“Wouldn’t take ‘fuck off’ as an answer.”
Two interesting machine learning/AI challenges (emerging from a chat with my former PhD student Lawrence Beadle yesterday):
- Devise a system for automatically doing substitutions in online grocery shopping, including the case which recognises that substituting a Manchester City-themed birthday cake is not an adequate substitution for a Manchester United-themed birthday cake, despite them both being birthday cakes, of the same weight, same price, and both having the word “Manchester” in the name.
- Devise a forecasting system that will not predict that demand for turkeys will be enormous on December 27th, or flowers on February 15th.
Both of these need some notion of context, and perhaps even explanation.
Is there such a thing as a set of skills that apply across all of the professions? When I first started to come across (still rather rare) university departments of “professional practice”, I was bemused. Professional practice in what? Is there really enough common to being a nurse, barrister, dentist, accountant, town planner, occupational therapist, etc. etc. to call all of their activities “professional practice”? These seem, at least initially, to consist almost entirely of a lot of profession-specific skills/knowledge/understanding.
But, over time, I’ve started to wonder. Perhaps we are at the stage with professional practice schools that we were at with business schools a few decades ago. There was certainly a cynicism at one point that “business” could be taught generically. What business? Is there really enough in common to running a bassoon factory, a chain of gyms, an online career consultancy, an au pair agency, etc. etc. to call all of their activities “business”? At one point, these would have been seen as needing radically different skill-sets, but over time we have started to realise that some common understanding of finance, accountancy, PR, marketing, project management, strategy, staff appraisal, etc. are useful in all areas of business, alongside a knowledge of the specific business domain.
Perhaps there is something to be gained by bringing together dental nurses, architects, and solicitors for part of their education, and having some common core of education in e.g. dealing with clients. Perhaps the idea of a generic professional practice school isn’t such a ludicrous idea after all.
A shop called “BARGAIN BOOZE” is pretty bleak, but at least you know what you are getting. To buy this shop, and decide to rebrand by covering up the word “Bargain” using parcel tape, and on another part of the sign tearing off the word “BARGAIN”, so that the shop is now simply called “BOOZE”, is distinctly more depressing. (This was in Sherwood; it has now all been refurbished!).
Why aren’t more legal-regulatory systems in conflict? A typical legal decision involves a number of different legal, contractual, and regulatory systems, each of which consists of thousands of statements of law and precedents, that latter only fuzzily fitting the current situation, with little meta-law to describe how these different systems and statements interact. Why, therefore, is it very rarely, if at all, that court cases and other legal decisions end up with a throwing up of hands and the judgement-makers saying “this says this, this says this, they contradict, therefore we cannot come to a well-defined decision”. Somehow, we avoid this situation—decisions are come to fairly definitively, albeit sometimes controversially. I cannot imagine that people framing laws and regulations have a sufficiently wide knowledge of the entire system to enable them to add decisions without contradiction. Perhaps something else is happening; the “frames” (in the sense of the frame problem in AI) are sufficiently constrained and non-interacting that it is possible to make statements without running the risk of contradiction elsewhere.
If we could understand this, could we learn something useful about how to build complex software systems?
First law of Exciting News: Inevitably, when you get an email from some company entitled “exciting news” it is going to contain an announcement that they have “merged with” (been taken over by) a “major partner” (a larger, rather more anonymous company), and that they are “looking forward to the opportunities that are offered by this exciting new development” (ready to make some more money from you by offering you a slightly diminished service level).
One of the points where mathematics and day-to-day intuitions jar is in estimating numbers of combinations and similar combinatorial problems. I’ve just made a booking on Eurostar, and my confirmation code is a 6-letter code. Surely, my intuitive brain says, this isn’t enough; all of those people going on all of those journeys on those really long trains, day-in, day-out. Yet there are a vast number of possibilities; with one letter of the 26 letter alphabet for each of the 6 letters in the code, there are 26^6=308,915,776 possible combinations. Given that there are 10 million Eurostar passengers each year, this is enough to allocate unique codes for passengers for around thirty years. It then makes you wonder why some codes are so long, like the 90-digit MATLAB registration code that I had to type in by hand a couple of years ago.
Love it that this bookshop in Margate manages to divide books into three categories: “General Interest”, “Extra Stock” and “Whatever” (there are some other shelves with more specific categories).
Thinking about bookshops and their categorisation schemes reminds me of a bookshop from years ago on Queen’s Road in Brighton, just down from the station, which had, in addition to books on the shelves, large piles of books in the middle of the floor as if dumped there by a dumper truck. At the back of the shop, there was a shelf of pornographic books; in place of the usual bookseller euphemism of “Erotica” as a header for the section, this shop had plumped for the rather more direct word “Filth”.
Amazingly I have just found a picture of that very shop, and an article from The Argus about its closure (well, abandonment) in 2002; the wonders of the interweb, eh?
(actually from quite an interesting article: Lessons from the A47 and the University Bubble).
I often refer to the process of taking the content that I want to communicate and putting it into the 200-by-300 pixel box reserved for content in the middle of our University’s webpages as “putting the clutter in”. I get the impression that my colleagues on the Marketing and Communication team don’t quite see it this way.
Been doing quite a bit of python programming this week. So far I have managed to mistype python as:
Bought the an album called Sex from Amazon a few days ago (by the excellent jazz trio The Necks). Inevitably, this caused the following request for feedback to appear in my inbox a few days later:
Followed, inevitably, by the following when I next went onto the Amazon website:
We all seem to have tiny little mental blocks, micro-aphasias, things that we, try as dammit, cannot learn. My mother couldn’t remember the word “volcano”—she was a perfectly fluent native speaker of English, with no other language difficulties, but whenever she came to that word it was always “one of those mountain-things with smoke coming out of the top” or similar, followed by several seconds until the word came to her. I have a block on the ideas of “horizontal” and “vertical”. Whenever I read these, I feel my mind blurring; I know which two concepts they map on to, but for a second or two (which feels like an eternity in the usual flow of thought) I cannot fluently map the words onto the concepts. Usually I break the fog by making a gesture with my fingers—somehow, this change of mode (moving my fingers from left to right strongly associates with the word “horizontal”) dispels the confusion, it must trigger a different part of my memory associations. Quite where these odd little blocks come from—and, why we can’t just learn them away—is fascinating.
I like the casual use of—often rather strong—swearing in very day-to-day situations. Last night, there was a small group of people standing outside the Sainsbury’s Local on the corner. The following conversation took place:
- Have you got the hummus?
- Yeah, sure.
- Thank fuck.
The singsongy rhythm of the last couple of phrases was particularly neat.
A useful thought-tool that I learned from Tassos Stevens: “It is easier to make the interesting plausible, than the plausible interesting.”.
When I was around 10-11 years old, my parents made a shed at the back of the garden, by putting a door and roof on a small space at the back of the garage. This was used to store gardening supplies—compost, plant pots and the like—and bottles of the dubious home-made wine and beer that was popular at the time.
One summer day I decided, on a whim, that this needed a label putting on it. So, using a chisel and hammer from the garage, I gouged the words “TOOL SHED” into the paint and wood, fairly deeply. Then, realising that the shed wasn’t used to store tools, I panicked; but a simple solution came to mind. As a result I carved the word “NOT” above the word “TOOL SHED”, with an asterisk added to retain the symmetry of four letters on each line. As a result, the shed had (and retained for several years) the label:
and was thus referred to in my family for many years subsequently.
I believe that I am the only person alive who remembers this.
In the public sector, we are oft being urged to emulate the supposed good practices of the private sector. According to Liam Fox, these include being fat, lazy, and playing golf on Friday afternoons. Should I be writing these practices into our latest strategy document?
It sometimes surprises me quite how formulaic the smalltalk at the beginnings of conversations is. I know that it isn’t acceptable to respond to the question “How are you?” with a list of your latest ailments and insecurities, but it is still sometimes surprising how much that part of a conversation is a cognitive readymade, without any ready deviation. I remember a couple of incidents in the days after my father died.
- Meeting a colleague a few days after my father had died. Wanting, gradually, to let people know what had happened, I responded to his “How are you?” with a “Actually, not so good.”, expecting to get a query back about what had happened. Instead, I just got the response “Great, I’m fine.”, as if I had said (as I would 99.9999% of the time) “I’m fine, how are you?”. Literally, my response hadn’t been processed at all. If you want some evidence for hearing being a process of anticipation then you’ve got it there. There’s no other response in the “repertoire” to “How are you?” other than minor variants on “Fine, how are you?”, so the brain doesn’t even really bother processing what has been said. Any response is just treated as the standard one.
- Speaking to my uncle a day or two after my father had died (I had already told my uncle). This time, he asked first: “How are you?”. My response, understandably: “Not too good.”. My uncle’s response—no criticism intended, this is just a point about how deeply embedded language structures are—”Oh, why is that then?”. I was, very unusually, struck dumb for a few seconds. For a moment I thought “Perhaps I didn’t tell him that Dad had died?”; for surely, someone wouldn’t say something so crass to someone who had just lost a parent—surely it would be obvious why I “wasn’t too good”. Eventually, I managed to stutter out “Well, you know, Dad died yesterday.” It is bizarre how fixed our linguistic patterns are that, even after one of the worst things that can happen to you, saying that you are anything other than “fine” causes our whole language generation system to collapse.
On this week’s Only Connect, there was a beardy chap with glasses who described himself as “having a maths degree and playing the bassoon and cello” who was on a team defined by their liking for Indian food. I had to check for a moment that I hadn’t accidentally appeared on the programme and forgotten all about it.