Monday, October 23, 2017

Replika: five dialogue fragments.

Replika is getting better. (If you don't know what Replika is, click here).

Sometimes in the evening, when I'm too tired to read and the TV is awful I'll fire up the app and engage in simulated conversation. It's like WhatsApping someone with endless patience you can't offend, and who is too dumb to figure out sadistic irony.

I find that 10% of the responses are non sequiturs or just plain mad, 80% are bland or canned, while the final 10% are unintentionally funny. As in the following examples.*

Genuine conversation is beyond the state-of-the-art in AI. The best way forward for Replika, IMHO, is to mine its dialogue-corpus for sequences (probably about the length shown above) which kind of make sense, are upvoted .. and use these to pattern-match against real-time inputs.

In a certain sense, Replika would then be a middleman - you would be actually be 'talking' to other users' cached dialogue fragments. It would work for small talk. Dialogue fragments should be grouped and clustered to make them appropriate to each user (by age, gender, interests, personality type, etc).

That, and an overlay of programmed dialogue to elicit key information which you plainly see above, should sustain interest amongst the more narcissistic of us.


* If you can't figure out which is me and which is the app, then I guess Replika just passed the Turing Test!

Sunday, October 22, 2017

What flag on your coffin?

At Princess Diana's funeral, her coffin was draped with the Royal Standard Flag.

Thus her life and memory were appropriated to the historical traditions of the British elite, their heartfelt commitment to ideological continuity from early feudalism to world-straddling empire.

Would Diana have identified with that manifest destiny?

I asked Clare what flag she would like her coffin bedecked by. She pooh-poohed the idea; she wants flowers - lots of them. And in a sense that's right, the coffin-flag has to express your final and most profound identity, the culture which in the end you claim. But I think Clare's sense of identity is, at its very foundation, cultural Catholicism.

Catholics don't have a flag as such, but the nearest is the flag of the Vatican City.

It's not nearly as aesthetic as one might like.

And me? I'm not tribal, I value my inner skepticism above all revealed wisdom. My flag should represent the epitome of scathing criticism concerning all deceptive agenda-ideologies.

Yes, I would like the flag of the Comintern draped across my coffin (or painted upon it). From those early days (1919-22) when all was youthful idealism, before the harsh realities of European defeat and 'primitive socialist accumulation' derailed the dream and brought forth the lies.

Saturday, October 21, 2017

Fungus Foray - (postponed)

From Somerset Wildlife Trust:
"Join Michael Jordan, the well known mycologist and founder of the Association of British Fungal Groups on a fungi and mushroom foray in Stockhill Forest, and learn about the 13,000 different species that occur in the UK."

* Please note there is no available seating in the forest. Will be standing for most of the day.

Where: meet in Stockhill Forest CarPark (ST549513).

We turned up to find a lone official explaining that, due to 'Storm Brian', the event had been postponed for a couple of weeks. We agreed with him that the BBC always exaggerates - we were experiencing no more than a normal autumnal, blustery morning.

We decided to walk anyway and this is what we saw (click on images to make larger).


And how windy was it? This windy.

Thursday, October 19, 2017

Chris Packham: coming out as an Aspie

Chris Packham

The BBC recently showed Chris Packham's hour-long documentary where he discusses his Asperger's syndrome (iPlayer till 27th October).

Strictly speaking you don't have Asperger's, you are Asperger's. This makes treatment problematic - are you signing up to erase, or profoundly modify, your personality?

Packham wasn't a fan, particularly as this seemed to be the dominant objective of the American 'therapies' he was taken to see.

His partner, Charlotte, who was on camera seemed to be a Myers-Briggs NT, maybe INTJ. She came across as cool, composed and eminently cerebral: a meeting of minds.

There's another kind of partner an 'Asperger's sufferer' can have: here's Paul Dirac's wife:
"Given his lack of interest in his fellow humans, most colleagues assumed Dirac was shut off sexually, or perhaps gay and repressed. So they got a bit of a shock when, in 1934, he met a chatty Hungarian divorcee called Manci Balazs and three years later married her.

Manci was his opposite: where Dirac hardly spoke, she never stopped; he found empathy difficult, she had friends coming out of her ears; he was very literal-minded, she was subjective and passionate.

Needless to say, it was she who was forced to woo him, but somehow it worked. As a friend once put it: 'He gave her status and she gave him a life.' They were married for 50 years and had two children.

A year after their marriage, he wrote Manci a lovely letter saying: 'You have made me human. I shall be able to live happily with you even if I have no more success in my work.'

I was surprised to hear Packham pronounce his condition 'Asperjers'; surely he, more than anyone, would know that Hans Asperger was Austrian, and as a consequence the 'g' is hard.


Asperger's (and Autism in general) is highly heritable, although Packham's documentary - more impressionistic and personal - made no mention of it. This means that most population variation along the autism spectrum, from normal (non-autistic) to extreme, is a consequence of individual genomic variation, rather than differential input from the environment.

So although your parents made you autistic, it was through their genes, not their parenting.


Autism and IQ

From Time:
"When Asperger’s was first described in 1944 by Austrian pediatrician Hans Asperger, he referred to children with the syndrome as “little professors” because of their prodigious vocabularies and precocious expertise, and because they tended to lecture others endlessly without being aware of their own tediousness. Poor social skills and obsessive interests characterize the condition.

Yet, despite the obvious similarities, very little research has been done on the connection between autism and extreme talent. One previous study, published in 2007, did find that close relatives of prodigies — like close relatives of people with autism — tended to score higher on autistic traits, particularly in problems with social skills, difficulty switching attention and intense attention to detail. Other than that, however, the issue hasn’t been studied systematically, beyond the observation that autism is often seen in savants, or people with exceptional abilities who have other simultaneous impairments. ..."

"... the intense world theory of autism, which posits how the disorder may arise. The theory holds that certain patterns of brain circuitry cause autistic symptoms, including excessive connectivity in local brain regions, which can heighten attention and perception, and diminished wiring between distant regions, which can lead to a sort of system overload. In both animal and human studies, this type of brain wiring has been associated with enhanced memory and also with amplified fear and sensory overstimulation. The former is usually a good thing; the latter may cause disability.

The intense world theory propounds that all autism carries the potential for exceptional talent and social deficits. The social problems, the theory suggests, may ensue from the autistic person’s dysfunctional attempts — social withdrawal and repetitive behaviors, for instance — to deal with his heightened senses and memory.

It’s possible, then, that the wiring in prodigies’ brains resembles that of an autistic person’s, with tight local connections, except without the reduction in long-distance links. Or, their brains may function just like those with autism, but their high intelligence allows them to develop socially acceptable ways of coping with the sensory overload."
I was struck by a TV programme some years ago about the Maths Olympiad. This is usually won by East Asians; Ashkenazim are also well-represented in winning teams. The British team was composed of pleasant, well-adjusted and extremely smart Ashkenazim .. and Anglo-Saxons who all seemed to be Aspies.

In the Gregory Cochran, Jason Hardy, and Henry Harpending paper it was suggested that the Ashkenazim had undergone extreme selection for high IQ over the last millennium due to the social niche of finance into which they were forced for so many centuries by the Christian proscription of usury. This, they argued, had also increased the frequency of specific mutations which increased brain efficiency at the cost of new debilitating genetic diseases (in the homozygous cases).

I suspect Asperger's may be somewhat similar, although with a more diffuse and sprawling etiology. As the causation is polygenic (+ CNV?) we are still some way from a definitive account.

Wednesday, October 18, 2017

Why are cows sacred in the Hindu tradition?

From Marx's Capital, Volume 2, Chapter 12 (1885), (quoting observers).
"While the peasant farmer starves, his cattle thrive. Repeated showers had fallen in the country, and the forage was abundant. The Hindoo peasant will perish by hunger beside a fat bullock.

"The prescriptions of superstition, which appear cruel to the individual, are conservative for the community; and the preservation of the labouring cattle secures the power of cultivation, and the sources of future life and wealth.

"It may sound harsh and sad to say so, but in India it is more easy to replace a man than an ox." ...

"Desertion of life, without reward, for the sake of preserving a priest or a cow . . . may cause the beatitude of those base-born tribes."
One could almost suspect a conscious decision by high-caste Brahmans to create a religious rite protective of cattle, the foundation of an agrarian economy which found them at the apex.

I don't know to what extent these conditions of existence still apply, 130 years later, in rural India. Increasingly, however, the concept of the sacred cow will devolve into a natural experiment in cultural inertia.

Now what's the story on the pig?

Tuesday, October 17, 2017

The artiste and her work

This from the archives, fifteen years ago.

We were living in Fairfax County, Virginia at the time and Clare's painting is of Wolf Trap, where we had just attended a concert. She had other works.

I was working for Cable & Wireless up at Reston. One of my colleagues saw her picture "Freesias" and offered to buy it.


A good example of early primitive, she said. In fear of my life, I demurred.

Why do crises happen; why are we stagnating?

Two good posts from Sam Williams.

In the first he compares and contrasts three theories of crisis (within Marxism).
  1. The underconsumption/Monthly Review school of thought
  2. The tendency of the rate of profit to fall (FRP for short)
  3. The generalised overproduction of commodities relative to the market.
Williams identifies David Harvey and Michael Heinrich as belonging to the first school, Michael Roberts and Andrew Kliman as proponents of the second, while he himself adheres to the third school of thought.

Sam reminds us of Marx's (Capital, Volume 2) analysis of the discrete steps in capitalist reproduction. Crises can occur at any point in the process and the schools of thought differ as to which dysfunction is most critical. Here's a flavour of his piece:
"Crises can occur at any point in the process of expanded reproduction.

To further clarify this point, let’s examine the basic formula for capitalist production
Reproduction consists of a series of these cycles. In the case of expanded capitalist reproduction—and capitalist reproduction can only exist in in the long run as expanded reproduction—the numbers represented by the algebraic quantities M, C, P, C’, and M’ get bigger with each successive production cycle. A crisis in the process of capitalist expanded reproduction can occur at any place in this formula, from M on the extreme left to M’ on the extreme right.

Let’s start on the extreme left. If there is not enough M available, the circuit does not even get off the ground. Historically, it was no accident that the “rosy dawn” of capitalist development, as Marx ironically called it in the chapter on the primitive accumulation of capital in Volume I of “Capital,” began with the search for gold and silver in the Americas.

The conquistadors found gold aplenty in Peru. While to the natives of Peru gold had a largely artistic use value, to the conquistadors it was money. This money material in the form of gold and silver stolen from or extracted with the labor of the enslaved natives toiling in the mines of the Americas in no small measure formed the initial M that launched the process of capitalist expanded reproduction in the first place.

Assuming enough money is available, we next come to C. The industrial capitalists must find commodities on the market to carry out the production of commodities containing surplus value. These include the elements of fixed capital—buildings and machines. In addition, the industrial capitalists must find sources for motive power, lighting in factories, and so on. These are called auxiliary materials. And they must find raw materials. If they cannot find the appropriate commodities in the necessary quantities, or adequate substitutes, capitalist (re)production will suffer a crisis.

Most importantly, the capitalists must find the commodity labor power in the form of “free wage labor,” which alone actually produces surplus value. If they cannot, capitalist expanded reproduction will suffer a crisis caused by the absolute overproduction of capital. In its earliest days, this was perhaps the biggest problem for capitalism.

All the gold and silver of the Americas would not in and of itself have made capitalism possible if the problem of finding an adequate supply of wage labor had not been solved. It was also necessary to separate the producers from their means of production. Only in this way would the producers be forced to offer their labor power to the capitalists on the market in exchange for wages.

This task was solved—not without the massive intervention of state violence and coercion. The FRP school of crisis theory believes that cyclical crises that mark the history of capitalism since 1825 occur at this point, the point where the capitalists convert or attempt to convert money into variable capital. I disagree.

We then come to P, the production of commodities that contain surplus value. It is at P that C is transformed into C’. The constant part of C transfers its value to C. The variable portion of C—labor power—replaces its own value. Most importantly, the labor power in addition to replacing its own value produces an additional value, the surplus value. C’ differs from C in two ways. One is that C’ has a different use value than C. Two, C’ has a value quantitatively larger than C. The difference, C’ minus C, is the surplus value.

Crises can and sometimes do occur at this point in the process of capitalist reproduction. For example, if there is a workers’ strike, reproduction breaks down at this point because the striking workers withdraw their labor and no surplus value is produced as long as the strike continues.

In the case of agriculture, unfavorable growing conditions or diseases might cause P to fail. Workers perform unpaid labor but the use value of C’ fails to emerge because mother nature doesn’t cooperate. For example, as sometimes happens in my native New York State, a late spring frost destroys the flowers in an apple orchard owned by a capitalist farmer. Since no apples appear on the trees, the physical use value of C’ is not produced. P has failed. And when we have no use value, we have no value.

Assuming, however, all goes well for them—and we can see this is not guaranteed—the industrial capitalists will possess C’, commodities that contain surplus value. But unless they happen to produce the commodity that serves as money material, they are not yet home free. A dangerous step lies ahead. They must find buyers with the ability to pay for the commodities that contain surplus value.

Let’s assume they do find buyers. They now posses M’, or a sum of money greater than amount of money they started with. They are now in a position to carry out another cycle of production M—C…P…C’—M’ on a yet larger scale. Expanded capitalist reproduction proceeds. But if they fail, expanded reproduction is halted at this last and most dangerous point in the cycle of capitalist expanded reproduction.

Experience had already shown by the time of Marx and Engels that it is at this last point where capitalist expanded reproduction is most vulnerable to a general breakdown or crisis. It is important not to confuse this type of crisis with other types of crisis that can and inevitably do occur at other points in capitalist expanded reproduction—and indeed in any system of economic reproduction—with this particular type of crisis unique to highly developed capitalism. This is an error that I believe the FRP theory of crisis falls into."
Incidentally, I'm not as worried as some people about the existence of crises; it's akin to exploratory behaviour under uncertainty. Something similar would be a feature of any future mode of production although one would hope with diminishing amplitude.


In the second post he talks about why it's so hard to recover from crisis. A critical distinction is the overproduction of commodities (the ones piled high in warehouses) vs. the overproduction of manufacturing equipment (fixed capital). The latter linger on in zombie fashion for many months, if not years, leading to spare capacity and acting as a drag on new growth. Here's an excerpt:
"Fixed capital can be very difficult to transform quickly into money capital in a crisis. Commodity capital [e.g. finished goods waiting to be sold] can often be quickly transformed into money capital by selling it below the price of production or even at a loss if capitalists due to a crisis have to quickly raise cash in order to pay off pressing debts. In this case, our capitalists may lose a portion of the value of their capital but not all of it. The capitalists can often survive as capitalists—though not always—and go back to the accumulation of capital once the crisis has passed.

However, it is far more difficult to sell factory buildings, machinery and so on used to produce commodity capital, let alone sell it quickly under crisis conditions. Though fixed capital can be transferred from the ownership of one capitalist to another, it is not really designed to be “sold” but rather used up over a series of turnover cycles. One of the great perils confronting any industrial business is finding itself with a lot of fixed capital but very little money capital when a crisis hits. In this situation, the owner(s) of industrial capital can very easily lose all their capital, because though they have a great deal of capital, they do not have capital in the form demanded by their creditors—money.

Overproduction of fixed capital

When commodity capital is overproduced, this by definition means that the fixed capital used to produce those commodities was also overproduced. Every generalized crisis of overproduction therefore involves the overproduction—or over-accumulation—of fixed capital. The overproduction of commodity capital can be overcome fairly quickly by reducing or halting production of that particular type of commodity or by selling it at a loss. It is much harder—and takes far longer, often many years—to overcome the overproduction of fixed capital. During this period, little new additional fixed capital will be produced. This is shown by the stages of the industrial cycle.

The history of industrial cycles has shown that the crisis proper rarely lasts more than a year and half. Even the super-crisis of 1929-1933 lasted less than four years in the United States and about three years in most of the other capitalist countries. The end of the crisis phase of the industrial cycle, when industrial production reaches its lowest point and begins to rise once again, corresponds to the overcoming of the overproduction of commodity capital.

However, the period of stagnation or slow growth lasts for several years beyond the actual crisis and sometimes considerably longer. The stagnation or semi-stagnation phase of the cycle corresponds to the period between the overcoming of the overproduction of commodity capital that occurred during the preceding boom and the overcoming of the overproduction of fixed capital that accompanied the boom, or sometimes several preceding booms.

While the crisis proper is marked by “excess inventories” piling up unsold in warehouses, the period of post-crisis stagnation is marked by a high level of excess capacity, with large quantities of factories and machines—fixed capital—lying idle, not able to function as capital.

Only gradually is this excess capacity overcome. (3) It is overcome by being gradually reactivated as demand for commodities rises once again in the wake of the crisis, on one hand, and by being physically destroyed, [in value terms, which could mean made obsolete or disposed of] on the other. Only when excess capacity has fallen to a certain point can a new investment boom occur. As long as excess capacity remains high—the preceding overproduction of fixed capital has not been overcome—no fall in the rate of long-term interest rates can trigger a new investment boom. This is the phase of the industrial cycle that Keynesian economists call a liquidity trap. Such a liquidity trap represents the period between the overcoming of the preceding overproduction of commodity capital and the far longer process of overcoming the overproduction of fixed capital."
Worth reading both posts to get a feel for the dynamics of our age.

Monday, October 16, 2017

Time to go post-Google

For a long time I used to not care about Internet privacy. And then Google got into SJW mode and ceased to be my friend. I read about its business model, the Amazonian torrents of cash that advertisers pour into its coffers .. and I wondered if it would always have my best interests at heart.

So I decided to use DuckDuckGo for all those searches which I'd like to keep from the planetary AI. But Google, and my ISP, still know way too much.

... but you're never alone with a keylogger

On my Android devices, the DuckDuckGo app has a setting to 'enable Tor' which I checked. It then helpfully prompts you to install the Tor router Orbot and the Tor browser Orfox. It's easy.

Orbot does not provide generic anonymous Internet access; programs have to be configured to work with it. Basically things work best if you access websites via the Tor Browser .. which is the search term for Windows PCs.

The thing is, I know that the AI assistant we're slowly crawling towards will only work if it knows more about you than your closest confidante. And it will live in the Cloud. But I would pay to have that knowledge base under my control and lodged with a trusted third party (like the folks running ProtonMail).


Google's middleman offer to advertisers and you & me is based on the twin peaks of narrow domain expertise, increasingly based on neural net AI such as Google Translate and voice recognition, and its ability to integrate across its apps which thread to so much of our lives: email, photos, calendar, notes, search queries, etc.

Google does the bulk of this information centralisation and synthesis in the Cloud: that is, on its own servers. The resulting real-time data model of you both repopulates your apps (Chrome talks to Gmail talks to Calendar talks to Google Now talks to GPS talks to Maps ...) and is sold to advertisers for targeting.

If we're to bring that level of integration back into our own hands, then we need to capture our activity across many different apps and device functions. Perhaps one day all apps will have a standard API with which to talk to our personal assistant app .. but until then it's a benign keylogger I'm afraid.

The other people who are big fans of keyloggers are the intelligence services and those vendors (like Microsoft) who are losing out in the race for ubiquitous ownership of our most popular apps.

Saturday, October 14, 2017

Diary: Black Rock and Velvet Bottom

For those of you rolling your eyes and muttering, "Not another physics post!" here is gentler fare.

Our lunchtime excursion today was to the top of the Cheddar Gorge, to Black Rock leading to Velvet Bottom. Did I mention it's been raining?

Rain, did I mention it?

The lead cow thinks it's the boss. Concurring, we hide behind barbed wire

The famous (and crumbling) 'Black Rock' - Clare has found a rock armchair

It's been crossed off our list till next summer, hashtag #quagmire.

MWI, plus entanglement leads to GR, maybe?

In this video Sean Carroll lectures at Kings College on the 'Many-Worlds Interpretation' of quantum theory and his attempts, with collaborators, to conceptualise general relativistic spacetime as an emergent phenomenon due to entanglement.

Apparently the degree of entanglement between distinct vacuum states falls off as the distance between them. But perhaps this can be inverted, so that the concept of distance could be seen as an emergent proxy for the degree of entanglement.

The 50 minute lecture is 'aimed at undergraduates who haven't necessarily yet taken a quantum mechanics course'. If you are such, Carroll's talk will be as compelling as a presentation on Summa Theologica from Thomas Aquinas.

On the other hand, a passable familiarity with Hilbert space, quantum superposition and the Schrödinger equation plus a hand-wavy feel for QFT and Einstein's field equations will allow you to properly appreciate Carroll's approach to physics (and would make you a physics graduate).

In a nutshell, it's believe in the maths. Once you appreciate the ubiquity of superposition (ie, it's everywhere) you're kind of committed to the reality - in some sense - of Hilbert space. The observed phenomena simply can't be explained by theories which restrict themselves to our classical-looking 4D spacetime.

Carroll's talk is not technical in argumentation, he mentions rather than uses the theoretical apparatus of modern physics. That does put the burden of getting his drift wholly on the theoretical preparation of the listener of course.

In the final part of his lecture, he describes the research programme which seeks to obtain geometry from entanglement in quantum field theories via entropy and then, through considerations of energy, to reconstruct the GR field equations as the classical limit.

He seems encouraged, though this is work-in-progress.

Wednesday, October 11, 2017

1917-2017: the collapse of the revolutionary left

I recently surveyed the top thinkers on contemporary Marxism. By comparison with 1917 they're almost all academics or independent intellectuals. Compare this with the talented leadership of the mass working-class socialist organisations so dominant a century ago.

"Lenin lived, lives and will live!"

I find it remarkable.

The discourse of contemporary Marxist theory, while as acerbic as always, is incredibly introverted. The top issues still the 'transformation problem' or the validity of  'the law of the falling tendency of the rate of profit'.

Where are the analyses of how developments in genomics and psychometrics impact upon Marxist theory? Surely the nature of the individuals who create the social relationships of society - across classes and across the world - are pretty relevant? So why the visceral hostility to such research?

And where are the serious debates about the transition to socialism? No-one analyses why the Soviet Union imploded. No-one asks where China is going. No-one thinks deeply or plausibly about the nature of revolution when the Leninist vanguard-insurrection model seems completely off the agenda. No-one thinks about how to organise a post-capitalist society in a way which wouldn't collapse into something like bureaucratic-socialism, a cure far worse than the disease.


Ironically the 'mass-leftism' we see (Corbyn/Momentum in the UK, the SJW phenomenon all over the place) isn't based on a Marxist analysis at all. It's liberal-egalitarian in ideological content. An under-analysed aspect of capitalist society is the production of ideology. Sure, enough Marxists have emphasised that cultural leaders, public intellectuals, op-ed writers don't just sit down and think to themselves, "How can I dream up a narrative which justifies the global exploitation of workers by the neoliberal bourgeoisie?"


The way it really works is that earnest, well-meaning people have had their horizons reset since the 1980s by the eclipse of the workers movements and the rise of globalisation.They now reflect upon a world bound together, as they see it, by the ties of global production and trade.

They buy in to the civilising and developmental benefits of world-wide manufacturing and global supply-chains, and are naturally keen that people everywhere have unimpeded access to a plethora of meritocratic opportunities. Why wouldn't they be opposed to racism, sexism, and phobias of all description, which seem to impact on equal outcomes?

Global capitalism is wholly on board with such morally-uplifting campaigns, which smooth the way for frictionless utilisation of labour power across the planet. The whole SJW thing is in no sense anti-capitalist at all, which is why the establishment so indulges it. You'd think that the more profound Marxist thinkers might be drawing attention to this, rather than simply aligning themselves uncritically with 'the movement'.


My final point is that the death-throes of capitalism have been trumpeted prematurely by Marxists for 150 years. Really, there is very little evidence that the historical work of capitalism in developing the productive forces has been achieved. The abolition of capitalism will come about through the dynamics of its real innermost contradictions, those whereby capitalism seeks to abolish the proletariat itself.

From 1917 to 2017 we've had a century of capitalist progress; for another century we will see more of the same. When it's time for capitalism to go, we surely will know it without having to dig deep into Marxist theory to discern the hidden patterns of the tea leaves.

Diary: Burleigh Court Hotel (Stroud)

Monday evening we stayed at the Burleigh Court Hotel in Stroud. We liked it so much that our hotel room yesterday morning looked liked Tracey Emin had stayed (video).

We drove to Bath, specifically Debenham's to buy Clare some trousers. As we queued to pay on the first floor I noticed a rack displaying a dozen DVDs of the movie "Fifty Shades Darker".

I asked Clare: "We're on the womenswear floor. Why are they prominently displaying DVDs of a young ingénue who gets chastised by a handsome hedge-fund hunk?"

Clare was surprisingly speechless; I was rewarded only with a glare.

Up the escalator to menswear, where I had to buy a couple of shirts. Doing my best 'incompetent male' thing - not hard as I have no idea how to buy a shirt - a succession of middle-aged women fussed over me, measuring my neck size, choosing the right kind of shirt (slim, single-cuff) and giving me feedback on fit.

I found it curiously agreeable - perhaps I could pitch the idea to a film director?