I’m writing this post effectively ‘offline’ into a word processor for later upload. I sometimes do this as I find it easier to write without the distractions of editing within a web browser, or because I am physically disconnected from the Internet. However, now I am connected, and indeed I can see I am connected as a FTP file upload is progressing, it is just that anything else network-related is stalled.

The reason that the FTP upload is ‘hogging’ the network is, I believe, due to a quirk in the UNIX scheduling system, which was, paradoxically, originally intended to improve interactivity.

UNIX, which sits underneath Mac OS, is a multiprocessing operating system running many programs at once. Each process has a priority, called its ‘niceness‘, which can be set explicitly, but is also tweaked from moment to moment by the operating system. One of the rules for ‘tweaking’ it is that if a process is IO-bound, that is if it is constantly waiting for input or output, then its niceness is decreased, meaning that it is given higher priority.

The reason for this rule is partly to enhance interactive performance in the old days of command line interfaces; an interactive program would spend lots of time waiting for the user to enter something, and so its priority would increase meaning it would respond quickly as soon as the user entered anything. The other reason is that CPU time was seen as the scarce resource, so that processes that were IO bound were effectively being ‘nicer’ to other processes as they let them get a share of the precious CPU.

The FTP program is simply sitting there shunting out data to the network, so is almost permanently blocked waiting for the network as it can read from the disk faster than the network can transmit data. This means UNIX regards it as ‘nice’ and ups its priority. As soon as the network clears sufficiently, the FTP program is rescheduled and it puts more into the network queue, reads the next chunk from disk until the network is again full to capacity. Nothing else gets a chance, no web, no email, not even a network trace utility.

I’ve seen the same before with a database server on one of Fiona’s machines — all my fault. In the MySQL manual it suggested that you disable indices before large bulk updates (e.g. ingesting a file of data) and then re-enable them once the update is finished as indexing is more efficient on lots of data than one at a time. I duly did this and forgot about it until Fiona noticed something was wrong on the server and web traffic had ground to a near halt. When she opened a console on the server, she found that it seemed quiet, very little CPU load at all, and was puzzled until I realised it was my indexing. Indexing requires a lot of reading and writing data to and from disk, so MySQL became IO-bound, was given higher priority, as soon as the disk was free it was rescheduled, hit the disk once more … just as FTP is now hogging the network, MySQL hogged the disk and nothing else could read or write. Of course MySQL’s own performance was fine as it internally interleaved queries with indexing, it is just everything else on the system that failed.

These are hard scenarios to design for. I have written before (“why software need never hang“) about the way application designers do not think sufficiently about potential delays due to slow networks, or broken connections. However, that was about the applications that are suffering. Here the issue is not that the FTP program is badly designed for its delays, it is still responding very happily, just that it has had a knock on effect on the rest of the system. It is like cleaning your sink with industrial bleach — you have a clean house within, but pollute the watercourse without.

These kind of issues are not related solely to network and disk, any kind of resource is limited and profligacy causes damage in the digital world as much as in the physical environment.

Some years ago I had a Symbian smartphone, but it proved unusable as its battery life rarely exceeded 40 minutes from full charge. I thought I had a duff battery, but later realised it was because I was leaving applications on the phone ‘open’. For me I went to the address book, looked up a number, and that was that, I then maybe turned the phone off or switched  to something else without ‘exiting’ the address book. I was treating the phone like every previous phone I had used, but this one was different, it had a ‘real’ operating system, opening the address book launched the address book application, which then kept on running — and using power — until it was explicitly closed, a model that is maybe fine for permanently plugged in computers, but disastrous for a moble phone.

When early iPhones came out iOS was criticised for being single threaded, that is not having lots of things running in the ‘background’. However, this undoubtedly helped its battery life. Now, with newer versions of iOS, it has changed and there are lots of apps running at once, and I have noticed the battery life reducing, is that simply the battery wearing out with age or the effect of all those apps running?

Power is of course not just a problem for smartphones, but for any laptop. I try to closedown applications on my Mac when I am working without power as I know some programs just eat CPU when they are apparently idle (yes, Firefox, it’s you I’m talking about). And from an environmental point of view, lower power consumption when connected would also be good. My hope was that Apple would take the lessons learnt in the early iOS to change the nature of their mainstream OS, but sadly they succumbed to the pressure to make iOS a ‘proper’ OS!

Of course the FTP program could try to be friendly, perhaps when it is not the selected window deliberately throttle its network activity. But then the 4 hour upload would take 8 hours, instead of 20 minutes left at this point, I’d be looking forward to another 4 hours and 20 minutes, and I’d be complaining about that.

The trouble is that there needs to be better communication, more knowledge shared, between application and operating system. I would like FTP to use all the network capacity that it can, except when I am interacting with some other program. Either FTP needs to say to the OS “hey here’s a packet, send it when there’s a gap”¹, or the OS needs some way for applications to determine current network state and make decisions based on that. Sometimes this sort of information is easily available, more often it is either very hard to get at or not available at all.

I recall years ago when internet was still mainly through pay-per-minute dial-up connections. You could set your PC to automatically dial when the internet was needed. However, some programs, such as chat, would periodically check with a central server to see if there was activity, this would cause the PC to dial-up the ISP. If you were lucky the PC also had an auto-disconnect after a period of inactivity, if you were not lucky the PC would connect at 2am and by the morning you’d find yourself with a phone bill more than your weeks’ wages.

When we were designing onCue at aQtive, we wanted to be able to connect to the Internet when it was available, but avoid bankrupting our users. Clearly somewhere in the TCP/IP stack, the layers of code over the network, at some level deep down it knew whether we were connected. I recall we found a very helpful function in the Windows API called something like “isConnected”². Unfortunately, it worked by attempting to send a network packet and returning true if it succeeded and false if it failed. Of course sending the test packet caused the PC to auto-dial …

And now there is just 1 minute and 53 seconds left on the upload, so time to finish this post before I get on to garbage collection.

1. This form of “send when you can” would also be useful in cellular networks, for example when syncing photos. [back]
2. I had a quick peek, and fund that Windows CE has a function called InternetGetConnectedState.  I don’t know if this works better now. [back]

If you are expecting the annual virtual cracker from me it is coming … but maybe not before Twelfth Night :-/

The New Year is bringing changes, not least, as many already know, I am moving my academic role and taking up a part-time post as professor down in Birmingham University.

At Birmingham I will be joining an established and vibrant HCI centre, including long-term colleague and friend Russell Beale.  The group has recently had substantial  investment from the University leading to several new appointments including Andrew Howes (who coincidentally also has past Lancaster connections).

The reasons for the move are partly to join this exciting group and partly to simplify life as Talis is based in Birmingham, so just one place to travel to regularly, and one of my daughters also there.

Of course this also means I will be leaving many dear colleagues and friends at Lancaster, but I do expect to continue to work with many and am likely to retain a formal or informal role there for some time.

As well as moving institutions I am also further reducing my percentage of academic time — typically I’ll be just one day a week academic.  So, apologies in advance if my email responses becomes even more sporadic and I turn down (or fail to answer ) requests for reviews, PhD exams, etc.

Although moving institutions, I will, of course, continue to live up in Tiree (wild and windy, but, at the moment, so is everywhere!), so will still be travelling up and down the country; I’ll wave as I pass!

… and there will be another Tiree Tech Wave in March

So is it as if it had never been?  I there anything left behind?  There are a few sheets of Magic Whiteboard on the walls, that I left so that those visiting the Rural Centre in the coming weeks can see something of what we were doing, and there are used teabags and fish-and-chip boxes in the bin, but few traces.

We trod lightly, like the agriculture of the island, where Corncrake and orchid live alongside sheep and cattle.

Some may have heard me talk about the way design is like a Spaghetti Western. In the beginning of the film Clint Eastwood walks into the town, and at the end walks away.  He does not stay, happily ever after, with a girl on his arm, but leaves almost as if nothing had ever happened.

But while he, like the designer, ultimately leaves, things are not the same.  The Carson brothers who had the town in fear for years lie dead in their ranch at the edge of town, the sharp tang of gunfire still in the air and the buzz of flies slowly growing over the elsewise silent bodies.  The crooked major, who had been in the pocket of the Carson brothers, is strapped over a mule heading across the desert towards Mexico, and not a few wooden rails and water buts need to be repaired.  The job of the designer is not to stay, but to leave, but leave change: intervention more than invention.

But the deepest changes are not those visible in the bullet-pocked saloon door, but in the people.  The drunk who used to sit all day at the bar, has discovered that he is not just a drunk, but he is a man, and the barmaid, who used to stand behind the bar has discovered that she is not just a barmaid, but she is a woman.

This is true of the artefacts we create and leave behind as designers, but much more so of the events, which come and go through our lives.  It is not so much the material traces they leave in the environment, but the changes in ourselves.

I know that, as the plane and ferry left with those last participants, a little of myself left with them, and I know many, probably all, felt a little of themselves left behind on Tiree.  This is partly abut the island itself; indeed I know one participant was already planning a family holiday here and another was looking at Tiree houses for sale on RightMove!  But it was also the intensity of five, sometimes relaxed, sometimes frenetic, days together.

So what did we do?

There was no programme of twenty minute talks, no keynotes or demo, indeed no plan nor schedule at all, unusual in our diary-obsessed, deadline-driven world.

Well, we talked.  Not at a podium with microphone and Powerpoint slides, but while sitting around tables, while walking on the beach, and while standing looking up at Tilly, the community wind turbine, the deep sound of her swinging blades resonating in our bones.  And we continued to talk as the sun fell and the overwhelmingly many stars came out , we talked while eating, while drinking and while playing (not so expertly) darts.

We met people from the island those who came to the open evening on Saturday, or popped in during the days, and some at the Harvest Service on Sunday.  We met Mark who told us about the future plans for Tiree Broadband, Jane at PaperWorks who made everything happen, Fiona and others at the Lodge who provided our meals, and many more. Indeed, many thanks to all those on the island who in various ways helped or made those at TTW feel welcome.

We also wrote.  We wrote on sheets of paper, notes and diagrams, and filled in TAPT forms for Clare who was attempting unpack our experiences of peace and calmness in the hope of designing computer systems that aid rather than assault our solitude.  Three large Magic Whiteboard sheets were entitled “I make because …”, “I make with …”, “I make …” and were filled with comments.  And, in these days of measurable objectives, I know that at least a grant proposal, book chapter and paper were written during the long weekend; and the comments on the whiteboards and experiences of the event will be used to create a methodological reflection of the role of making in research which we’ll put into Interfaces and the TTW web site.

We moved.  Walking, throwing darts, washing dishes, and I think all heavily gesturing with our hands while taking.  And became more aware of those movements during Layda’s warm-up improvisation exercises when we mirrored one another’s movements, before using our bodies in RePlay to investigate issues of creativity and act out the internal architecture of Magnus’ planned digital literature system.

We directly encountered the chill of wind and warmth of sunshine, the cattle and sheep, often on the roads as well as in the fields.  We saw on maps the pattern of settlement on the island and on display boards the wools from different breeds on the island. Some of us went to the local historical centre, An Iodhlann [[ http://www.aniodhlann.org.uk/ ]], to see artefacts, documents and displays of the island in times past, from breadbasket of the west of Scotland to wartime airbase.

We slept.  I in my own bed, some in the Lodge, some in the B&B round the corner, Matjaz and Klem in a camper van and Magnus – brave heart – in a tent amongst the sand dunes.  Occasionally some took a break and dozed in the chairs at the Rural Centre or even nodded off over a good dinner (was that me?).

We showed things we had brought with us, including Magnus’ tangle of wires and circuit boards that almost worked, myself a small pack of FireFly units (enough to play with I hope in a future Tech Wave), Layda’s various pieces she had made in previous tech-arts workshops, Steve’s musical instrument combining Android phone and cardboard foil tube, and Alessio’s impressively modified table lamp.

And we made.  We do after all describe this as a making event!  Helen and Claire explored the limits of ZigBee wireless signals.  Several people contributed to an audio experience using proximity sensors and Arduino boards, and Steve’s CogWork Chip: Lego and electronics, maybe the world’s first mechanical random-signal generator.  Descriptions of many of these and other aspects of the event will appear in due course on the TTW site and participants’ blogs.

But it was a remark that Graham made as he was waiting in the ferry queue that is most telling.  It was not the doing that was central, the making, even the talking, but the fact that he didn’t have to do anything at all.  It was the lack of a plan that made space to fill with doing, or not to do so.

Is that the heart?  We need time and space for non-doing, or maybe even un-doing, unwinding tangles of self as well as wire.

There will be another Tiree Tech Wave in March/April, do come to share in some more not doing then.

Who was there:

• Alessio Malizia – across the seas from Madrid, blurring the boundaries between information, light and space
• Helen  Pritchard – artist, student of innovation and interested in cows
• Claire  Andrews – roller girl and researching the design of assistive products
• Clare  Hooper – investigating creativity, innovation and a sprinkling of SemWeb
• Magnus  Lawrie – artist, tent-dweller and researcher of digital humanities
• Steve Gill – designer, daredevil and (when he can get me to make time) co-authoring book on physicality TouchIT
• Graham Dean – ex-computer science lecturer, ex-businessman, and current student and auto-ethnographer of maker-culture
• Steve Foreshaw - builder, artist, magician and explorer of alien artefacts
• Matjaz Kljun – researcher of personal information and olive oil maker
• Layda Gongora – artist, curator, studying improvisation, meditation and wild hair
• Alan Dix – me

I recall my own interest in the origins of language began young, as a seven year old over breakfast one day, asking whether ‘night, was a contraction of ‘no light’.  While this was an etymological red herring, it is very much the kind of change that Deutscher documents in detail showing the way a word accretes beginnings and ending through juxtaposition of simpler words followed by erosion of hard to pronounce sounds.

One of my favourites examples was the French “aujourd’hui”.  The word ‘hui, was Old French for ‘today’, but was originally Latin “hoc die”, “(on) this day”. Because ‘hui’ is not very emphatic it became “au jour d’hui”, “on the day of this day” , which contracted to the current ‘aujourd’hui’. Except now to add emphasis some French speakers are starting to say “au jour aujourd’hui”, “on the day on the day of this day”!  This reminds me of Longsleddale in the Lake District (inspiration for Postman Pat‘s Greendale),  a contraction of “long sled dale”, which literally means “long valley valley” from Old English “slaed” meaning “valley” … although I once even saw something suggesting that ‘long’ itself in the name was also “valley” in a different language!

Deutscher gives many more prosaic examples where words meaning ‘I’, ‘you’, ‘she’ get accreted to verbs to create the verb endings found in languages such as French, and how prepositions (themselves metaphorically derived from words like ‘back’) were merged with nouns to create the complex case endings of Latin.

However, the most complex edifice, which Deutscher returns to repeatedly, is that of the Semitic languages with a template system of vowels around three-consonant roots, where the vowel templates change the meaning of the root.  To illustrate he uses the (fictional!) root ‘sng’ meaning ‘to snog’ and discusses how first simple templates such as ‘snug’ (“I snogged”) and then more complex constructions such as ‘hitsunnag’ (“he was made to snog himself”) all arose from simple processes of combination, shortening and generalisation.

“The Unfolding of Language” begins with the 19th century observation that all languages seem to be in a process of degeneration where more complex  forms such as the Latin case system or early English verb endings are progressively simplified and reduced. The linguists of the day saw all languages in a state of continuous decay from an early linguistic Golden Age. Indeed one linguist, August Schleicher, suggested that there was a process where language develops until it is complex enough to get things done, and only then recorded history starts, after which the effort spent on language is instead spent in making history.

As with geology, or biological evolution, the modern linguist rejects this staged view of the past, looking towards the Law of Uniformitarianism, things are as they have always been, so one can work out what must have happened in the pre-recorded past by what is happening now.  However, whilst generally finding this convincing, throughout the book I had a niggling feeling that there is a difference.  By definition, those languages for which we have written records are those of large developed civilisations, who moreover are based on writing. Furthermore I am aware that for biological evolution small isolated groups (e.g. on islands or cut off in valleys) are particularly important for introducing novelty into larger populations, and I assume the same would be true of languages, but somewhat stultified by mass communication.

Deutscher does deal with this briefly, but right at the very end in a short epilogue.  I feel there is a whole additional story about the interaction between culture and the grammatical development of language.  I recall in school a teacher explained how in Latin the feminine words tended to belong to the early period linked to agriculture and the land, masculine words for later interests in war and conquest, and neuter for the still later phase of civic and political development. There were many exceptions, but even this modicum of order helped me to make sense of what otherwise seemed an arbitrary distinction.

The epilogue also mentions that the sole exception to the ‘decline’ in linguistic complexity is Arabic with its complex template system, still preserved today.

While reading the chapters about the three letter roots, I was struck by the fact that both Hebrew an Arabic are written as consonants only with vowels interpolated by diacritical marks or simply remembered convention (although Deutscher does not mention this himself). I had always assumed that this was like English where t’s pssble t rd txt wth n vwls t ll. However, the vowels are far more critical for Semitic languages where the vowel-less words could make the difference between “he did it” and “it will be done to him”.  Did this difference in writing stem from the root+template system, or vice versa, or maybe they simply mutually reinforced each other?

The other factor regarding Arabic’s remarkable complexity must surely be the Quran. Whereas the Bible was read for a over a millennium in Latin, a non-spoken language, and later translated focused on the meaning; in contrast there is a great emphasis on the precise form of the Quran together with continuous lengthy recitation.  As the King James Bible has been argued to have been a significant influence on modern English since the 17th century, it seems likely the Quran has been a factor in preserving Arabic for the last 1500 years.

Early in “The Unfolding of Language” Deutscher dismisses attempts to look at the even earlier prehistoric roots of language as there is no direct evidence. I assume that this would include Mithin’s “The Singing Neanderthals“, which I posted about recently. There is of course a lot of truth in this criticism; certainly Mithin’s account included a lot of guesswork, albeit founded on paleontological evidence.  However, Deutscher’s own arguments include extrapolating to recent prehistory. These extrapolations are based on early written languages and subsequent recorded developments, but also include guesswork between the hard evidence, as does the whole family-tree of languages.  Deutscher was originally a Cambridge mathematician, like me, so, perhaps unsurprisingly, I found his style of argument convincing. However, given the foundations on Uniformitarianism, which, as noted above, is at best partial when moving from history to pre-history, there seems more of  a continuum rather than sharp distinction between the levels of interpretation and extrapolation in this book and Mithin’s.

Deutscher’s account seeks to fill in the gap between the deep prehistoric origins of protolanguage (what Deutscher’s calls ‘me Tarzan’ language) and its subsequent development in the era of media-society (starting 5000BC with extensive Sumerian writing). Rather than seeing these separately, I feel there is a rich account building across various authors, which will, in time, yield a more complete view of our current language and its past.

“The Singing Neanderthal”, as its title suggests, is about the role of music in the evolutionary development of the modern human. We all seem to be born with an element of music in our heart, and Mithin seeks to understand why this is so, and how music is related to, and part of the development of, language. Mithin argues that elements of music developed in various later hominids as a form of primitive communication², but separated from language in homo sapiens when music became specialised to the communication of emotion and language to more precise actions and concepts.

The book ‘explains’ various known musical facts, including the universality of music across cultures and the fact that most of us do not have perfect pitch … even though young babies do (p77). The hard facts of how things were for humans or related species tens or hundreds of thousands of years ago are sparse, so there is inevitably an element of speculation in Mithin’s theories, but he shows how many, otherwise disparate pieces of evidence from palaeontology, psychology and musicology make sense given the centrality of music.

Whether or not you accept Mithin’s thesis, the first part of the book provides a wide ranging review of current knowledge about the human psychology of music. Coincidentally, while reading the book, there was an article in the Independent reporting on evidence for the importance of music therapy in dealing with depression and aiding the rehabilitation of stroke victims³, reinforcing messages from Mithin’s review.

The topic of “The Singing Neanderthal” is particularly close to my own heart as my first personal forays into evolutionary psychology (long before I knew the term, or discovered Cosmides and Tooby’s work), was in attempting to make sense of human limits to delays and rhythm.

Those who have been to my lectures on time since the mid 1990s will recall being asked to first clap in time and then swing their legs ever faster … sometimes until they fall over! The reason for this is to demonstrate the fact that we cannot keep beats much slower than one per second⁴, and then explain this in terms of our need for a mental ‘beat keeper’ for walking and running. The leg shaking is to show how our legs, as a simple pendulum, have a natural frequency of around 1Hz, hence determining our slowest walk and hence need for rhythm.

Mithin likewise points to walking and running as crucial in the development of rhythm, in particular the additional demands of bipedal motion (p150). Rhythm, he argues, is not just about music, but also a shared skill needed for turn-taking in conversation (p17), and for emotional bonding.

In just the last few weeks, at the HCI conference in Newcastle, I learnt that entrainment, when we keep time with others, is a rare skill amongst animals, almost uniquely human. Mithin also notes this (p206), with exceptions, in particular one species of frog, where the males gather in groups to sing/croak in synchrony. One suggested reason for this is that the louder sound can attract females from a larger distance. This cooperative behaviour of course acts against each frog’s own interest to ‘get the girl’ so they also seek to out-perform each other when a female frog arrives. Mithin imagines that similar pressures may have sparked early hominid music making. As well as the fact that synchrony makes the frogs louder and so easy to hear, I wonder whether the discerning female frogs also realise that if they go to a frog choir they get to chose amongst them, whereas if they follow a single frog croak they get stuck with the frog they find; a form of frog speed dating?

Mithin also suggests that the human ability to synchronise rhythm is about ‘boundary loss’ seeing oneself less as an individual and more as part of a group, important for early humans about to engage in risky collaborative hunting expeditions. He cites evidence of this from the psychology of music, anthropology, and it is part of many people’s personal experience, for example, in a football crowd, or Last Night at the Proms.

This reminds me of the experiments where a rubber hand is touched in time with touching a person’s real hand; after a while the subject starts to feel as if the rubber hand is his or her own hand. Effectively our brain assumes that this thing that correlates with feeling must be part of oneself⁵. Maybe a similar thing happens in choral singing, I voluntarily make a sound and simultaneously everyone makes the sound, so it is as if the whole choir is an extension of my own body?

Part of the neurological evidence for the importance of group music making concerns the production of oxytocin. In experiments on female prairie voles that have had oxytocin production inhibited, they engage in sex as freely as normal voles, but fail to pair bond (p217). The implication is that oxytocin’s role in bonding applies equally to social groups. While this explains a mechanism by which collaborative rhythmic activities create ‘boundary loss’, it doesn’t explain why oxytocin is created through rhythmic activity in the first place. I wonder if this is perhaps to do with bipedalism and the need for synchronised movement during face-to-face copulation, which would explain why humans can do synchronised rhythms whereas apes cannot. That is, rhythmic movement and oxytocin production become associated for sexual reasons and then this generalises to the social domain. Think again of that chanting football crowd?

I should note that Mithin also discusses at length the use of music in bonding with infants, as anyone who has sung to a baby knows, so this offers an alternative route to rhythm & bonding … but not one that is particular to humans, so I will stick with my hypothesis

Sexual selection is a strong theme in the book, the kind of runaway selection that leads to the peacock tail. Changing lifestyles of early humans, in particular longer periods looking after immature young, led to a greater degree of female control in the selection of partners. As human size came close to the physical limits of the environment (p185), Mithin suggests that other qualities had to be used by females to choose their mate, notably male singing and dance – prehistoric Saturday Night Fever.

As one evidence for female mate choice, Mithin points to the overly symmetric nature of hand axes and imagines hopeful males demonstrating their dexterity by knapping ever more perfect axes in front of admiring females (p188). However, this brings to mind Calvin’s “Ascent of Mind“, which argues that these symmetric, ovoid axes were used like a discus, thrown into the midst of a herd of prey to bring one down. The two theories for axe shape are not incompatible. Calvin suggests that the complex physical coordination required by axe throwing would have driven general brain development. In fact these forms of coordination, are not so far from those needed for musical movement, and indeed expert flint knapping, so maybe it was this skills that were demonstrated by the shaping of axes beyond that immediately necessary for purpose.

Mithin’s description of the musical nature of mother-child interactions also brought to mind Broomhall’s “Eternal Child“. Broomhall ‘s central thesis is that humans are effectively in a sort of arrested development with many features, not least our near nakedness, characteristic of infants. Although it was not one of the points Broomhall makes, his arguments made sense to me in terms of the mental flexibility that characterises childhood, and the way this is necessary for advanced human innovation; I am always encouraging students to think in a more childlike way. If Broomhall’s theories were correct, then this would help explain how some of the music making more characteristic of mother-infant interactions become generalised to adult social interactions.

I do notice an element of mutual debunking amongst those writing about richer cognitive aspects of early human and hominid development. I guess a common trait in disciplines when evidence is thin, and theories have to fill a lot of blanks. So maybe Mithin, Calvin and Broomhall would not welcome me bringing their respective contributions together! However, as in other areas where data is necessarily scant (such as sub-atomic physics), one does feel a developing level of methodological rigour, and the fact that these quite different theoretical approaches have points of connection, does suggest that a deeper understanding of early human cognition, while not yet definitive, is developing.

In summary, and as part of this wider unfolding story, “The Singing Neanderthal” is an engaging and entertaining book to read whether you are interested in the psychological and social impact of music itself, or the development of the human mind.

… and I have another of Mithin’s books in the birthday pile, so looking forward to that too!

1. See particularly my essay on the role of imagination in bringing together our different forms of ‘specialised intelligence’. “The Prehistory of the Mind” highlighted the importance of this ‘cognitive fluidity’, linking social, natural and technological thought, but lays this largely in the realm of language. I would suggest that imagination also has this role, creating a sort of ‘virtual world’ on which different specialised cognitive modules can act (see “imagination and rationality“). [back]
2. He calls this musical communication system Hmmmm in its early form – Holistic, Multiple-Modal, Manipulative and Musical, p138 – and later Hmmmmm – Holistic, Multiple-Modal, Manipulative, Musical and Mimetic, p221. [back]
3. “NHS urged to pay for music therapy to cure depression“, Nina Lakhani, The Independent, Monday, 1 August 2011 [back]
4. Professional conductors say 40 beats per minute is the slowest reliable beat without counting between beats. [back]
5. See also my previous essay on “driving as a cyborg experience“. [back]

The shortness is a boon that I wish many writers would follow (including me). As with so many single issue books (e.g. Blink), there is s slight tendency to over-sell the main argument, but this is forgiveable in a short delightful book, in a way that it isn’t in 350 pages of less graceful prose.

I know I have a tendency, which can be confusing or annoying, to give, paradoxically for fear of misunderstanding, the caveat before the main point. Still, despite knowing this, in the early chapters I did find myself occasionally bristling at Maeda’s occasional overstatement (although in accordance with simplicity, never hyperbole).

One that particularly caught my eye was Maeda’s contrast of the MIT engineer’s RFTM (Read The F*cking Manual) with the “designer’s approach” to:

marry function with form to create intuitive experiences that we understand immediately.

Although in principle I agree with the overall spirit, and am constantly chided by Fiona for not reading instructions³, the misguided idea that everything ought to ‘pick up and use’ has bedeviled HCI and user interface design for at least the past 20 years. Indeed this is the core misconception about Heidegger’s hammer example that I argued against in a previous post “Struggling with Heidegger“. In my own reading notes, my comment is “simple or simplistic!” … and I meant here the statement not the resulting interfaces, although it could apply to both.

It has always been hard to get well written documentation, and the combination of single page ‘getting started’ guides with web-based help, which often disappears when the web site organisation changes, is an abrogation of responsibility by many designers. Not that I am good at this myself. Good documentation is hard work. It used to be the coders who failed to produce documentation, but now the designers also fall into this trap of laziness, which might be euphemistically labelled ‘simplicity’⁴.

Personally, I have found that the discipline of documenting (in the few times I have observed it!) is in fact a great driver of simple design. Indeed I recall a colleague, maybe Harold Thimbleby⁵, once suggested that documentation ought to be written before any code is written, precisely to ensure simple use.

Some years ago I was reading a manual (for a Unix workstation, so quite a few years ago!) that described a potentially disastrous shortcoming of a the disk sync command (which could have corrupted the disk). Helpfully the manual page included a suggestion of how to wrap sync in scripts that prevented the problem. This seemed to add insult to injury; they knew there was a serious problem, they knew how to fix it … and they didn’t do it. Of course, the reason is that manuals are written by technical writers after the code is frozen.

In contrast, I was recently documenting an experimental API⁶ so that a colleague could use it. As I wrote the documentation I found parts hard to explain. “It would be easier to change the code”, I thought, so I did so. The API, whilst still experimental, is now a lot cleaner and simpler.

Coming back to Maena after a somewhat long digression (what was that about simplicity and brevity?). While I prickled slightly at a few statements, in fact he very clearly says that the first few concrete ‘laws’ are the simpler (and if taken in their own simplistic), the later laws are far more nuanced and suggest deeper principles. This includes law 5 “differences: simplicity and complexity need each other”, which suggest that one should strive for a dynamic between simplicity and complexity. This echoes the emphasis on texture I often advocate when talking with students; whether in writing, presenting or in experience design it is often the changes in voice, visual appearance, or style which give life.

the simplest interface?

I wasn’t convinced by Maeda’s early claim that simple designs were simpler and cheaper to construct.  Possibly true for physical prodcuts, but rarely so for digital interfaces, where more effort is typically needed in code to create simpler user interfaces.  However, again this was something that was revisited later, especially in the context of more computationally active systems (“law 8, in simplicity we trust”), where he contrasts “how much do you need to know about a system?” with “how much does the system know about you?”.  The former is the case of more traditional passive systems, whereas more ‘intelligent’ systems such as Amazon recommendations (or even Facebook news feed) favour the latter.  This is very similar to the principles for incidental and low-intention interaction that I have discussed in the past⁷.

Finally “The Laws of Simplicity” is beautifully designed in itself.  It includes  many gems not least those arising from Maeda’s roots in Japanese design culture, including aichaku, the “sense of attachment one can feel for an artefact” (p.69) and omakase meaning “I leave it to you”, which asks the sushi chef to create a meal especially for you (p.76).  I am perhaps too much of a controller to feel totally comfortable with the latter, but Maeda’s book certainly inspires the former.

1. In fact there are 108 pages in the main text, but 9 of these are full page ‘law/chapter’ frontispieces, so 99 real pages.  However, if you include the 8 page introduction that gives 107 … so even the 100 page cap is perhaps a more subtle concept than a strict count. [back]
2. See his full 10 laws of simplicity at lawsofsimplicity.com [back]
3. My guess is that the MIT engineers didn’t read the manuals either. [back]
4. Apple is a great — read poor — example here as it relies on keen technofreaks to tell others about the various hidden ways to do things — I guess creating a Gnostic air to the devices. [back]
5. Certainly Harold was a great proponent of ‘live’ documentation, both Knuth’s literate programming and also documentation that incorporated calculated input and output, rather like dexy, which I reported after last autumn’s Web Art/Science camp. [back]
6. In fairness, the API had been thrown together in haste for my own use. [back]
7. See ‘incidental interaction” and HCI book chapter 18. [back]

As well as time at Talis offices in B’ham and at Lancs (including exam board week), travels have taken me to Pisa for a workshop on ‘Supportive User Interfaces’, to Koblenz for Web Science conference giving a talk on embodiment issues and a poster on web-scale reasoning , to Newcastle for British HCI conference doing a talk on fridge, to Nottingham to give a talk on extended episodic experience, and back to Lancs for a session on creativity! Why can’t I be like sensible folks and talk on one topic!

Supportive User Interfaces

Monday 13th June I attended a workshop in Pisa on “Supportive User Interfaces“, which includes interfaces that adapt in various ways to users.  The majority of people there were involved in various forms of model-based user interfaces in which various models of the task, application and interaction are used to generate user interfaces on the fly. W3C have had a previous group in this area; Dave Raggett from w3c was at the workshop and it sounds like there will be a new working group soon.  This clearly has strong links to various forms of ‘meta-level’ representations of data, tasks, etc..  My own contribution started the day, framing the area, focusing partly on reasons for having more ‘meta-level’ interfaces including social empowerment, and partly on the principles/techniques that need to be considered at a human level.

Also on Monday was a meeting of IFIP Working Group 2.7/13.4. IFIP is the UNESCO founded pan-national agency that national computer societies such as as the BCS in the UK and ACM and IEEE Computer in the US belong to.  Working Group 2.7/13.4 is focused on the engineering of user interfaces.  I had been actively involved in the past, but have had many years’ lapse.  However, this seemed a good thing to re-engage with with my new Talis hat on!

SUI: paper:

• Opening the Box: Meta-level Interfaces Needs and Solutions

Web Science Conference in Koblenz

Jaime Teevan from Microsoft gave the opening keynote at WebSci 2011.  I know her from her earlier work on personal information management, but her recent work and keynote was about work on analysing and visualising changes in web pages.  Web page changes are also analysed alongside users re-visitation patterns; by looking at the frequency of re-visitation Jaime and her colleagues are able to identify the parts of pages that change with similar frequency, helping them, inter alia, to improve search ranking.

Had many great conversations, some with people I know previously (e.g. the Southampton folks), but also new, including the group at Troy that do lots of work with data.gov.  I was particularly interested in some work using content matching to look for links between otherwise unlinked (or only partly inter-linked) datasets.  Also lots of good presentations including one on trust prediction and a fantastic talk by Mark Bernstein from Eastgate, which he delivered in blank verse!

My own contribution included the poster that Dave@Talis prepared, which was on the web-scale spreading activation work in collaboration with Univ. Athens.  Quite a niche area in a multi-disciplinary conference, so didn’t elicit quite the interest of the social networking posters, but did lead to a small number of in depth discussions.

In addition I gave talk on the more cognitive/philosophical issues when we start to use the web as an external extension to / replacement of memory, including its impact on education.  Got some good feedback from this.

Closing keynote was from Barry Wellman, the guy who started social network analysis way before they were on computers.  At one point he challenged the Dunbar number¹. I wondered whether this was due to cognitive extension with address books etc., but he didn’t seem to think so; there is evidence that some large circles predate web (although maybe not physical address books).  Made me wonder about itinerant tradesmen, tinkers, etc., even with no prostheses. Maybe the numbers sort of apply to any single content, but are repeated for each new context?

WebSci papers:

• Spreading Activation for Web Scale Reasoning: Promise and Problems
• A Shifting Boundary: the dynamics of internal cognition and the web as external representation

The HCI Conference – Newcastle

I attended the British HCI conference in Newcastle. This was the 25th conference, and as my very first academic paper in computing² was at the first BHCI in 1984, I was pleased to be there at this anniversary.  The paper I was presenting was a retrospective on vfridge, a social networking site dating back to 1999/2000, it seemed an historic occasion!

As is always the case presentations were all interesting. Strictly BHCI is a ‘second tier’ conference compared with CHI, but why is it that the papers are always more interesting, that I learn more?  It is likely that a fair number of papers were CHI rejects, so it should be the other way round – is it that selectivity and ‘quality’ inevitably become conservative and boring?

Gregory Abowd gave the closing keynote. It was great to see Gregory again, we meet too rarely.  The main focus of his keynote was on three aspects of research: novelty, value and reliability and how his own work had moved within this space over the years.  In particular having two autistic sons has led him in directions he would never have considered, and this immediately valuable work has also created highly novel research. Novelty and value can coexist.

Gregory also reflected on the BHCI conference as it was his early academic ‘home’ when he did his PhD and postdoctoral here in the late 1980′s.  He thought that it could be rather than, as with many conferences, a second best to getting a CHI paper, instead a place for (not getting the quote quite perfect) “papers that should get into CHI”, by which he meant a proving ground for new ideas that would then go on to be in CHI.

However I initially read the quote differently. BHCI always had a broader concept of HCI compare with CHI’s quite limited scope. That is BHCI as a place that points the way for the future of HCI, just as it was the early nurturing place of MobileHCI.  However CHI has now become much broader in it’s own conception, so maybe this is no longer necessary. Indeed at the althci session the organisers said that their only complaint was that the papers were not ‘alt’ enough – that maybe ‘alt’ had become mainstream. This prompted Russell Beale to suggest that maybe althci should now be real science such as replication!

Gregory also noted the power of the conference as a meeting ground. It has always been proud of the breadth of international attendance, but perhaps it is UK saturation that should be it’s real measure of success.  Of course the conference agenda has become so full and international travel so much cheaper than it was, so there is a tendency to  go to the more topic specific international conferences and neglect the UK scene.  This is compounded by the relative dearth of small UK day workshops that used to be so useful in nurturing new researchers.

I feel a little guilty here as this was the first BHCI I had been to since it was in Lancaster in 2007 … as Tom McEwan pointed out I always apologise but never come! However, to be fair I have also only been twice to CHI in the last 10 years, and then when it was in Vienna and Florence. I have just felt too busy, so avoiding conferences that I did not absolutely have to attend.

In response to Gregory’s comments, someone, maybe Tom, mentioned that in days of metrics-based research assessment there was a tendency to submit one’s best work to those venues likely to achieve highest impact, hence the draw of CHI. However, I have hardly ever published in CHI and I think only once in TOCHI, yet, according to Microsoft Research, I am currently the most highly cited HCI researcher over the last 5 years … So you don’t have to publish in CHI to get impact!

And incidentally, the vfridge paper had NOT been submitted to CHI, but was specially written for BHCI as it seemed the fitting place to discuss a thoroughly British product

vfridge paper:

• Anatomy of an Early Social Networking Site

Nottingham MRL

I was at Mixed Reality Lab in Nottingham for Joel Fischer‘s PhD viva and while there did a seminar the afternoon on “extended episodic experience” based on Haliyana Khalid‘s PhD work and ideas that arose from it. Basically, whereas ‘user experience’ has become a big issue most of the work is focused on individual ‘experiences’ whereas much of life consists of ongoing series of experiences (episodes) which together make up the whole experience of interacting with a person or place, following a band, etc.

I had obviously not done a good enough job at wearing Joel down with difficult questions in the PhD viva in the morning as he was there in the afternoon to ask difficult questions back of his own

Docfest – Digital Economy Summer School

The last major event was Docfest, which brought together the PhD students from the digital economy centres from around the country. Not sure of the exact count but just short of 150 participants I think. They come from a wide variety of backgrounds, business, design, computing, engineering, and many are mature students with years of professional experience behind them.

This looked like being a super event, unfortunately I was only able to attend for a day  However, I had a great evening at the welcome event talking with many of the students and even got to ride in Steve Forshaw‘s Sinclair C5!

My contribution to the event was running the first morning session on ‘creativity’. Surprise, surprise this started with a bad ideas session, but new for me too as the largest group I’ve run in the past has been around 30.  There were a number of local Highwire students acting as facilitators for the groups, so I had only to set them off and observe results . At the end of the morning I gave some the theoretical background to bad ideas as a method and in understanding (aspects of) creativity more widely.

Other speakers at the event included Jane Prophet, Chris Csikszentmihalyi and Chris Bonnington, so was sad to miss them; although I did get a fascinating chat with Jane over breakfast in the hotel hearing about her new projects on arts and neural imaging, and on how repetitious writing induces temporary psychosis … That is why the teachers give lines, to send the pupils bonkers!

1. The idea that there are fundamental cognitive limits on social groups with different sized circles family~6, extended family~20, village~60, large village~200 [back]
2. I had published previously in agricultural engineering. [back]

Of course if ou are a touch typist things may be different.

I was especially struck by the image of the keyboard on the iPad, which on the ‘F’ and ‘J’ keys has an image of the tiny tactile bar that allows easy finger positioning by touch alone.  However, this is of course a flat screen and I am feeling I am in some sort of surreal world.

Along with Fitts’ Law and Miller’s 7+/-2 this “five is enough” must be among the most widely cited, and most widely misunderstood and misused aphorisms of HCI and usability. Indeed, I feel that this post belongs more in ‘Myth Busters” than in my blog.

So, do I think five is enough? Of course, the answer is (annoyingly), “it depends”, sometimes, yes, five is enough, but sometimes, fewer: one appropriate user, or maybe no users at all, and sometimes you need more, maybe many more: 20, 100, 1000. But even when five is enough, the reasons are usually not directly related to Nielsen and Landauer’s original paper, which people often cite (although rarely read) and Nielsen’s “Why You Only Need to Test With 5 Users” alert box (probably more often actually read … well at least the title).

The “it depends” is partly dependent on the particular circumstances of the situation (types of people involved, kind of phenomenon, etc.), and partly on the kind of question you want to ask. The latter is the most critical issue, as if you don’t know what you want to know, how can you know how many users you need?

There are several sorts of reasons for doing some sort of user study/experiment, several of which may apply:

1. To improve a user interface (formative evaluation)

2. To assess whether it is good enough (summative evaluation)

3. To answer some quantitative question such as “what % of users are able to successfully complete this task”

4. To verify or refute some hypothesis such as “does eye gaze follow Fitts’ Law”

5. To perform a broad qualitative investigation of an area

6. To explore some domain or the use of a product in order to gain insight

It is common to see HCI researchers very confused about these distinctions, and effectively perform formative/summative evaluation in research papers (1 or 2) where in fact one of 3-6 is what is really needed.

I’ll look at each in turn, but first to note that, to the extent there is empirical evidence for ‘five is enough”, it applies to the first of these only.

I dealt with this briefly in my paper “Human–Computer Interaction: a stable discipline, a nascent science, and the growth of the long tail” in the John Long Festschrift edition of IwC, and quote here:

In the case of the figure of five users, this was developed based on a combination of a mathematical model and empirical results (Nielsen and Landauer 1993). The figure of five users is:

(i) about the optimal cost/benefit point within an iterative development cycle, considerably more users are required for summative evaluation or where there is only the opportunity for a single formative evaluation stage;

(ii) an average over a number of projects and needs to be assessed on a system by system basis; and

(iii) based on a number of assumptions, in particular, independence of faults, that are more reasonable for more fully developed systems than for early prototypes, where one fault may mask another.

We’ll look at this in more detail below, but critically, the number ’5′ is not a panacea, even for formative evaluation.

As important as the kind of question you are asking, are the kind of users you are using. So much of modern psychology is effectively the psychology of first year psychology undergraduates (in the in 1950s it was male prisoners). Is this representative? Does this matter? I’ll return to this at the end, but first of all look briefly at each kind of question.

Finally, there is perhaps the real question “will the reviewers of my work think five users is enough” — good publications vs. good science. The answer is that they will certainly be as influenced by the Myth of Five Users as you are, so do good science … but be prepared to need to educate your reviewers too!

formative evaluation – prototyping cycle

As noted formative evaluation was the scope of Nielsen and Landauer’s early work in 1993 that was then cited by Nielsen in his Alert Box in 2000, and which has now developed mythic status in the field.

The 1993 paper was assuming a context of iterative development where there would be many iterations, and asking how many users should be used per iteration, that is how many users should you test before fixing the problems found by those users, and then performing another cycle of user testing, and another. That is, in all cases they considered, the total number of users involved would be far more than five, it is just the number used in each iteration that was lower.

In order to calculate the optimal number of subjects to use per iteration, they looked at:

(i) the cost of performing a user evaluation

(ii) the number of new faults found (additional users will see many of the same faults, so there are diminishing returns)

(iii) the cost of a redevelopment cycle

All of these differ depending on the kind of project, so Nielsen and Landauer looked at a range of projects of differing levels of complexity. By putting them together, and combining with simple probabilistic models of bug finding in software, you can calculate an optimal number of users per experiment.

They found that, depending on the project, the statistics and costs varied and hence the optimal number of users/evaluators (between 7 and 21), with, on the whole, more complex projects (with more different kinds of bugs and more costly redevelopment cycles) having a higher optimal number than simpler projects. In fact all the numbers are larger than five, but five was the number in Nielsen’s earlier discount engineering paper, so the paper did some additional calculations that yielded a different kind of (lower) optimum (3.2 users — pity the last 0.2 user), with five somewhere between 7 and 3 … and a myth was born!

Today, with Web 2.0 and ‘perpetual beta’, agile methods and continuous deployment reduce redevelopment costs to near zero, and so Twidale and Marty argue for ‘extreme evaluation‘ where often one user may be enough (see also my IwC  paper).

The number also varies through the development process; early on, one user (indeed using it yourself) will find many, many faults that need to be fixed. Later faults become more obscure, or maybe only surface after long-term use.

Of course, if you use some sort of expert or heuristic evaluation, then the answer may be no real users at all!

And anyway all of this is about ‘fault finding’, usability is not about bug fixing but making things better, it is not at all clear how useful, if at all, literature on bug fixing is for creating positive experiences.

summative evaluation – is it good enough to release

If you are faced with a product and want to ask “is it good enough?” (which can mean, “are there any usability ‘faults’?”, or, “do people want to use it?”), then five users is almost certainly not enough. To give yourself any confidence of coverage of the kinds of users and kinds of use situations, you may need tens or hundreds of users, very like hypothesis testing (below).

However, the answer here may also be zero users. If the end product is the result of a continuous evaluation process with one, five or some other number of users per iteration, then the number of users who have seen the product during this process may be sufficient, especially if you are effectively iterating towards a steady state where few or no new faults are found per iteration.

In fact, even when there has been a continuous process, the need for long-term evaluation becomes more critical as the development progresses, and maybe the distinction between summative and late-stage formative is moot.

But in the end there is only one user you need to satisfy — the CEO … ask Apple.

quantitative questions and hypothesis testing

(Note, there are real numbers here, but if you are a numerophobe never fear, the next part will go back to qualitative issues, so bear with it!)

Most researchers know that “five is enough” does not apply in experimental or quantitative studies … but that doesn’t always stop them quoting numbers back!

Happily in dealing with more quantitative questions or precise yes/no ones, we can look to some fairly solid statistical rules for the appropriate number of users for assessing different kinds of effects (but do note “the right kind of users” below). And yes, very, very occasionally five may be enough!

Let’s imagine that our hypothesis is that a behaviour will occur in 50% of users doing an experiment. With five users, the probability that we will see this behaviour in at least one user is 1 in 32, which is approximately 3%. That is if we do not observe the behaviour at all, then we have a statistically significant result at 5% level (p<0.05) and can reject the hypothesis.

Note that there is a crucial difference between a phenomenon that we expect to see in about 50% of user iterations (i.e. the same user will do it about 50% of the time) and one where we expect 50% of people to do it all of the time. The former we can deal with using a small number of users and maybe longer or repeated experiments, the latter needs more users.

If instead, we wanted to show that a behaviour happens less than 25% of the time, then we need at least 11 users, for 10% 29 users. On the other hand, if we hypothesised that a behaviour happens 90% of the time and didn’t see it in just two users we can reject the hypothesis at significance level of 1%. In the extreme if our hypothesis is that something never happens and we see it with just one user, or if the hypothesis is that it always happens and we fail to see it with one user, in both cases we can reject our hypothesis.

The above only pertains when you see samples where either none or all of the users do something. More often we are trying to assess some number. Rather than “does this behaviour occur 50% of the time”, we are asking “how often does this behaviour occur”.

Imagine we have 100 users (a lot more than five!), and notice that 60% do one thing and 40% do the opposite. Can we conclude that in general the first thing is more prevalent? The answer is yes, but only just. Where something is a simple yes/no or either/or choice and we have counted the replies, we have a binomial distribution. If we have n (100) users and the probability of them answering ‘yes’ is p (50% if there is no real preference), then the maths says that the average number of times we expect to see a ‘yes’ response is n x p = 100 x 0.5 = 50 people — fairly obvious. It also says that the standard deviation of this count is sqrt(n x p x (1-p ) ) = sqrt(25) = 5. As a rule of thumb if answers differ by more than 2 standard deviations from the expected value, then this is statistically significant; so 60 ‘yes’ answers vs. the expected 50 is significant at 5%, but 55 would have just been ‘in the noise’.

Now drop this down to 10 users and imagine you have 7 ‘yes’s and 3 ‘no’s. For these users, in this experiment, they answer ‘yes’ more than twice as often as ‘no’, but here this difference is still no more than we might expect by chance. You need at least 8 to 2 before you can say anything more. For five users even 4 to 1 is quite normal (try tossing five coins and see how many come up heads); only if all or none do something can you start to think you are onto something!

For more complex kinds of questions such as “how fast”, rather than “how often”, the statistics becomes a little more complex, and typically more users are needed to gain any level of confidence.

As a rule of thumb some psychologists talk of 20 users per condition, so if you are comparing 4 things then you need 80 users. However,  this is just a rule of thumb and some phenomena have very high variability (e.g. problem solving) whereas others (such as low-level motor actions) are more repeatable for an individual and have more consistency between users. For phenomena with very high variability even 20 users per condition may be too few, although within subjects designs may help if possible. Pilot experiments or previous experiments concerning the same phenomenon are important, but this is probably the time to consult a statistician who can assess the statistical ‘power’ of a suggested design (the likelihood that it will reveal the issue of interest).

qualitative study

Here none of the above applies and … so … well … hmm how do you decide how many users? Often people rely on ‘professional judgement’, which is a posh way of saying “finger in the air”.

In fact, some of the numerical arguments above do still apply (sorry numerophobes). If as part of your qualitative study you are interested in a behaviour that you believe happens about half the time, then with five users you would be very unlucky not to observe it (3% of the time). Or put it another way, if you observe five users you will see around 97% of behaviours that at least half of all users have (with loads and loads of assumptions!).

If you are interested in rarer phenomena, then you need either lots more users (for behaviour that you only see in 1 in 10 users, then you have only a 40% chance of observing it with 5 users, and perhaps more surprisingly, only 65% chance of seeing it with 10 users).

However, if you are interested in a particular phenomenon, then randomly choosing people is not the way to go anyway, you are obviously going to select people who you feel are most likely to exhibit it; the aim is not to assess its prevalence in the world, but to find a few and see what happens.

Crucially when you generalise from qualitative results you do it differently.

Now in fact you will see many qualitative papers that add caveats to say “our results only apply to the group studied …”. This may be necessary to satisfy certain reviewers, but is at best disingenuous – if you really believe the results of your qualitative work do not generalise at all, then why are you publishing it – telling me things that I cannot use?

In fact, we do generalise from qualitative work, with care, noting the particular limitations of the groups studied, but still assume that the results are of use beyond the five, ten or one hundred people that we observed. However, we do not generalise through statistics, or from the raw data, but through reasoning that certain phenomena, even if only observed once, are likely to be ones that will be seen again, even if differing in details. We always generalise from our heads, not from data.

Whether it is one, five or more, by its nature deep qualitative data will involve fewer users than more shallow methods such as large scale experiments or surveys. I often find that the value of this kind of deep interpretative data is enhanced by seeing it alongside large-scale shallow data. For example, if survey or log data reveals that 70% of users have a particular problem and you observe two users having the same problem, then it is not unreasonable to assume that the reasons for the problem are similar to those of the large sample — yes you can generalise from two!

Indeed one user may be sufficient (as often happens with medical case histories, or business case studies), but often it is about getting enough users so that interesting things turn up.

exploratory study

This looking for interesting things is often the purpose of research: finding a problem to tackle. Once we have found an interesting issue, we may address it in many ways: formal experiments, design solutions, qualitative studies; but none of these are possible without something interesting to look at.

In such situations, as we saw with qualitative studies in general, the sole criteria for “is N enough” is whether you have learnt something.

If you want to see all, or most of the common phenomena, then you need lots of users. However, if you just want to find one interesting one, then you only need as many as gets you there. Furthermore whilst you often choose ‘representative or ‘typical’ users (what is a typical user!) for most kinds of study and evaluation, for exploratory analysis, often extreme users are most insightful; of course you have to work out whether your user or users are so unusual that the things you observe are unique to them … but again real research comes from the head, you have to think about it and make an assessment.

In the IwC paper I discuss some of the issues of single person studies in more detail and Fariza Razak’s thesis is all about this.

the right kind of users

If you have five, fifty or five hundred users, but they are all psychology undergraduates, they are not going to tell you much about usage by elderly users, or by young unemployed people who have left school without qualifications.

Again the results of research ultimately come from the head not the data: you will never get a complete typical, or representative sample of users; the crucial thing is to understand the nature of the users you are studying, and to make an assessment of whether the effects you see in them are relevant, and interesting more widely. If you are measuring reaction times, then education may not be a significant factor, but Game Boy use may be.

Many years ago I was approached by a political science PhD student. He had survey data from over 200 people (not just five!), and wanted to know how to calculate error bars to go on his graphs. This was easily done and I explained the procedure (a more systematic version of the short account given earlier). However, I was more interested in the selection of those 200 people. They were Members of Parliament; he had sent the survey to every MP (all 650 of them) and got over 200 replies, a 30% return rate, which is excellent for any survey. However, this was a self-selected group and so I was more interested in whether the grounds for self-selection influenced the answers than in how many of them there were. It is often the case that those with strong views on a topic are more likely to answer surveys on it. The procedure he had used was as good as possible, but, in order to be able to make any sort of statement about the interpretation of the data, he needed to make a judgement. Yet again knowledge is ultimately from the head not the data.

For small numbers of users these choices are far more critical. Do you try and choose a number of similar people, so you can contrast them, or very different so that you get a spread? There is no right answer, but if you imagine having done the study and interpreting the results this can often help you to see the best choice for your circumstances.

being practical

In reality whether choosing how many, or who, to study, we are driven by availability. It is nice to imagine that we make objective selections based on some optimal criteria — but life is not like that. In reality, the number and kind of users we study is determined by the number and kind of users we can recruit. The key thing is to understand the implications of these ‘choices’ and use these in your interpretation.

As a reviewer I would prefer honesty here, to know how and why users were selected so that I can assess the impact of this on the results. But that is a counsel of perfection, and again good science and getting published are not the same thing! Happily there are lovely euphemisms such as ‘convenience sample’ (who I could find) and ‘snowball sample’ (friends of friends, and friends of friends of friends), which allow honesty without insulting reviewers’ academic sensibilities.

in the end

Is five users enough? It depends: one, five, fifty or one thousand (Google test live with millions!). Think about what you want out of the study: numbers, ideas, faults to fix, and the confidence and coverage of issues you are interested in, and let that determine the number.

And, if I’ve not said it enough already, in the end good research comes from your head, from thinking and understanding the users, the questions you are posing, not from the fact that you had five users.

references

A. Dix (2010)  Human-Computer Interaction: a stable discipline, a nascent science, and the growth of the long tail. Interacting with Computers, 22(1) pp. 13-27. http://www.hcibook.com/alan/papers/IwC-LongFsch-HCI-2010/

Nielsen, J. (1989). Usability engineering at a discount. In Salvendy, G., and Smith, M.J. (Eds.), Designing and Using Human–Computer Interfaces and Knowledge Based Systems, Elsevier Science Publishers, Amsterdam. 394-401.

Nielsen, J. and Landauer, T. K. 1993. A mathematical model of the finding of usability problems. In Proceedings of the INTERACT ’93 and CHI ’93 Conference on Human Factors in Computing Systems (Amsterdam, The Netherlands, April 24 ? 29, 1993). CHI ’93. ACM, New York, NY, 206?213. http://doi.acm.org/10.1145/169059.169166

Jakob Nielsen’s Alertbox, March 19, 2000: Why You Only Need to Test With 5 Users. http://www.useit.com/alertbox/20000319.html

Fariza Razak (2008). Single Person Study: Methodological Issues. PhD Thesis. Computing Department, Lancaster University, UK. February 2008. http://www.hcibook.net/people/Fariza/

Michael Twidale and Paul Marty (2004-) Extreme Evaluation.  http://people.lis.uiuc.edu/~twidale/research/xe/

The event started on Friday night with a lovely meal in the restaurant at the British Museum. The museum was partially closed in the evening, but in the open galleries Rosetta Stone, Elgin Marbles and a couple of enormous totem poles all very impressive. … and I notice the BM’s website when it describes the Parthenon Sculptures does not waste the opportunity to tell us why they should not be returned to Greece!

Treasury of Atreus

I was fascinated too by images of the “Treasury of Atreus” (which is actually a Greek tomb and also known as the Tomb of Agamemnon. The tomb has a corbelled arch (triangular stepped stones, as visible in the photo) in order to relieve load on the lintel. However, whilst the corbelled arch was an important technological innovation, the aesthetics of the time meant they covered up the triangular opening with thin slabs of fascia stone and made it look as though lintel was actually supporting the wall above — rather like modern concrete buildings with decorative classical columns.

how web killed the hypertext star

On Saturday, the camp proper started with Paul de Bra from TU/e giving a sort of retrospective on pre-web hypertext research and whether there is any need for hypertext research anymore. The talk brought out several of the issues that have worried me also for some time; so many of the lessons of the early hypertext lost in the web¹.

For me one of the most significant issues is external linkage. HTML embeds links in the document using <a> anchor tags, so that only the links that the author has thought of can be present (and only one link per anchor). In contrast, mature pre-web hypertext systems, such as Microcosm², specified links eternally to the document, so that third parties could add annotation and links. I had a few great chats about this with one of the Southampton Web Science DTC students; in particular, about whether Google or Wikipedia effectively provide all the external links one needs.

Paul’s brief history of hypertext started, predictably, with Vannevar Bush‘s  “As We May Think” and Memex; however he pointed out that Bush’s vision was based on associative connections (like the human mind) and trails (a form of narrative), not pairwise hypertext links. The latter reminded me of Nick Hammond’s bus tour metaphor for guided educational hypertext in the 1980s — occasionally since I have seen things a little like this, and indeed narrative was an issue that arose in different guises throughout the day.

While Bush’s trails are at least related to the links of later hypertext and the web, the idea of associative connections seem to have been virtually forgotten.  More recently in the web however, IR (information retrieval) based approaches for page suggestions like Alexa and content-based social networking have elements of associative linking as does the use of spreading activation in web contexts³

It was of course Nelson who coined the term hypertext, but Paul reminded us that Ted Nelson’s vision of hypertext in Xanadu is far richer than the current web.  As well as external linkage (and indeed more complex forms in his ZigZag structures, a form of faceted navigation.), Xanadu’s linking was often in the form of transclusions pieces of one document appearing, quoted, in another. Nelson was particularly keen on having only one copy of anything, hence the transclusion is not so much a copy as a reference to a portion. The idea of having exactly one copy seems a bit of computing obsession, and in non-technical writing it is common to have quotations that are in some way edited (elision, emphasis), but the core thing to me seems to be the fact that the target of a link as well as the source need not be the whole document, but some fragment.

Paul de Bra's keynote at Web Art/Science Camp (photo Clare Hooper)

Over a period 30 years hypertext developed and started to mature … until in the early 1990s came the web and so much of hypertext died with its birth … I guess a bit like the way Java all but stiltified programming languages. Paul had a lovely list of bad things about the web compared with (1990s) state of the art hypertext:

Key properties/limitations in the basic Web:

1. uni-directional links between single nodes
2. links are not objects (have no properties of their own)
3. links are hardwired to their source anchor
4. only pre-authored link destinations are possible
5. monolithic browser
6. static content, limited dynamic content through CGI
7. links can break
8. no transclusion of text, only of images

Note that 1, 3 and 4 are all connected with the way that HTML embeds links in pages rather than adopting some form of external linkage. However, 2 is also interesting; the fact that links are not ‘first class objects’. This has been preserved in the semantic web where an RDF triple is not itself easily referenced (except by complex ‘reification’) and so it is hard to add information about relationships such as provenance.

Of course, this same simplicity (or even that it was simplistic) that reduced the expressivity of HTML compared with earlier hypertext is also the reasons for its success compared with earlier more heavy weight and usually centralised solutions.

However, Paul went on to describe how many of the features that were lost have re-emerged in plugins, server enhancements (this made me think of systems such as zLinks, which start to add an element of external linkage). I wasn’t totally convinced as these features are still largely in research prototypes and not entered the mainstream, but it made a good end to the story!

demos and documentation

There was a demo session as well as some short demos as part of talks. Lots’s of interesting ideas. One that particularly caught my eye (although not incredibly webby) was Ana Nelson‘s documentation generator “dexy” (not to be confused with doxygen, another documentation generator). Dexy allows you to include code and output, including screen shots, in documentation (LaTeX, HTML, even Word if you work a little) and live updates the documentation as the code updates (at least updates the code and output, you need to change the words!). It seems to be both a test harness and multi-version documentation compiler all in one!

I recall that many years ago, while he was still at York, Harold Thimbleby was doing something a little similar when he was working on his C version of Knuth’s WEB literate programming system. Ana’s system is language neutral and takes advantage of recent developments, in particular the use of VMs to be able to test install scripts and to be sure to run code in a consistent environments. Also it can use browser automation for web docs — very cool

Relating back to Paul’s keynote this is exactly an example of Nelson’s transclusion — the code and outputs included in the document but still tied to their original source.

And on this same theme I demoed Snip!t as an example of both:

1. attempting to bookmark parts of web pages, a form of transclusion
2. using data detectors a form of external linkage

Another talk/demo also showed how Compendium could be used to annotate video (in the talk regarding fashion design) and build rationale around … yet another example of external linkage in action.

… and when looking after the event at some of Weigang Wang‘s work on collaborative hypermedia it was pleasing to see that it uses a theoretical framework for shared understanding in collaboratuve hypermedia that builds upon my own CSCW framework from the early 1990s

sessions: narrative, creativity and the absurd

Impossible to capture in a few words, but one session included different talks and discussion about the relation of narrative and various forms of web experiences — including a talk on the cognitive psychology of the Kafkaesque. Also discussion of creativity with Nathan live recording in IBIS!

what is web science

I guess inevitably in a new area there was some discussion about “what is web science” and even “is web science a discipline”. I recall similar discussions about the nature of HCI 25 years ago and not entirely resolved today … and, as an artist who was there reminded us, they still struggle with “what is art?”!

Whether or not there is a well defined discipline of ‘web science’, the web definitely throws up new issues for many disciplines including new challenges for computing in terms of scale, and new opportunities for the social sciences in terms of intrinsically documented social interactions. One of the themes that recurred to distinguish web science from simply web technology is the human element — joy to my ears of course as a HCI man, but I think maybe not the whole story.

Certainly the gathering of people from different backgrounds in a sort of disciplinary bohemia is exciting whether or not it has a definition.

1. see also “Names, URIs and why the web discards 50 years of computing experience“ [back]
2. Wendy Hall, Hugh Davis and Gerard Hutchings, “Rethinking Hypermedia:: The Microcosm Approach, Springer, 1996. [back]
3. Spreading activation is used by a number of people, some of my own work with others at Athens, Rome and Talis is reported in “Ontologies and the Brain: Using Spreading Activation through Ontologies to Support Personal Interaction” and “Spreading Activation Over Ontology-Based Resources: From Personal Context To Web Scale Reasoning“. [back]