Monday, 12 December 2011

Rural Relaxation: A Winter Sunset in Surrey

Sunset seen through a break in the trees
Bookham Common, Surrey
When I am feeling trapped I find that there is no better way to relax than to take a rural walk.

Brain Storms – 8 – Was Douglas Adams right about the Dolphins?


Douglas Adams, in The Hitch Hikers Guide to the Galaxy, suggested that dolphins were more intelligent than humans.

Perhaps his conclusion was based on the finding of a vast conference in with all the animal species that ever lived sent one delegate to decide on which of them had the best brain. This caucus was called because they were all fed up with Humans unilaterally claiming that they were more intelligent than all the other animals – by simply defining intelligence as “those mental activities which humans can do which other animals can't do.”

Continues below the fold

Wednesday, 7 December 2011

Brain Storms - 7 - Getting rid of those pesky numbers


There has been a break in the brain storming – in part because I have been distracted by other things and in part because my ideas had become stuck in a box – and I hadn't realised the implications.

The box I was stuck in was the one relating to the handling of numbers in conventional computer systems . I had already decided that because hunter-gatherer man would not be doing much, if any, counting – much less arithmetic - I should strip the arithmetic facilities out of the CODIL model if I was to relate it to how the brain works. However when I came to think about mapping onto a neural net there were some problems

Perhaps the easiest way is to relate the problem to CODIL's history. The original ideas developed in a very large commercial data processing department where numbers (quantities, prices, dates, customer identity numbers, etc.) were of paramount importance. It was also concerned with the real world text names of objects such as people, goods for sale, and places. In developing the CODIL paradigm I did many unconventional things but I still kept the equivalent of the conventional computing concept of a data field – which had a name so that the computer could address it and a value which was directly linked to something in the real world. The result was the CODIL item, such as the following examples:

QUANTITY > 27
PRODUCT = Diesel

This structure has been retained in al subsequent developments – when the aim was to design a flexible and user friendly tool and the idea of having a “data base” package that could not handle numbers would have been ridiculous.

... But in modelling the brain I am not building a “user friendly” tool, I am trying to model how the pre-civilization brains worked. I note that there is no difference between Australian Aborigine brains and Western European brains – despite the fact that in evolutionary terms they are at least 50,000 years apart – while modern civilization is no more than 10,000 years old. This means that in developing a brain model anything which could be an artefact resulting from the development of civilization can safely be excluded.

So back to the little example. “PRODUCT” is the name given to a set which includes “Diesel” ... but wait - isn't “Diesel” the name given to a set which includes different grades of diesel oil ... If we get rid of numbers (and also some of the features for manipulating strings of characters) we can redefine the item value as a set name where the set contains a single member – the name of the set being a representation of its value. This could be represented (with trivial changes to the existing software) by representing the item “PRODUCT = Diesel” as:

PRODUCT DIESEL

However at the paradigm level it is possible to say that an item is just a pair of symbolic set names where one set is a member of the other. What we have done is to simplify the CODIL model by discarding a component (the item value) and made the approach more general.

The implications of this will be discussed in later “Brain Storms” posts.

Earlier Brain Storms
  1. Introduction
  2. The Black Hole in Brain Research
  3. Evolutionary Factors starting on the African Plains
  4. Requirements of a target Model
  5. Some Factors in choosing a Model
  6. CODIL and Natural Language

Sunday, 9 October 2011

The "being bullied at school" box

Before I went to Dartington I was bullied a lot at school and the lack of self-confidence this engendered has remained with me for over 60 years. Following the recent school reunion (see Escaping from the Box - The Dartington Experience) I have been following the school site on Facebook and as a result found a collection of interviews "and in 2010 - bullying" of pupils who had suffered bullying before they went to Sands School, Ashburton

My experience at Dartington, from the age of 14, showed that schools could be run without bullying, and the interviews demonstrate that for many of today's pupils things in many schools are little better. Hopefully Sands School (which is the "successor" to Dartington) will help these pupils get over some of the damage they have already suffered - as there is little worse than being a child trapped in a school where you are always fearful of being victimized by other pupils, and sometimes even the teachers.

Are you playing safe and keeping well inside your box? Why not take a risk occasionally

I have just read The Blue Collar Athiest's blog "The Most Dangerous Thing You Can Do: Play Safe".

It made me think about how we often feel safer if we never do anything risky and stay well inside our mental and physical boxes. Clearly this happened to me. Because my research often caused significant criticism I didn't promote it as actively as perhaps I should, and when I was viciously bullied for doing unconventional research which didn't have funding (at a time when I was recovering from the effects of a family suicide) I just gave up the research to do something safer ...

OK about a year ago I decided to stick my head above the parapet and started publicizing my former research - fully aware that because it is unconventional most people will automatically ignore it - and some will be very critical.

So are you stuck in a comfortable box in which you believe that all "significant  research is carried out in prestige institutions and has been published in the last six months? If so you will be in good company - because such an approach avoids thinking outside the box - and will be right for much of the time. But is ignoring ideas which differ from the establishment view good science?

Isn't it about time you took a risk? Why not abandon your "What is good research" box and have a good look at what I am saying on this blog . If you like it say so - and if you don't like it be as rude as you like - as long as you make it clear why you don't agree. If I am prepared to come out of my shell and take a risk, why don't you?

Thursday, 6 October 2011

Am I being foolhardy in considering CODIL to be a model how the brain works?


In discussing Artificial Intelligence Steven Pinker, in The Language Instinct writes (my highlighting):

Since innate similarity spaces are inherent to the logic of learning, it is not surprising that human-engineered learning systems in artificial intelli­gence are always innately designed to exploit the constraints in some domain of knowledge. A computer program intended to learn the rules of baseball is pre-programmed with the assumptions under-lying competitive sports, so that it will not interpret players' motions as a choreographed dance or a religious ritual. A program designed to learn the past tense of English verbs is given only the verb's sound as its input; a program designed to learn a verb's dictionary entry is given only its meaning. This requirement is apparent in what the designers do, though not always in what they say. Working within the assumptions of the Standard Social Science Model, the computer scientists often hype their programs as mere demos of powerful general-purpose learning systems. But because no one would be so foolhardy as to try to model the entire human mind, the researchers can take advantage of this allegedly practical limitation. They are free to hand-tailor their demo program to the kind of problem it is charged with solving, and they can be a deus ex machina funnelling just the right inputs to the program at just the right time. Which is not a criticism; that's the way learning systems have to work!

In various past and future postings on this blog I am arguing that CODIL is a model of at least some aspects of how the brain works and in making wide claims it is important that I ask myself whether I am being foolhardy, and whether I am making extravagant claims for a model general purpose learning system for the brain.

Wednesday, 5 October 2011

The Establishment rejected research which later got the Nobel Prize

Greg Laden has reported on 
"There can be no such creature" - Chemistry Nobel: Daniel Shechtman

At first Shechtman's results were interpreted by other experts in the field as a simple mistake that only an idiot would make. They told Shechtman that he was observing a "double crystal" and mixing up his observations. One snarky colleague gave him a school textbook on crystals implying that he should brush up on his basics. A manuscript Shechtman sent to the Journal of Applied Physics was sent back unopened.
Over time, using his personal connections to a grad school colleague and, in turn, his colleague's connections, Shechtman got his work looked at more closely by other crystal experts, who concurred that the results were both a) real and b) impossible. Finally, in 1984, a paper coauthored by Shechtman and those experts (John Cahn, Denis Gratias and his schoolmate Ilan Blech) was published in Physical Review Letters. The world of crystallography was stunned, and the single most important pillar of dogma of the science of crystals ... all crystals consist of repeating periodic patterns ... was under serious question.
There are several interesting lessons in that Odyssey of Shechtman. One, don't discount your personal and professional relationship, or the cultural aspects of scientific networks. A pure, scientific, or even "skeptical" view of Shechtman's work failed. Shechtman needed to get someone to look at his results on blind trust that there could be something there, rather than approaching it from what was known to be very well established fact. On the other hand, it must be remembered that 99.99% of the time that a scientist gets an email from someone telling them that the basic tenets of their science are wrong, it is some crazy guy who sees things in rocks. But not this time. Ultimately, this is a case of science being conservative, which is usually appropriate, but learning something new. The turnaround time between nobody knowing this thing and key scientists getting it in print was only a couple of years.
Mu comments:
This story makes you wonder how many discoveries were suppressed as "defies science" before the age of arxiv and instant distribution. Nice to see he didn't get scooped and got credit in the end.
When I hesitatingly proposed in 1967 that it should be possible to build a human-friendly "white box" information processing system (as a opposed to an inherently unfriendly "black box" conventional computer system) the idea would never have even started to get off the ground if it had not been a very supportive boss, George Stern, and his boss, John Aris (see How ICL came to axe the CODIL project). However when ICL closed and I was declared redundant for wanting to follow up the idea I got the same kind of hostility as Shechtman - and as someone who is not good at exploiting contacts (due to vicious childhood bullying)I struggled with the research, getting many rejections because "I was an idiot" until I finally had a breakdown and abandoned the idea.


It is only recently that I have dusted down the old files and it looks as if the idea could be a partial model of a symbolic language that could bridge the gap between the brain's neural net and natural language. Whether this is true a study of how my research was "suppressed" could well give clues as to how other very interesting ideas have been consigned to the garbage heap.



Tuesday, 4 October 2011

Putting Faster Garbage into the Westminster Box


We all live in mental, physical and social boxes – and while we may sometimes between boxes there is one box which we share and which we cannot escape – The Earth. And this box is getting very crowded and the society in which we live is looking more like a House of Cards every day – with financial crises shaking the structure and climate change weakening the foundations.

So what are our politicians doing about it.

Wednesday, 21 September 2011

Children are better Scientists than Adults

The Scientific American has just posted an item More than Child's Play: Ability to think Scientifically Declines as Kids Grow Up describing research on 4 and 5 year old children at Massachusetts Institute of Technology and Stanford University.

In 1980 I had a problem. The university where I worked had just installed their first fully interactive computer service and I had the task of introducing 125 first year undergraduates (mainly about 18 years old) to the joys of computing. Many had never used a keyboard of any kind, some had studied computer science at school and a small number had their own home computers. Could I set an introductory piece of coursework which would not allow the budding computer geeks to show off in front of the others and scare off those who were nervous of the technology.

About a year earlier I had prepared a small demonstration package for a TV programme called Jim'll Fix It where children wrote to ask Jimmy Saville to help them to do something - and one eleven year old child had asked that a computer could help him do his homework.

I adapted and extended the package (which was written in CODIL) two include a number of additional games and data bases and the course work was:
FIXIT is a package designed for an eleven year old but it has some problems. Try it out and write a two page account of what you found.
Psychologically this worked well.  Even novices did not want to be beaten by a package for an eleven year old, while the need to write a critical account in English gave some of the budding computer whiz-kids a bit of a headache - as they had come convinced that computing was all about writing programs in BASIC.

Afterwards I gave a debriefing session to explain the significance of different "games" in the package to different sections of their future studies - and I also discussed some of the limitations of the package - especially the ones none of them had spotted. For instance FIXIT contained information on a number of British Birds and if you typed in the name of a bird it came up with some information about it. Not one of them discovered what happened if you typed in the name of a different kind of bird - i.e. a girl's name! The KINGS section asked you to type in a date and it told you the English King at the time and a number complained that they would have liked to be able to type in the name of the king and be told his dates.  Of course any eleven year old would have tried it, rather than complain, and would have found that it worked!

To me it is obvious that as we get older we organize our knowledge into mental boxes and the more we cram in the more reluctant we are to ask questions which fail to conform to our mental models. 

Monday, 19 September 2011

How ICL came to axe the CODIL project


On the 3rd of May 1968 John Aris, who had worked on Leo Computers, was then Manager of Applied Systems in English Electric Computers, and later Director of the National Computer Centre wrote: about a new research project called DORIS (later renamed CODIL):

DORIS, a 'Data ORiented Information System', is an original and radically new; approach to the systems/programming of computers, devised by Dr. C. Reynolds of this Department. It involves holding a large proportion of the decision-making logic, normally built into user programs, as part of the data files, and analyzing it not with a variety of special applications programs, but with one general purpose decision making routine. If this proves viable it could result in the virtual demise of the conventional application program in many fields, replacing it by the said general purpose routine and groups of calculation routines. Its data structure, being closer to the ordinary modes of human thought than computer files generally are, could allow the ultimate user far greater freedom to use the computer as an information handler in a familiar way, and without the need for expensive commercial programmers. This would, among other advantages, ease the implementation of management information systems. A further important consequence could be a new approach to CPU design.

Saturday, 10 September 2011

Review: The Origin of Our Species – by Chris Stringer


As a young man I was very interested in the caves of Devon, and the early work of the Victorian archaeologist William Pengelly in Kent's Cavern, Devon, so I was delighted to learn that human bones from this cave have now been shown to be about 40,000 years old and hence among the oldest homo remains found in Europe. Fifty years later my interests in human-computer communication have widened in an attempt to understand the evolution of human-human communication and I needed a crash course on what is currently known on the subject.

Thursday, 8 September 2011

Voices of the Deep - Marine mammals and the cost of language

I have just seen the BBC film Voices of the Deep, which is part of the BBC Ocean Giants series currently available on I-player. and the shoot of, for example, the different ways in which individual pods of orcas (killer whales) catch their prey clearly shows a high degree of culture. Other recent items I have seen on line include research into Dolphin language  and the following video on Youtube which showed how orcas could disable and kill the Great White Shark

Escaping from the Box – The Dartington Experience


Some reflection following the Dartington Hall School reunion, 2nd-4th September, 2011

Dartington Hall School was one of the most progressive schools anywhere in England and between 1952 and 1956 I found myself way outside the conventional educational “box” as a pupil there. At the reunion a few days ago there was much discussion about old times and also the effect that going to such an unusual school had on one's career. I was one of the pupils that Bill Curry (the headmaster at the time) rescued from the evils of conventional education In discussions I said I would write something of my experiences - and I am doing so here because the observations are very relevant to the way we all become trapped by the mental, physical and social boxes we grow up in.

Tuesday, 30 August 2011

Language is based on what the brain can do – rather than the brain evolving to accommodate language.


As a result of a post What makes Humans Tick on Babel's Dawn my attention has been drawn to a paper Language Has Evolved to Depend on Multiple-Cue Integration by Morten H. Christiansen to appear in R. Botha & M. Everaert (Eds). The Evolutionary Emergence of Human Language, Oxford, Oxford University Press.
In discussing Language as a Culturally Evolved Linguistic System Morten writes:
A key question for language evolution research is to explain why language is the way it is, and how it got to be that way. The cultural evolution perspective suggests that the structure of language derives primarily from processes of cultural transmission involving repeated cycles of learning and use, constrained by the properties of the human brain. Thus, instead of asking, “Why is the brain so well-suited for learning language?”, we need to turn the question upsidedown and ask, “Why is language so well-suited to being learned by the brain?”
He goes on to say
Consequently, what has evolved is not a set of neural structures specific to language; rather, cultural evolution produces a system of linguistic constructions specific to a given speech community (i.e., a language).
This is really saying what I am trying to say on this blog – but coming from a very different direction.
The basic idea which underlies CODIL started as an attempt to design a sales accounting system in 1967 which would handle perhaps a quarter of a million very different customers, and perhaps 5,000 different products in a dynamic market. Details of the CODIL research and the sad reasons why the project eventually folded are given elsewhere on this blog.
To summaries the key factors: CODIL uses a conseptually very simple recursive set structure to represent information (there is no formal distinction between program and data) and a very simple “decision making unit” which scans through the stored information. Despite this simplicity it has been demonstrated to work, albeit sometimes on a small scale, on a wide range of non-numerical applications including some which would classify as artificial intelligence, or involve fuzzy logic or dynamic learning.
Its relevance to the evolution of language (I will be posting further information about this in September) is that it appears that both the information structure and the processing routines will translate comparatively easily from the sequential central process of the original model onto a neural net where the amount of information done by any one node is acceptably small. It could well be that the processing power already demonstrated in the original sequential model can be reproduced in a neural net and is sufficient to support many aspect of human natural language.
If a CODIL type model can be sustained for the inner workings of the brain and can be extended to support the plethora of human languages, this would fully demonstrate the feasibility of the ideas represented in the above quotations.
However don't expect me to come up with all the answers. I am an old age pension who retired from university research over 20 years ago, with deteriorating eyesight, and with no access to academic library or computer facilities. As far as I am concerned this blog site will have been successful if I can persuade someone younger and fitter than I am, and who has adequate resources, to follow up the ideas.

Babel's Dawn and the Evolution of Vocalisation


A copy of Edmund Bolles' book, Babel's Dawn, A Natural History of the Origins of Speech, has just arrived and tells a fascinating story of how humanity has developed from just being a great ape over the last six million years. He uses a series of verbal snapshots to reconstruct what life was like for the Last Common Ancestor and their descendants. At each stage he discusses the findings of recent research, the uncertainties of reconstructing the past of our species, and the many gaps that still have to be filled. For each episode there are brief notes that re-direct you to the extensive bibliography.
So far, so good, but I have two major problems with the book. Before I retired I used to review books for the New Scientist, and also was the book review editor for an online discussion forum called HICOM, which was an early UK based interactive forum discussing a very wide range of subjects relating to the way humans interacted with information and computers. This book might even have been included on the possible review list as it is about how humans came to communicate with each other.
The first question I would ask in picking up any book to review myself, or to allocate a suitable specialist to review, is “What is the market for this book?” In the case of Babel's Dawn I am not sure. The story is told as if the reader was was walking through a museum with dioramas representing different periods in the past, with a verbal commentary being given as you go. For school children and the generally interested public the book would have been far more effective if each vista had been described with an associated picture with the pre-human figure in a reconstructed view of the landscape, and the amount of more formal scientific discussion significantly reduced. On the other hand, the serious scientific reader might find the faction approach annoying and the lack of an index frustrating. However my knowledge of the modern US university student book market is very limited, and it could be a useful reader, rather than reference book, in introducing a subject which many students might find controversial because of their religious beliefs – a problem which is far less an issue in the UK.
I also had a serious problem with Edmund's treatment of vocalisation and its relevance to speech, Without vocalisation there can be no speech as we know it and in my opinion the book fails to examine the evolutionary pressures that lead to the changes in the vocal tract which makes speech possible.
To be fair there is a very real difficulty in reconstructing this aspect of mankind's part as sound is about as ephemeral as one can get, and there is no possibility of any direct fossil evidence. Even indicator evidence, such as the changes in a tiny bone in the larynx, is hard to find, as the bone is rarely found in fossil deposits. In practice all one can do is to speculate on the evolutionary pressures that might lead to the evolution of vocalisation, which in turn lead to speech.
So lets step away from the book and speculate as to how the story might have been told.
Undoubtedly a very important aspect of the evolution of the human body was the move from living in forests to living in far more open and dryer environments. As the forests dried out they would have initially fragmented, depending on soil, altitude and drainage conditions locally, leading to a series of forested “islands” that would eventually vanish entirely. Forest species trapped on these islands, including our ancestors, would either have to adapt or become extinct. Such environmental changes, temporally trapping creatures in small habitat areas which are in the process of disappearing, are powerful evolutionary workshops and if we look at the human body most of the changes make sense. Descending from the trees and walking in more open areas is an obvious driver for changes in our lower limbs. Such changes also free our forelimbs for gathering food, tool making and throwing things. As a vulnerable animal, which no longer has trees to climb to escape, and who is not agile enough to catch fast moving prey or run from the larger predators, early man would need a degree of cunning and planning which will be helped by a larger and more imaginative brain.
The relevant evolutionary pressures are not so obvious when it comes to vocalisation. The modern human has a very powerful and flexible vocal system – which can produce a wide range of noises, including whistles and clicks, with significant control over pitch, volume and timing. However a modern human language may only use about 30 phonemes to express meaning – so it would seem that our vocal system has developed to a far greater extent than is necessary to support speech.
Of course social bonding might have been a factor, as Edmund suggests, but our great ape ancestors would almost be using facial expressions, body language, grooming and gestures to communicate when they are close together, and it has been suggested that the reason that we have developed whites to our eyes is to make it easier to “read minds” by observing our social companion's eye movements. It is easy to see why, if vocalisation changed for other reason it might have been adapted for social communication but it is not obvious why, if stronger social bonding was an advantage, that this should not have been done by a cultural change affecting the existing bonding techniques, rather than the far slower method of, for example, evolving a more versatile larynx.
One must realise that when man's forebears first came down from the trees the use of sound would have had its snags. Using sound is a very public way of communicating – and while someone might be able to signal to a colleague on the other side of a valley one must remember that every prey or predator in the neighbourhood would pick up the sound and turn to look at where the sound indicating danger or food came from.
However there is one area where improved vocal ability could have a significant advantage and that is if used as a tool when hunting. After all, in evolutionary terms getting enough to eat must be the top priority after avoiding being eaten yourself. If the aim is to catch a medium-sized herbivore such as an antelope a good technique is to drive the animal towards an ambush, but this needs a number of hunters working as a team, and this involves signalling at a distance. The ambushers need to tell the drivers when they are ready while it could increase success if the drivers warn the ambushers that the prey is approaching the killing area. Initially an ape-like grunt to indicate that the ambusher is ready might be adequate – but over time evolutionary pressures on the prey would alert it to the fact that ape noises can indicate danger – so the signalling method would need to change. As the long time battle between prey and hunter continues the hunter must improve his tactics – so an innocent imitation call of an owl hooting, or the whistling call of a bird would be a far better signal. But as the hunters vocal skills develop it will become possible to lure other animals by making, for example, mating calls at the right time of year.
If we consider that the ability to mimic other animals is an advantage to the increasingly intelligent hunter we now have an evolutionary pathway that will lead to the development of a very wide range of vocalisations - which is what the human species possesses. But this also leads to a possible start for using vocal symbols to communicate. After all, using an animal call to symbolise an animal is the equivalent of using a simple picture as a symbol in the early days of writing. Perhaps, in the same way that pictorial languages simplifies to alphabetic languages using words, vocal languages developed in a similar way by using special combinations of sounds to symbolise names of individual members of the group, perhaps naming them after animals. Vocalisations might then be extended to other objects. Further moves towards the rudiments of language would involve reducing the number of phonemes needed to be accurately recognised and generated by combining groups of sounds into words.
But that is a story for another day.

Sunday, 28 August 2011

The Old Codger comes up for Air

The recent lack of posts on this blog is because August has been rather chaotic for me. On Doctor's order I have to loose weight and get plenty of exercise and frequent rural walks - plus doing shopping by walking rather using the car - takes time. A visit to the hospital brought some good and bad news. My left eye, which was hit by glaucoma a few years ago, is stable and if I needed to use it I can still read the computer screen with 4 times magnification. The problem is that my "good" right eye has started to develop a cataract, and while it has not developed far enough yet to justify an operation I am finding reading smaller text difficult.  While my wife is getting much more mobile after a knee operation, I find that I still spend as much time shopping and preparing the meals, and the number of days spent on social activities has increased   because she can get around better!

One particular problem this month is that both my wife and I are hoarders - and last year we had new insulation installed in the loft. This involved downloading much of the stuff stored up there into the spare bedroom - where it has stayed - as one of the effects of old age is that neither of us want to climb up a ladder to put it back. This would not matter except that we are having an Australian visitor to stay over September - and it was only last night that the room was clear - after many visits to the local charity shop, and filling bins for recycling.

On the computer side I find that I have far more that I would like to do that I have time to do it. For example my genealogy site has over 3,500 pages of information about Hertfordshire and Hertfordshire people, and I am slowly working through the towns and village pages upgrading them, and introducing new material. The problem is that even if I averaged one town or village a week I would be approaching 90 years old before I finished - and in reality the process is more like the never-ending task of painting the Forth Bridge. At the beginning of August I decided to cut back on this activity but have been "rewarded" by an increased flow of challenging questions. To give myself a break I have shut down the "Ask Chris" facility throughout September.

Despite all the above I have been thinking quite a lot about this site - with quite a lot of background reading and even starting to draft an occasional blog - but nothing has emerged on blog. So I have decided to move this blog back up the priority list.

By the end of the August I plan to post a review of Edmund Bolles new book "Babel's Dawn: A Natural History of the Origins of Speech" and early in September I plan to post an article suggesting why my earlier work on CODIL actually provides a possible insight into the brain's "symbolic assembly language".  I may also post details of the old school reunion at Dartington Hall - which was a very unconventional school which greatly stimulated my interest in science - which encouraging the idea that the established viewpoint was not always the right one.

Thursday, 4 August 2011

The explosive evolution of langauge

Greg Laden's blog has just posed the following talk by Mark Pagel under the title "How language transformed humanity"



I am very interested in the way that language suddenly appeared on the scene, and the accelerating speed at which it has helped to transform the world in which we live. While Mark clearly recognizes the importance of cultural factors I was not happy with some of his supposed facts - and while I think that the concept of "social learning" can be useful in discussing the cultural explosion that coincides with, and is possibly driven by, the appearance of language I was not always happy with some of his examples. As a result I published the following comment on Greg Laden's blog.

Oh Dear - he really runs down the capability of our ape and early human relatives and misses the significance of fact that many of the areas where there are many languages with few speakers involve primitive societies.


I would agree that once language reaches a certain tipping point it greatly facilitates the transfer of information between generations - and also motivates further cultural development of the language itself. Basically you get what in chemistry I would call an auto-catalytic reaction. Once the tipping point is reached the each development in language skills makes further development easier - and there is an accelerating chain reaction. This will not only affect language - but also all the cultural activities and objects of the time.


The question which he does not address is what came before "language" which he does not really define, except by a crude analogy. Once the human line split off there has been a steady increase in brain size and almost certainly vocal abilities - and the majority of these changes will have been before "language" evolved.


What were the evolutionary pressures to drive these "improvements" if "language" did not exist. Humans were hunting on the African plains and needed skill and good team working to catch and kill larger animals. For instance in setting an ambush those lying in wait need to alert the drivers that they are ready. What better than sound - but if the sound is obviously human the prey will evolve to treat obviously human utterances as a danger - so what could be better than to evolve the ability to mimic other animals such as an owl hooting, etc., and to make a wide range of clicks, whistles and yodels, and to be able to change pitch


Before a proper language had developed there could well have been a simple hunting symbolic language which would involve a very wide range of phonemes - and each hunting group might use them in different way because they were hunting different prey in different environments. Recent research on attempting to date the development of language in Africa start with the assumption that the more primitive languages use a larger number of phonemes. Such an origin might also explain that the majority of languages with very small numbers of speakers involve tribes which are either still hunter gathers or were so until comparatively recent times.


Where did the tipping point come. My own guess is around the camp fire of an evening, when the hunters returned - and started to use their hunting calls to tell the story of the day. Children would learn about the methods and dangers of hunting without being exposed to them - thus acquiring a wider range of hunting skills. Once language reaches the stage of "Tell about the time you killed the lion, Daddy" the art of story telling has got underway. Legends are born and those of the Australian Aborigines go back several tens of thousands of years to the Dream Time, when some now extinct animals were alive.


An interesting feature of this model of language development is that language starts from a cultural tipping point and there may be no simple genetic factor at all. OK - more brain capacity could help handling far more concepts - and better vocal skills might make it easier - but the fact that the more advanced languages use less phonemes would suggest that a wide range of vocal skills is not essential for language.

Monday, 1 August 2011

The Curse of Knowledge

In this week's New Scientist (30 July) Richard Fisher writes:

There are many virtues in being ignorant. We all aspire to have the smarts, but it now seems knowing less can sometimes be an asset. It can make you a better teacher, a more perceptive student and a happier person overall.

The article includes details of a study by Nate Kornell, of Williams College, Williamstown, Massachusetts:

Kornell and Son staged a trivia quiz for mathematicians and historians. The pair asked these academics 90 questions about other experts in their field. Try two of the questions yourself: is Johannes de Groot a famous mathematician? What about Benoit Thoron? Would you answer yes, no, or don't know?

When asked about mathematicians, those in the same field were more likely to give a definite answer, yes or no. Yet while they might know their differential equations inside out, the mathematicians' confidence in naming members of their field was unfounded. They gave more wrong answers to these questions than the historians, who were more willing to confess their ignorance. But the historians weren't humble when it came to their own field - they made the same blunders over the questions about their peers. “They assumed their knowledge was great, but it wasn't in this case” says Kornell. “Experts should say 'don't know' in their own field of expertise more often.”

Are you honest about the limits of your knowledge? How did you answer the mathematician question? De Groot is a Dutch mathematician who died in 1972. Thoron is fictional.

My experience in developing CODIL is very relevant. My first job was as an information scientist working on research and development issues in a large international organisation. I was dealing with scientists and managers (up to board level) with very different backgrounds and specialities. The idea of one information worker having complete knowledge of what the organisation was doing was ridiculous. Everyone in such a large company, including the information workers, is ignorant of much of the detail of what was going on. What the information worker needs is the ability to identify sources of information when required – plus the “teaching skills” of being able to interpreting specialist reports in more widely understandable words.

I then moved to the data processing department of a very large commercial company and took my approach to knowledge and ignorance with me. I was aware that there were specialists in selling aviation fuel to the United States Air Force, in selling central heating oil to domestic houses, in selling tarmac to road builders, etc. I was also aware the the market place was dynamically changing. To me the idea that a group of systems analysts and programmers could produce an exact model which not only accurately reflected today's requirements but also allow flexibility to meet the commercial challenges that might appear tomorrow was something in cloud cuckoo land. Based on my experience of manual information process I took the idea of ignorance in my stride and suggested the answer was not to draw up an exact model of the invoicing process but rather to model the way that sales staff thought about invoicing. The aim was to allow each sales specialist within the company to directly control his area of the market without the need to conform to a globally pre-agreed approach which would always be incomplete and out-of-date.

How this idea later expanded the design of CODIL is describe in detail elsewhere on this blog. However the difficulty I had in getting the idea accepted is directly linked to the concept of ignorance. The computer establishment was, and still is, based on the idea that it is necessary to precisely predefine the application in advance. Ignorance of the application is automatically taken as a sign of incompetence. There is no doubt that much of the opposition to CODIL was that it was taken for granted that it is impossible to program a computer from a position of ignorance of the task – therefore I must have created the computer equivalent of the perpetual motion machine.

This blog is trying to take the CODL ideas further and say that in as far as you can model the human brain you can use the model to handle situations involving ignorance in a way that mimics the way that humans also tackle the problem .
~ ~ ~ ~
There is an example where too much knowledge has slowed down the progress in the “Brain Storms” on this blog. Over the last week or two I have been trying to complete the first stage of remapping CODIL onto a neural net. Most of the basic processes move relatively easily but one basic facility – that of asking a question of the knowledge base – would not go in a way that I felt was satisfactory if one wanted to move to parallel processing.. Basically CODIL uses the question as “criteria” (a little bit like a conventional program) and the knowledge base “file” (long term memory) was loaded statement by statement into the “facts” (equivalent to human short term memory). The problem was that in originally designing this part of CODIL I had too much knowledge about how conventional computers work and had incorporated this in the design.

Once I had realised the “block” existed I went right back to first principles. In fact it is obvious that the important thing, in terms of brain activity, is the question – so this should go into the Facts (short term memory). Once there the knowledge base file is asked to say what it contains which is relevant – rather as if it was acting as a kind of program, rather than as data. A new “Brain Storm” should appear in the next few days covering this point.

Wednesday, 27 July 2011

Brain Storms - 6 - CODIL and Natural Language

CODIL was designed to be the symbolic assembly language of an information processing system which reflected the way some people thought about open ended tasks (where it was difficult to predefine a processing algorithm). It arose, not from an academic attempt to model human thought processes, but from a pragmatic attempt to help sales staff control the processing of contracts in a very large commercial organisation, and was later generalised to handle a wide range of basically non-numerical tasks.

The interesting question is whether CODIL works because it was also modelling the “symbolic assembly language” of the brain, and it is therefore very relevant to ask whether the approach could be made to work on a neural net. Is it possible, starting from the CODIL model to reverse engineer and intelligent brain model that is capable of being explained in evolutionary terms?

Before going further it is appropriate to consider how a language like CODIL relates to natural language.

In a stored program computer system the symbolic assembly language relates to moving numbers (which represent data) between the memory and registers and carrying out processes and explicit logic tests. This mechanism is remote from normal human thought processes – which is why the stored program computer is a black box whose internal working is utterly incomprehensible to the average user of the “box”. Billions of pounds have been spent over the years to develop elaborate software packages, layer upon layer, like an onion, to hide the mysterious inner workings and provide a more friendly interface.

In CODIL (at least if considered in terms of the original hardware proposals) the symbolic assembly language represents not numbers, but sets and subsets, with the set names being those decided on by the user of the system. All addressing is done associatively by set names (which the user understands) and most processing is done automatically with the decision making unit (the equivalent of the stored program computer's CPU) carrying out simple set operations.

Take a simple CODIL statement:

1 MURDERER = Macbeth,
2    VICTIM = Duncan,
3       WEAPON = Dagger.

This can easily and quickly be recognised that someone called Macbeth used a dagger to kill someone called Duncan. While the statement is not written in English its meaning is immediately recognised by English speakers. If you look at the many other examples in the Publications about CODIL associated with the blog you will find that in many cases you can convert the statements into English with comparative easy. It is almost as if humans have the mechanism to convert such set descriptions into natural language., and there is no reason to think that this would not be true for any other human “natural” language. (I exclude formal mathematical systems from this observation.)

This raises the question of what extra “facilities” might one need in CODIL (and the brain) in order to convert set descriptions (held in parallel on a neural net in the case of the brain) into the serial string of spoken words, and also the reverse process. In terms of brain modelling what functional changes, if any, are needed to an animal's brain in order to handle language, excluding those functions explicitly connected with speech generation (i.e. muscle control of the voice box, tongue, lips, etc.) or simple capacity (number of neurons available.)

If one looks at CODIL (as implemented in MicroCODIL) the following statements will generate “Macbeth murders Duncan with a Dagger” in the DISPLAY window. (TEXT is a reserved set name – and instead of the item being moved to the FACTS the value is displayed on the screen.

1 MURDERER,
2    VICTIM,
3       TEXT := MURDERER.
3       TEXT = murders.
3       TEXT := VICTIM.
3       WEAPON,
4          TEXT = with a.
4          TEXT := WEAPON.

A more general approach would include statements such as

1 MURDERER,
2    VICTIM,
3       SUBJECT := MURDERER.
3       VERB = Murder.
3       OBJECT := VICTIM.
....

with further statements on how to modify verbs such as murder and word order depending on context, to generate sentences such as “Will Macbeth murder Duncan with a dagger?” or “Macbeth used a dagger to murder Duncan”.

When I was engaged in full time research over 20 years ago I was unable to explore the possibility of assessing how good the CODIL approach might be to generating or understanding natural languages or for translating between languages and if anyone is interested in following this approach I am happy to advise. The important thing is that if CODIL (as mapped onto a neural net) is a reasonable model of the brain's “symbolic assembly language” and it is capable of supporting something approximating to natural language, there is no need for any significant new “logic processing” capability between the great apes and humans – simply that humans have greater capacity – to handle the extra language information needed – and also have better sound-generating capabilities.

Held to ransom over Winzip Archive Files

When I acquired my first PC in 1992 I started to use WinZip to archive data - some of it transferred from other computers and dating from 1988 or earlier. However as storage has become so cheap, I haven't used it to archive important files since 2003 although I used it extensively to store records of ebay sales and purchases between 2003 and 2005.  However I have used it occasionally to unpack old files or zipped folders supplied by other people.

I recently received an invitation to try out a new version of WinZip (version 15.5) for free. I thought it might be useful to have a quick look - and decided not to purchase. Several months later I wanted to unzip some of my old archives and found that the trial version wanted payment - and HORROR of HORRORS - it had deleted the old version. I needed to access the files - so reluctantly I paid up. It then became apparent that the version I had brought would stop functioning if I didn't continue to make annual payments - and so I would loose access to 20 year old archives unless continued to pay up indefinitely. There is nothing to say in the terms and conditions that it will remove old (but licensed) versions of the software,  or about from the vague word "limited", indicate that you will be unable to access your ancient files unless you pay an annual fee ...

WinZip appear to have forgotten that the whole part of archiving information is to be able to retrieve it in the future. I am quite happy to pay an annual fee when I am actively using the software to archive new files - as long as there is a guarantee that the fee I paid to create the archive also allows me to recover the data at any time in the future.

To add insult to injury, the software came with the benefit of free software - which is no more that a sales promotion package in that if you run it you are told you can only use it properly if you pay even more money ....

The "access to old files" situation is rather like the trick that Microsoft played with Word some years ago - when an upgrade meant that it would no longer read text files I had archived in the 1990s. As a result I am keeping running an old PC which has a version of Word which will read the old files - and reformat them into a form that newer Microsoft software will still read. I have also switched to OpenOffice (Free!) which can read the old files (although with a few formatting errors).

The only benefit of all this is that I search my hard disc for all WinZip folders (5211 of them) and listed them in date order. As a result I have found a number of important folders containing files relating to CODIL from the 1980s which I had thought I had lost.

Monday, 18 July 2011

Rural Relaxation at College Lake

The Marsh at College Lake, Near Tring
Pyramid Orchid
Having opted out of the science rat race over 20 years ago I am beginning to realize how retirement has affected my view of the world. I have never been the best organizer of my own time - but at least when I was lecturing there was a fixed timetable of required activities to provide a framework for the day and the week.

Moorhen and chick on rippled waters
The trouble is that I have always been a workaholic with an overflowing in-tray - and I still am - despite being retired. With few fixed dates it is easy to put things off until tomorrow - and sometimes I suddenly realize that several matters have become urgent. Because of what happened to my daughters I need to keep my stress levels down - and the best way to do this is to take a country walk - and things such as this blog just have to wait.

Wayfaring Tree
Fortunately there are plenty of opportunities to relax in the area around Tring, but one of my favourites is College Lake, which is run by the Beds, Bucks and Oxon Wildlife Trust. When I first visited this large hole in the ground some 20 years ago work had started on establishing a nature reserve at one end - while quarrying was continuing at the other. Last year a fine new visitors centre was opened - which means I can end a walk with a cup of coffee and a piece of cake!

Teasel
There are a wide range of habitats - a marshy area, shown above, a deep water lake, woodland - including a newly planted area which will look wonderful in 50 or so years time, heath land areas and an area where crops are growing in the old way - with all the associated weeds! There are also areas of bare chalk, left from the quarrying which is beginning to be covered with vegetation, and rare breed sheep and cattle graze the grassy area. Whatever tine if year you visit there is always something for the nature lover to see. For instance only a few days ago I spent a happy half an hour watching a hobby flying over the march area catching dragonflies.

Photographs all taken in July 2011












Friday, 15 July 2011

Squeezed into the insurance company's box

We all have problems in trying to fill in computerized forms which give us the choice of options none of which fit the bill. As a result  I was very much amused by the blog post Misinformation and muppets by Azaria Frost. She describes the problems of getting car insurance when your occupation is not on the list. Several people made comments about how a different description such as "office worker" - which could apply to many people who also have a specialist title - can reduce your car insurance premium. One comment also explains that one of the reasons is because of the need to share information.

The problem is one that is pretty universal. While it is useful to classify information into boxes the real world is not like that. Often there is not clear borderline - for instances the classification of living things into families, species and subspecies in always going to turn up intermediate cases. Attempts to analyse data statistically can throw up problems. One of my first blog posts was I discover Babel's Dawn which had a review entitled Last Common Language was in Africa which I find very interesting. However when you look at original paper Phonemic Diversity Supports a Serial Founder Effect Model of Language Expansion from Africa in depth there are problems with the quality of the source data. This comes from classified data about a large number of different languages. While this does not invalidate the analysis the results must retain a considerable degree of uncertainty because of the classification assumptions which underlie the research.  


The important thing to remember is that in classifying and statistically analyzing any body of data there are likely to be difficult cases. Most computer systems simply sanitize the data by shoe-horning the "messy data" into a standard category - and "conveniently" forgetting that there was ever any uncertainty.  

Sunday, 10 July 2011

Brain Storms - 5 - Some factors in Choosing a Model


As CODIL is an artificial language which has been implemented on a stored program computer my experience has been that many people jump to conclusions about what is does and how it does it. For this reason it is worth spending a little time discussing the limitations of models in understanding science.

My own approach to scientific models is greatly influenced by the fact that as a student I studied chemistry at a university with a strong interest in the underlying theories, and went on to do a Ph.D. with a strong theoretical content. As such I was well aware that it helped to have different models for different purposes – and each model had its strengths and weaknesses – and it was often possible to represent the mathematics in different way. The difference between the Ancient and the Copernican models of the Solar System are little more than a change in the mathematical origin (which simplified calculations) and a change in the belief that humans must be at the centre of creation.

In the context of “Trapped by the Box” it is also important to realise that our brain constructs models of reality which also have their strengths and weaknesses. We are all aware that the brain/eye can be fooled with various optical illusions. The “image” of what we see is an incomplete view of the reality, which is filtered by what we expect to see. This was brought home to me when I visited a nearby zoo with a young child who suddenly pointed with an excited cry of “Look”. Everyone looked in the direction she was pointing and saw nothing that could possibly be relevant, such as a bear with its cub. Nothing – so the child pointed more directly and we could all see the ant on the barrier to the enclosure!

Everyone also has models for the human languages that they speak – and these can influence the way they think about the world. So what about computer languages. It is well know that if someone becomes very familiar with one programming language this will influence how they use a new language. In fact it is quite difficult to find anyone that does not take the stored program computer approach for granted, even if only to admit they they are not clever enough to write programs. The more knowledgeable will talk about the Turing Universal Machine to demonstrate the power of the almost universally accepted approach.

But all models have practical limitations and the Turing Universal Machine and the stored program computer are no exceptions. They are based on the assumption that you can define the algorithm (program) in advance. Whatever the theoretical position might be if we had infinite knowledge, there are many aspects of the world where we do not have enough prior knowledge to define an algorithm. It is in these areas that CODIL was designed to work. I would also suggest that this area of incomplete and inexact knowledge is the one in which the brain (both animal and human) evolved.

A detailed assessment of CODIL will come in later postings but it is useful to think of CODIL as the “symbolic assembly language” of a non “Von Neumann” computer. This table highlights some of the differences.

Feature
Conventional Computer
CODIL system
Information stored as:
Numbers
Sets
Recursion as part of memory structure
No
Yes – sets can contain sets
Information addressed by:
Numerical position in linear store
Associatively by Set Name
Distinction between program and data
Yes
No
Address calculation
By program
Automatic
User friendly at the processor level
No
Yes (people understand sets!)
Algorithm required in advance
Yes
No
Will basic approach map easily onto a neural net
No
In part very easily but more research needs to be done

I can just imagine some of the computer purists looking at this – noting that CODIL has been made to work on a stored program computer – and blowing a large raspberry. But as I discovered when I studied Chemistry – you need different models for different purposes – and often different mathematical models prove to be equivalent. For those who wish to dismiss the idea may I suggest they remember that there are sets of numbers and it should be possible (and I leave it to others to prove it) that the stored program computer is a special subset of the CODIL approach. For instance if you said all set names are unique numbers in sequence, and all sets had a single numeric value, and eliminated recursion, the CODIL memory would become equivalent to the stored program computer memory.  

Friday, 8 July 2011

Brain Storms - 4 - Requirements of a target model

The human brain consists of a very large number of interconnected neurons and is capable of handling a large number of sophisticated concepts. Animals (by which I include the great apes and possibly all vertebrates) have brains made up of similar neurons and can take actions which relate to their memories of the environment in which they live. As the earlier posting, The Black Hole in Brain Research, suggests little is known about the way in which changes at the neuron level use these memories to work their way through to the resulting complex decisions leading to actions.

One way of trying to find out what is happening is to try and build a model of the likely processes and memory structures and find out what the model can do, and how far it fits with observations. This post aims to outline one possible approach to modelling the relevant thought processes and some of the observations which a successful model will need to explain.

We know enough about the brain to know that we need a large number of processors each of which is linked to a memory which can store a pattern, with communication links between processors.. The simplest action would be that a processor receives a signal over a link from another processor, compares the signal with its memory, and depending whether there is a match sends a signal to another processor. This can be considered as a single neuron acting as a logic gate.

The aim is to see how far we can model different aspects of brain activity by expanding this simple model in a way that will support complex decisions and actions. In the subsequent postings I will be looking at how research on an unconventional computer language, called CODIL, carried out some years ago, provides a good starting point for such an approach. However the work done so far falls far short of a complete brain model and it is appropriate to list some of the “goals” which will need to be addresses so that that areas where the model needs to be improved can be identified.

  1. The model must start – like a new-born baby – knowing nothing and be able to boot-strap itself up to a fully working system.
  2. A built in learning mechanism is essential
  3. There must be a mechanism to recognise patterns
  4. The pattern matching process may sometimes be exact, but will sometimes need to be “fuzzy” especially in early learning stages.
  5. There must be some way of organising patterns into categories and events.
  6. Information relating to time and distance need to be handled on a relative way.
  7. At any one time there needs to be a focus of activity (around short term memory?) but some processes will be “subconscious”
  8. Counting (beyond 1, 2, many) and formal mathematics and logic will only be relevant as tasks which the model can learn to do.
  9. There will need to be some housekeeping activities – including forgetting!
  10. The model should be capable of evolving from simple models.

Human language raises some interesting points. The brain's neural net working in parallel while speech is a sequential process. If information is to be transferred from one human to another by speech there has to be conversion processes (parallel > serial > parallel), introducing syntax. An important question is whether there is any need for special facilities in the brain to carry out this translation type task. The model will need to be able to support different cultures and languages and people with very different skills or political/religious views.

There is one factor that needed to allowed for in discussing any model. This is recursion. I have not defined a “pattern” or how it is represented – but in discussing examples it will be useful to be able to say that a particular pattern identifies a concept equivalent to the English word “Elephant”. However, in the same way that the word “Elephant” can be broken down into individual letters the pattern may also be able to be broken down into separate sub-patterns. For this reason I will use the term “Node” to represent a processor/pattern combination without needing to ask whether it is based on a single neuron, or a linked network of neurons, where the pattern is actually defined by links.