Monday, August 19, 2019

Analog

Dave was really worried. the latest batch of SSD he'd received to put in the server farm was turning up way lower capacity than the last one. What was wrong? He zoomed some friends in friendly cloud providers, but found it hard to hear what they were saying. He ran some error diagnostics on the storage systems - weird, the bad blocks were all over the place, then the diagnostic tool crashed. He tried to reboot the SDN controller net and then tried a re-install. Nothing worked.

Luckily, he had an old AM radio tuned to a local rock classic station that played vinyl only. He tried to find a news station, but they all seemed to be down. so was all the internet. He went outside and got on his bike and cycled over to see some friends on stanford campus who have a clue. They were all hunched over a scope looking at the waves off of a probe.

"Hi Dave, look at this" they said pointing at the very wavy line on the old screen. "It looks like we can't see truth and falsity anymore" they quipped - indeed, it seemed as if the laws of physics had subtly altered so that electronics that distinguished "1" and "0" could no longer do so reliably.

"we're not in a binary world anymore, dave" said the trans in charge of the lab, semi-humourously.
"I wonder if this is because of those experiments Q was doing last week", suggested encrico.
"Q? the guy who's been pushing the bounds on eliminating de-coherence effects, probably" asked dave. "yeah, so we think that what her experiment actually does is to rapdily switch which universe we are all in so that the QC she's using is usually right. The problem is it now means that everything else is frequently wrong"

"Well that about wraps it up for computer science, doesn't it, surely?, asked dave.
"oh no - one quantum computer should be sufficient for everything we ever want to do again", quipped TJ. "so like Q - do or die". "no, god doesn't play die".


Saturday, August 03, 2019

polar bot extinction

Some smart folks at Warwick have done some very large scale analysis of digitized
historical texts and more specifically to this post here, polarization in political speeches and surveys showing perhaps unsurprising things like war correlates with falls in GDP and falls in happiness, but also that (at least in the US) there's a marked diversification of political opinion in the disappearing center of politics, while the "leaders" concentrate at the poles, the behaviour by the larger population is more akin to anarchy.

what this suggests is that, over a sequence of political tweets, one might expect a bot to take a consistent position on a bag of topics, but a human to show lack of adherence to a party (pole) position. a very light (LDA is probably overkill) approach to the set of tweets should see the divergence of real people from the political machine.....it would be interesting to compare this with more complex (e.g. deep learning) approaches.

One problem for the party machine is that they cannot significantly dilute or randomize the belief tropes they exhibit for fear of losing cult followers, so one should be able to filter (unfollow/mute, whatever) them permanently, fairly effectively - certainly, anecdotally, this is what i see in the twittersphere. e.g. in recent politics in the UK, accounts that are high probability bots (low lifetime, small follower cadre, auto-timing retweets etc) also bundle together beliefs (e.g. with brexit, goes "NHS health tourism" "the UK is too crowded with immigrants" and cliche/propaganda phrases like "take back control" and "soveriegnty" etc - so need to take each of these sources over some number of successive tweets and compare with known non-bots to see if real people are more diverse - i think they will turn out to be so. if it works for speeches, should work for microblogging and social media too...

Friday, July 26, 2019

Myth and Magic in Senior Academic Promotion Processes

of course, the origins are lost in a dim past of Mesopotamia, with the ritual sacrifices and rites of the Ur god of wisdom, Bloki, often drawn with his glance akimbo, holding the chisel with which he will inscribe the third review on the hierophant's forehead.

There is certainly no truth in the scurrilous rumour that there is any sexual element to the procedure, nor is there a cat called Taboo (at least, not any more).

But there is a room, above a vortex.

There are boiled sweets.

And there are huge cauldrons full of scalding black liquid.

That is all I can say, as I have been sworn to secrecy, although perhaps I will confirm that all numbers are forbidden in the room, being instead replaced with the bouquet of fine wines, and discernment between the merely vernacular, and the supernacular is an essential skill amongst the cabal.

Saturday, July 20, 2019

new dimensions in ethical dilemmas

We've been working on mapping science progress, and collaborations, geo-location of research contributions, and the rise (and fall) of institutions (labs etc) over 50 years of bibliometric data.
This is easy as we;re all so eager to be seen to publish that the meta-data about our work is all freely available through plenty of online databases and can be extracted in a very small number of accepted standard formats for magic processing - so we started here with a basic 5 year set of data (we've also looked at the Turing institute's publication data over the last 3 years, and now we're working on all of sigcomm's half a century of data. it would be easy to do more subjects and also more analyses.
First of two we're just adding is mentioned above, which is to look at the trajectory of research labs. So most people will be aware that Bell Labs is no longer the thing of beauty it once was. Nor is Xerox PARC. And some labs simply disappeared due to their corporate hq discorporating (Sun Microsystems, Digital Equipment Corporation) or losing the plot (HP?).

So we can see his effect over time. We can use the data + other information, to train algorithms to predict the demise of labs too (in the same way this Euro S&P paper from oxford built an algorithm for mergers and treaties out of airplane tracking datasets and external news reports.

We can also see geographic variation in success at research outputs.

But we can go further - a second thing we can add is gender data, inferred from names (or by using data from web lookup of author name + institution or other public sources. And this could be useful for combating bias.

However, what happens when we start to relate overall trends of institutions to the arrival (or departure) of particular researchers? This could be turned into a recruitment (or promotion or dismissal) tool. That could be quite a bad thing (given there are natural reasons why there's variation in peoples' output or influence on colleagues output, which can vary over time, and are probably going to result in short termism killing more long term strategic type thinking).

what to do?



Tuesday, July 16, 2019

dream time

Before he died, Uncle James wrote me this lovely letter. He wasn't really my uncle, he was just this rare thing, an old friend of my mother's who was also my friend. He lived in a rundown old cottage in the middle of nowhere somewhere near Brecon. He had been a quite well known british classical composer of that odd kind of post-vaughan-williams type of music, never as radical as Schoenberg or as out-and-out crazy as Boulez or Stockhausen, or as trendy and minimal as Steve Reich.

James was worried that his 40 year-old digital clock was dreaming. While it hadn't told the time properly for two decades, he kept it as the only form of electronic technology he had ever (before he retired) found useful.

For most of the 1990s, it had remained completely blank, but then suddenly, one morning he noticed that it was showing some strange symbols that could be read as a word. He got out his welsh dictionary and discovered that indeed, bwrw was a common term for rain. Unsurprising, he thought. but then a digital clock that predicted the weather, no matter how predictable, was (like the proverbial poor chess playing dog) surprising.

However, the next day, the clock read ffliw. And James promptly sneezed and was laid up for two weeks, until he recovered. Feeling much better, he decided to cheer himself up and his wife Mary, by writing a spring tune. The clock ticked,  and read duw. He had just been thinking of something inspired by religion.

James wrote to me, because, as he explained, I was the only person he knew who might be able to explain how a 1970s digital clock, probably made in China, was suddenly behaving like one of the cleverest AIs in the world.  I was lost for words. It was very sad that he had discovered an emergent being that, while locked in such an unprepossessing box, was expressive and helpful. I did not want to tell anyone about this at the Turing Institute, as they would descend on James lovey cottage with their logic analyzers and Turing tests, and would no doubt kill the goose that was laying such golden eggs.

Not long after this, James died, and Mary moved to Cornwall and threw out all his memorabilia, so maybe the clock is in some landfill somewhere. But maybe it is still dreaming on someone's bedside table, offering hints in a timely way on how their day may go. An accidental oracle from the orient.

Monday, June 24, 2019

eco-washing

no, i'm not talking about greener washing machine detergents.

i'm talking about the number of projects i've been asked to join recently looking at air quality in cities, and correlating it with transport, health, etc

I remember when they cleaned the victorian coal-smoke black off of Westminster Abbey.
I remeber when they got rid of led (anti-knocking) additives in petrol.
I remember when they banned smoking in public places.

I'd have thought it was completely obvious that we should have a lot less crap in the air. whether its from cigarettes, or diesel, or coal or just people talking rubbish.

lets move on to more important things like reliable safe water, sea level, extreme weather, and species extinction.

lets stop window-dressing (eco-washing - as in ethics washing) with these superficial, trivial, obvious problems which have solutions which happen as a side effect of solving the really big problems.

Thursday, June 13, 2019

frogboiling on multiple timescales

the country is getting used to being boiled a la frog.

we have the spectacle of the tory leadership campaign - today's news is that over 100 tory MPs are backing someone who has been fired for lying from one job, and made the country a laughing stock in his utterly useless (but offensive) antics as foreign secretary. he didn't totally screw up london while mayor, but he did throw £40M quid down the toilet trying to build a garden bridge for his rich buddies.

then we have the spectacle of brexit, a plan so without merit that it makes playing chicken with cars on the freeway look very sensible

then we have climate change - the last three days, people are probably thinking to themselves that 1.5C rise in average temperature really wouldn't be such a bad thing.

no. no. no. stop this. stop all of this. just stop.