancy and flash, with all of the latest gizmos and ideas, but ( so often ) rather less on producing something that real people actually want to use, something that actually helps people do what they want to do. And something that helps people feel like people, not some strange appendage to a computer system.

People are people, they are important. Computer applications that miss this are rubbish, no matter how popular, expensive, leading edge or curvy.

Welcome!

2007-10-16T20:37:00.000+01:00

This is the first entry in a new blog, to try and express some of what I am learning about human computer interfaces, which I am studying for a PhD.

And, of course, anything else that I want to talk about.

I doubt that anyone else will be interested, but if you are - enjoy, and let me know!