Tuesday, November 23, 2004
Website changes
I've changed over to Blogger commenting, and still trying to iron out some of the glitches, so let me know if you have any commenting problems. [You'll notice all the comments from the old system have gone - there doesn't seem to be a simple way of keeping these.]
You may have noticed that there's been an archiving problem on the blog for a while - well, since Blogger introduced page posting and the commenting system back in the summer. I'll endeavour to fix it this week.
And having swapped over to the Mozilla Firefox browser, I'm now seeing the entire website as a little 'dirty', with some odd things happening here and there. Another task for a rainy day...
1:42 PM|
link to this item
|
0 comments
Visceral classification
Last week, I mentioned to a couple of people - who are discussing their home book organising arrangments - that I'd seen a couple of blog posts recently on classification by colour, so here they are: one from Peterme, and another on a reorg at the Adobe bookshop from McSweeneys, which Tom Coates posted on Plasticbag.
My friends got tested out on CD colours, and it seemed to work, so they've decided to go down the colour route.
I think this is a great method for items you know. Colour is more visceral than methods such as class or author.
1:34 PM|
link to this item
|
0 comments
Monday, November 22, 2004
E-voting panel report
NMK has posted a report of the NMK/IPPR joint event E-Voting: Policy and Practice, that I spoke at a couple of weeks ago.
11:02 AM|
link to this item
|
1 comments
Deyan Sudjic and the parodic Puup
Deyan Sudjic, a writer on design who has just been awarded an RSA medal for promotion of design - and was also one of the contributors to Touching the State - has a think piece on design in yesterday's Observer, which is a version of his RSA talk.
In How Ikea sold us a Puup - referring to Ikea's parodic Van Den Puup advertising campaign - Sudjic argues that the design debate should move away from style versus substance issues: "we need both".
Sudjic discusses Dieter Rams, who "has devoted himself to designing perfect objects that could defeat fashion", but found himself designing objects, such as a calculator and a record player, where the entire category of objects became redundant. He argues that attempts to put design beyond fashion are just as doomed in terms of life expectancy as Starck's or Van Den Puup's creations.
I'd argue that what's doomed is any attempt at being precious, whether with regard to style or substance. Design serves many purposes: entertainment or pleasure, safety, ease of use. But in the end, society moves on, and new products and services are required. Just as well, or we'd all be out of a job.
11:02 AM|
link to this item
|
0 comments
Friday, November 19, 2004
System failure
This has not been a good year for major public sector IT projects, or for IT projects in general.
An inquest into the Court Service's Libra system was published last year, and we're now onto the fourth project designed to provide the court service with a case management system. A recently leaked government document revealed that the Department for Constitutional Affairs is concerned about software errors and fit for purpose issues.
Last month, Computer Weekly reported:Magistrates' courts across England and Wales are largely autonomous and many have their own way of doing things. But the NAO said the Department for Constitutional Affairs "sought automation as a priority before questioning the existing business processes".
In response the department said it recognised that development of best business processes should have come before seeking an IT solution. However, with legacy equipment in danger of failing, it was under pressure to deliver new systems.
Also, the department, like the NHS in its implementation of a £2.3bn IT-led modernisation programme, pointed out that it did not have the authority to impose business process change on the independent Magistrates' Courts Committees. And it was reluctant to attempt further major change while the committees were going through other reforms.
But putting IT before consideration of business processes made it "difficult to obtain a single view of IT requirements across the various committees and this contributed to the difficulties in developing the new system", according to the report of the Public Accounts Committee. Which is a long way of saying that there was a failure to research existing processes and contexts of use.
The result is a system that cannot cope, for example, with offenders who move house, from one court's jurisdiction to another.
Last week, higher education minister Dr Kim Howells faced the music before the Education and Skills Committee over the UKeU (UK eUniversities) debacle. He admitted that a lot of money was put into constructing a platform and not enough into content - and there was a lot of ignorance about courses and added The people in charge should have been much more sensitive to the needs of students around the world. I doubt that Howells is the person who knows most about why this project failed - he has many other responsibilities. The fact that he has highlighted the lack of research concerning context of use speaks volumes. The Education and Skills Committee is expected to report on UKeU early next year.
This week, we've seen the resignation of the chief of the Child Support Agency (CSA), following computer chaos and a rapidly rising payments backlog. It seems the CSA system is unable to cope with real life, and suffered from unclear specifications from the outset. Webcast of Committee proceedings.
Also this week, Jonathon Ford resigned as managing director of the National Assessment Agency, which oversees national curriculum tests. This followed publication of a critical report which, among other things, documented IT failures such as a "poorly planned" implementation of web publishing of test results. The website suffered from a range of problems and collapsed when accessed by a high volume of users.
There have also been private sector failures - for example Sainbury's recent scrapping of a new logistics system which has apparently "not delivered". But we tend to hear less about them, largely I suspect because the private sector is better at keeping failures under wraps.
I suggest that there are two common threads running through these stories. The first is an attitude of 'we know best' on the part of senior decision makers, leading to unwarranted assumptions about how people work, erroneous requirements specifications, and finally systems that are not fit for purpose. If the requirements aren't right, it doesn't really matter whether there are software errors: perfect code will be useless if the system doesn't do what is needed.
The second thread is a belief that IT, merely through its existence, changes business processes and people's behaviour. If IT systems cannot deal with necessary processes and real life, people will be forced to work around them, forced back to manual systems, forced to make up for system failures themselves. That's why we have a CSA that currently has several systems, including manual ones, trying to do the same job, as staff try to reconcile people's lives and inadequate software; why we have hospital information systems with severely out-of-date data that cannot be trusted; and why court service workers have to use multiple terminals to answer the simplest question about a single person.
Commons Work and Pensions Select Committee chairman Sir Archy Kirkwood, speaking about the CSA case, said this week that, "This is not just about computers". He was talking about a general management failure within the CSA, but he could as easily have been referring to a problem with the way management thinks about projects involving computers.
1:02 PM|
link to this item
|
0 comments
Thursday, November 18, 2004
Google Scholar beta
New applications and services are arriving thick and fast. The latest is Google Scholar (in beta), which is a result of Google's collaboration with a scientific and academic publishers. It's being billed as a first stop for researchers looking for scholarly literature such as abstracts and peer-reviewed papers.
1:56 PM|
link to this item
|
0 comments
Predict and provide?
There's an interesting article in the December issue of Prospect (paying subscription required) on the work of Danish architect-town planner Jan Gehl.He always starts his lectures with the same story, about the English architect Ralph Erskine, resident in Sweden for several decades now. When asked what makes a good architect, Erskine replied, "First of all, you must love people." Gehl claims that all his work is dedicated to making cities places in which people want to live, work and play. To that end, he's convinced Copenhagen to increase pedestrian space from 15,800 square metres to over 100,000, gradually remove car parking by taking away a few spaces at a time...Copenhagen now has more outdoor seating than Melbourne or Perth.
I like the story that he tells about skunks: It concerns a man living in a small town in midwest America who finds a skunk in his basement one morning. He asks a neighbour for advice on how to get rid of it. "Easy: lay a trail of breadcrumbs from the basement back out into the woods." The following morning he has two skunks in his home. There's a lesson here for doing research rather than trying to predict, or rather guess. People - and skunks - need to be studied to see how they really behave.
Gehl wrote a report for Transport for London this summer, Towards a fine City for People.
11:53 AM|
link to this item
|
0 comments
Wednesday, November 17, 2004
Design Engaged
Material is starting to emerge from Design Engaged, which took place in Amsterdam last weekend. Check out Matt Jones, Chris Heathcote, some sketches of participants and a personal compass from David Erwin, Design Engaged photos, and a post from Ann Galloway bringing together a number of links to participant contributions. There should also be a post from Fabio Sergio at some point.
9:21 AM|
link to this item
|
0 comments
Tuesday, November 16, 2004
Microsoft Office updates and Firefox
Warning: You are viewing this page with an unsupported Web browser. This Web site works best with Microsoft Internet Explorer 5.01 or later or Netscape Navigator 6.0 or later. Click here for more information on supported browsers.
Internet Explorer version 5.0 or later is required Help
The Office Update site can perform detection on your computer and tell you exactly which updates you need! To take advantage of this service, please return to this site with MSN Explorer or Microsoft Internet Explorer version 5.0 or later installed. So we must run - certain versions of IE - to get Office updates? I'll take a guess on when MS intends to add Firefox as a supported browser: sometime never. And I'll also take a guess that this kind of strategy - irritating the customer - is going to backfire on MS.
By the way, I really like Firefox. Although Google search iss integrated, I hope that Google implements a full Google toolbar for Firefox, with PageRank etc.
1:11 PM|
link to this item
|
0 comments
Monday, November 15, 2004
Oldboy
Oldboy is a bizarre but compelling Korean film, with the main character being the victim of something between military PsyOps and the Skinner box. Go see it.
6:13 PM|
link to this item
|
0 comments
Design Museum blues
I went to the Design Museum with Whitney Quesenbery on Saturday. Two interaction people in a design museum, post-James Dyson's resignation as chair of the board. Some thoughts.
Constance Spry exhibition: this show gives designers a bad name. It gives women a bad name. How many women get a chance to get featured as designers at a London museum? Why Spry? Spry is not about design, though it may be about art, or about style. And the curating of the show itself leaves a lot to be desired. Surely something a little more analytical, more critical, should have been in order.
Marc Newson exhibition: fun, lots of interesting stuff, but again, a seeming lack of interpretation. [By the way, he'll be speaking at the Museum on 25 January.]
The Design Museum in general: someone on the staff clearly has a thing about chairs. And chairs. And more chairs.
I like chairs. A lot. But there is other stuff in the world. The bulk of stuff here is about interior or product design, and a very narrow approach to both at that (cars, chairs and record sleeves make up around three quarters of the exhibition space at present, not counting Spry and Newsom special exhibitions). It's a rag-bag, a jumble, without any kind of analysis, giving the impression that there is no curator. There's nothing on interaction design (or did I miss something?) and little that addresses the issue of what design is or isn't.
The bookshop follows in the same vein, with lots of glossy, coffee-table material, and not so much on the analytical, thinking front.
Nice diversion - and I did enjoy the Marc Newson - but rather a disappointment overall.
In my view, Dyson was absolutely right.
5:39 PM|
link to this item
|
0 comments
Saturday, November 13, 2004
The end of..
Dirk Knemeyer's The End of Usability Culture, Redux is raising a bit of a storm in the usability community.
7:33 PM|
link to this item
|
0 comments
UPA salary survey
The UK chapter of the Usability Professionals' Association is running its annual salary survey. If you're in the business, do take part.
10:42 AM|
link to this item
|
0 comments
Friday, November 12, 2004
Form and function
Amusing to see the juxtaposition of two London events on my events listing page being blogged by Matt - BlackBelt - Jones. Maybe I don't see this so much as a contrast: Concorde etc. (the boffins) are just as much about form as function, and the iPod is just about as much about function as form.
And how I wish I could scrape all that events info. Unfortunately, it's all pretty human-based. A labour of love really.
1:06 PM|
link to this item
|
0 comments
Office refurbishments
The new office is coming along nicely. Eight-foot-plus-high wall-to-wall shelving, made to measure, all now painted and largely filled; new desktop machine which arrive 7pm last night, plus backup hardware; ergonomic chairs...Next steps are are a new, large office table and, I think, a large comfy sofa, the latter to promote thinking time. Having all the books visible is a definite plus (see Don Norman's book Things That Make Us Smart).
But do I put the sofa next to the four large sash windows overlooking multifarious trees and wildlife (what *is* that extraordinary bird?) and move the desks onto a boring wall (near the whiteboards), so better to concentrate on work? Or keep the two desks with computers next to the windows overlooking the birds, bees and weather sweeping in across London? Decisions, decisions...
11:45 AM|
link to this item
|
0 comments
Tuesday, November 09, 2004
Visual representation of information
The Guardian G2 supplement has a lovely column entitled Why I love... (always written by an enthusiast), which this week took a look at Edward Tufte, a marvellous writer on the visual representation of information. His various books are well worth exploring.
And on a topical note, check out this University of Michigan page from Michael Gastner, Cosma Shalizi, and Mark Newman, on ways of representing the 2004 US election results. Their point is that things are not as black and white - or red and blue - as they might appear from the graphics published in the press.
Perhaps not as funny as this redrawing of the national boundaries in North America, which apparently first saw the light of day in a German-language publication in Switzerland.
And on another humorous note, in last month's Atlantic magazine (paying subscription required), Ben Birnbaum fights the red-blue divide with his Crayola Nation, a partly completed geography to be completed with a range of colours such as camouflage (waitress named Lureen), pesto (year-round family-yard-sale cluster), moss (place that could be mistaken for Canada), melon (cellar with Pilates equipment in original carton on floor behind furnace), and pineforest (courthouse with spittoons).
5:59 PM|
link to this item
|
0 comments
Job titles
Many years ago, I spent some considerable time trying to get IT projects off the ground at BT. While I worked at the common or garden BT HQ, I was aware of a somewhat different breed, raised and nurtured by BT Martlesham (now Adastral Park) in East Anglia, who came from an astonishing range of disciplines (doctors of entomology etc.) and were pretty much left to their own devices, presumably to come up with bright ideas of use to the firm and humanity in general.
This week I received an email telling me that BT has an 'enterprise embryologist'. He's shortly to give a talk at the University of Sunderland, as follows: In an infinite world and a finite time to decide, how do we answer the most rudimentary of questions 'that faced with this predicament what do I do next'. This talk discusses the role of information and regulation in complex systems.
ABSTRACT:
I remember a story of a family who had grown up in a mixed race society. The house of the family stood on the street that divided the multi-racial community. The father of the family had lived in the neighbourhood all his life and had listened intently to what all his neighbours had thought and said and learned of their fundamentally different views. One day his recently graduated son asked him a question concerning their wide-ranging viewpoints. "Who is right?" His son asked, "They all are" replied his father. "But this cannot be" his son argued from a Aristotelian viewpoint, "Only one of them can have the truth!" he insisted.
"The problem is not the truth" his father insightfully answered, "The problem is trust". The problem is understanding, the problem is understanding understanding; the problem is making decisions based upon in principle undecidable questions.
This presentation looks at how large complex enterprise can decide upon a common course of action by its senior managers by utilising a soft systems approach to decision making. I strongly suspect that Stephen Brewis has not spent his life studying embryos, and this abstract begins to sound almost Rumsfeldesque towards the end, but sounds interesting nevertheless.
5:15 PM|
link to this item
|
0 comments
More 'blame the user'
On the US elections, from Kablenet, this week:According to IT industry representatives, human error had accounted for most of the problems. They maintained that technology passed the test. Grrr. Design a poor system. Blame the users. Continue to use poorly designed system. Blame the users. And so on.
I am, I suppose, in the IT industry. But I fail to identify with this position. Because it's completely pointless and gets us nowhere.
We build this stuff for people, not for other machines. These machines do not exist in a vacuum, they're part of a wider system that includes people, citizens. If people have problems with the stuff we design, we need to rethink the design. And if we are designing for everyone - as we are with voting - including your grandmother and mine, we need to build something our grandmothers can use. There's no point declaiming 'human error' or 'user error' ad infinitum, like some mantra, and carrying on with business as usual back in the developer cave.
4:02 PM|
link to this item
|
0 comments
Sunday, November 07, 2004
Voter intention, data and measurement
In the same article, David Dill's views on data are addressed: But Dill noted that the data that voting machine vendors and academics generally use to evaluate the integrity of e-voting machines doesn't include the kinds of problems that voters have been reporting. Generally, the number of undervotes and overvotes on a machine are used to measure their effectiveness.
Undervotes occur when a machine records no choice for a particular race -- either because the machine failed to record it or the voter chose to skip the race. Overvotes occur when voters choose more options or candidates than the race allows. E-voting machines are supposed to make it impossible for voters to overvote.
But "recording a vote for a wrong candidate is not something that shows up in the statistics," Dill said. It doesn't show up in the statistics because officials have no way to know whether a machine incorrectly recorded votes. Without a paper trail or some other way to independently verify that the votes on the machines are the votes voters intended, there's no way to truly measure the accuracy of the machines. User intention is core to any usability analysis of systems or equipment. While Dill argues that introducing a paper trail will be the answer, usability testing is a standard approach that has been used for many years, and that can be used prior to any machines being put into the field and used 'in anger'. It does seem, though, that there is widespread ignorance of the field of usability, among vendors, among governments, and among academics in other fields.
I have to say I'm a little stunned to find that political scientists and lawyers are happy with just measuring over- and under-voting, and ignoring voter intention, whatever the voting technology.
It's rather like measuring the usability of an ecommerce website by seeing whether people can buy something, anything, on it, without taking into account that when they wanted to buy a toaster to send to their mother, they ended up ordering a ghastly three-piece suite that was despatched to their old address.
1:00 PM|
link to this item
|
0 comments
Blaming the voters...again
Various publications have reported on US voting machines that failed to start on election morning, or those that were not plugged in (and whose batteries died as the day wore on).
More interesting perhaps was the story in Wired relating to calibration and usability problems.
For example, the machines in Texas from Hart InterCivic: Michelle Shafer, spokeswoman for Hart InterCivic, said the problem that occurred in Texas with her company's machines were caused by voters rather than by the machines. The Hart machines are not touch-screen machines but instead use a wheel that voters turn to make their selections. Shafer said after choosing the straight-party option, many voters turned the wheel to manually go through the races and click their choices individually to emphasize them, not realizing that in doing so they de-selected their choices. Shafer said they probably then mistakenly moved the wheel to select a candidate from another party.
"It's not a machine issue," Shafer said. "It's voters not properly following the instructions." Shafer, if - in your own words - "many voters" did something you didn't expect and cater for in your machine design, then you haven't done your user research. Voters don't have time to learn, or to 'read the manual', especially with long queues of voters at many polling stations. Voting equipment needs to be 'walk up and use'.
You're right, it's not a machine issue, but it's not a voter issue either: it's a vendor issue. A vendor that fails to do its user research will produce equipment that many people will have problems with. David Beirne, spokesman for Harris County [in Texas], where some of the problems occurred, said voters had made the same mistake two years ago when political parties instructed voters to go back through the ballot and emphasize their choices.
"I think that often times the voter information passed out to voters is incorrect," he said. "We encourage voters to take their time and ask questions and watch the videotape demonstration that's provided." Well, if voters made the same "mistake" two years ago, and nothing has been done to improve how the machines function, then it's also an issue of county or state responsibility. If the county knew there was a problem, and took no action so that the same thing occurred again...
And no, videos won't be the answer, unless people can see them at a time/place that makes sense. A video is provided, but he doesn't say where/when people can see it. When do people get to see the videos? At home? While they're standing outside in a long queue, waiting to vote? I suspect not.
11:41 AM|
link to this item
|
0 comments
Saturday, November 06, 2004
about, with and for presentations
Many of the presentations from the 'about, with and for' conference that took place in Chicago a couple of weeks ago have been posted to the Web.
Speaker presentations
Panel presentations
Workshop presentations
4:00 PM|
link to this item
|
0 comments
Friday, November 05, 2004
Voting machines as black holes
It's amazing what some vendors sell. USA Today reports that in one North Carolina county, just one voting machine 'lost' several thousand votes. Each machine could apparently store only 3,005 votes, though the county was told by the vendor that the figure was 10,500.
Voters gaily voted in their thousands, only for many of their votes to disappear into a black hole.
According to the vendor, machines flash a warning message when there is no more room for storing ballots: "Evidently, this message was either ignored or overlooked" according to the head of the firm concerned.
Evidently, he's not a very bright guy.
How is it that first a - transient? - warning message is sent to the screen, where it will be seen only by voters, not election officials, and then the machine continues to allow votes to be cast, seemingly without limit, when none of the data can be stored? Presumably the one person who saw the message didn't report it (or more probably didn't notice it at all).
Is it really sensible to design a system that puts the onus for the smooth operation of a polling station, and the accurate recording of all votes cast, on one voter?
[And 3,005 stored ballots seems a tiny number; perhaps a way to sell more machines?]
More comment from Professor David Wagner at Evoting-Experts.
1:59 PM|
link to this item
|
0 comments
New blogger on the block
User-centred design consultant Michael Andrews - who has recently moved to New Zealand from London - has decided to take the plunge and start blogging. Michael's a thoughful practitioner who writes well, so Modules and Wholes promises to be a good read.
11:06 AM|
link to this item
|
0 comments
Thursday, November 04, 2004
Make it simple
Make it Simple, an Economist technology survey, is pretty on-message about designing systems. So much so that some usability consultancies are already sending it out to clients.
You can access the rest of the survey stories from the same page.
10:38 AM|
link to this item
|
0 comments
Wednesday, November 03, 2004
Touching the State
The Design Council has now posted the Touching the State magazine to its website. (5.5MB, pdf)
I have a piece in there titled 'The X Factor'.
2:34 PM|
link to this item
|
0 comments
Tuesday, November 02, 2004
Voting problem data
MSNBC is running a voting problem line for the US election, with summary data available online on its website. Check out in particluar the breakdown by voting system/method (link in top right corner).
4:04 PM|
link to this item
|
0 comments
Usability problems with ballots
US interaction designer Karen Schriver last week addressed some of the usability problems with various Florida ballots for the Orlando Sentinel (free registration required). She considers issues such as all-caps instructions (more difficult to read), wording of instructions to voters (changing just one or two words can make the difference between clarity and confusion), and the types of marks that voters are required to make for optically scanned ballots (people work better with some types than others).
Over on the Miami Herald, a shorter article back in the summer addressed the same theme, but concluded with some unfortunate words from a political sciencist: Ansolabehere of the Voting Technology Project said studies have shown that while arrows create more confusion among voters, scanners can better process those that have been filled out correctly than ballots that have been bubbled in.
''Those two things cancel each other out,'' said Ansolabehere, a political science professor at MIT. We ought to be able to design ballots that can be handled easily by BOTH voters and machines, not just accept ballots with poor usability.
10:47 AM|
link to this item
|
0 comments
ICTs in the public sector
Emily Keaney of IPPR has put together a summary of the proceedings of an IPPR seminar I attended last week on whether ICTs can make public sector workplaces more productive.
9:59 AM|
link to this item
|
0 comments
Monday, November 01, 2004
Farenheit 9/11
There'll be many US election parties around the world tomorrow. For anyone wanting to show the film Farenheit 9/11, it's now available online: http://marc.perkel.com/archives/000468.html (Due to be taken down on 2 November)
1:15 PM|
link to this item
|
0 comments
E-voting: Policy and Practice
A reminder that the IPPR/NMK event on e-voting is taking place this Thursday. Both the Electoral Commission and the Office of the Deputy Prime Minister will be represented, so a good opportunity to chat to the decision-makers. I'll be speaking about design, usability and acessibility issues.
I have one free invitation to give away, so let me know if you'd like to come along.
More info: http://www.nmk.co.uk/event/2004/11/04/e-voting
1:05 PM|
link to this item
|
0 comments
|
