On Counting

Kat Jung­nick­el kindly invited me to a two day meet­ing as part of her con­tinu­ing series of Trans­mis­sions and Entan­gle­ments events. Amidst oth­ers work­ing through new meth­ods and pro­cesses, here’s what I had to say for myself on count­ing:

What is it to count and to be coun­ted?

One way I have made sense of my work over the last 10 years at Microsoft has been to see it as a way of get­ting to grips with count­ing and in some ways com­ing to terms with being coun­ted.

I could tell a few stor­ies about num­bers and counts, but let me say a bit about just two, that are, in dif­fer­ent ways, import­ant for me.

The first one is admit­tedly a dry example, but I hope it might at least set us on our way..

About 4 years ago, I found myself part of a small team of sci­ent­ists from sys­tems and cell bio­logy, and com­puter sci­ence. The chal­lenge was to take a tool that had been devised to test bio­lo­gic­al mod­els for what is known as sta­bil­ity and make it some­thing access­ible to a wider com­munity of bio­lo­gists, to those who would be deterred from work­ing with bio­lo­gic­al mod­els pro­duced through lines of code and num­bers.

I won’t go into the details of the com­pu­ta­tion here. I do want to say, though, that some­thing struck me in the work. This was how, through a very soph­ist­ic­ated way of mak­ing counts and see­ing rela­tions, the par­tic­u­lar tool we were deal­ing with had the the­or­et­ic­al capa­city to test bio­lo­gic­al sys­tems with an infin­ite num­ber of states! By manip­u­lat­ing the way num­bers or counts ref­er­ence one anoth­er, the tool could work through every pos­sible situ­ation to determ­ine wheth­er some stable end point was always achieved. Simple mod­els could be tested in this way in a mat­ter of minutes, more com­plex ones in hours.

This is fant­ast­ic by itself, but more inter­est­ing for me was how an intrins­ic fea­ture of bio­logy, and espe­cially wet-lab bench work, was dis­rup­ted by this com­pu­ta­tion­al accom­plish­ment. Some­thing that is so inter­leaved in the work of exper­i­ment­al bio­logy, time, and more spe­cific­ally bio­lo­gic­al time, ceased here to be present, at least in any recog­nis­able way. Instead, a com­pu­ta­tion­al time comes to count in which the meas­ures are pro­duced through the steps taken in a sequence of lem­mas (roughly trans­lated as con­di­tion­al argu­ments’ in logic).

So we begin to see here how count­ing and being coun­ted can entangle. Some highly spe­cial­ized and com­pu­ta­tion­ally soph­ist­ic­ated tech­niques for trans­lat­ing bio­lo­gic­al states into clusters of counts means that life, cell life, comes to count dif­fer­ently. The cel­lu­lar mod­els have no way of enu­mer­at­ing the changes occur­ring in a tem­por­al sequence. Through a dif­fer­ent fig­ur­ing, the cel­lu­lar life being enacted by the mod­els is done through a sort of state space where it is the dens­ity and weight giv­en to the rela­tions that make a dif­fer­ence. So the tech­niques of enu­mer­a­tion and cal­cu­la­tion fun­da­ment­ally alter what mat­ters in the cel­lu­lar sys­tem.

To put it anoth­er way, the counts, bound up with a form­al and algorithmic logic, are a mat­ter of life and (I need to be care­ful here) death: for this tool is an exper­i­ment­al one tar­geted at mod­el­ling, for instance, healthy skin cell devel­op­ment and lean­ing more about those cases in which can­cer­ous rather than healthy cells pro­lif­er­ate.

I want to say here, then, that the modes of count­ing and how things come to count appear tightly entangled. I’ve missed too many of the import­ant details here, but hope­fully ever so faintly we catch a glimpse how count­ing can become a way to see and do lived worlds dif­fer­ently.

To turn to my second example, I’d like now to think through the data flows of London’s rent­al bikes and how I’ve used my own count­ing meth­ods to intro­duce, let us say, some trouble into the entan­gle­ments.

I see there to be two broad ways in which the Bor­is bike’ data (made freely avail­able by the pub­lic author­ity, Trans­port for Lon­don) are being used. One is tar­geted at sup­port­ing the users of the sys­tem, provid­ing them with, for example, live counts of bike avail­ab­il­ity for the roughly 700 dock­ing sta­tions across the city. You can down­load apps, for instance, that show the nearest dock­ing sta­tions and the num­ber of bikes avail­able to rent.

The second com­mon use of the data is to visu­al­ise the usage, pic­tur­ing the pop­ular­ity of dock­ing sta­tions and some indic­a­tion of the fre­quency of jour­neys between them. The res­ult is often a col­our­ful map of nodes (dock­ing sta­tions) and lines of vary­ing dens­ity between them (indic­at­ing jour­ney fre­quency).

The first thing I want to say about these geo­spa­tial counts of bicycles will be of little sur­prise to us. These bikes and their data are bound intim­ately to a polit­ics of the city. Yes, the Bor­is bikes were launched in 2010 by the con­tro­ver­sial con­ser­vat­ive may­or of Lon­don, Bor­is John­son (hence their col­lo­qui­al name), and yes, the system’s status as a public-private part­ner­ship is often used as an exem­plary case for par­tis­an­ship on both sides of the public/private own­er­ship debates.

Things go deep­er than this though. Inter­weaved with the spa­tial con­fig­ur­a­tion of the city and a spe­cif­ic set of eco­nom­ic, tech­nic­al and com­pu­ta­tion­al modes, we find a geo­graphy emer­ging from the entan­gle­ments. Most obvi­ously this is mani­fest in the free 30 minute win­dow users have before they are charged on a per-minute basis for using the sys­tem. With about 95% of all jour­neys fall­ing inside this count down, a car­to­gram of the city is pro­duced that has some fairly well-defined regions and bound­ar­ies. These, more often than not, paint a pic­ture of a patch­worked city with hubs in the fin­an­cial dis­tricts and dense spokes fun­nelled to the res­id­en­tial neigh­bour­hoods that ser­vice them. Large areas to the East and South East are rendered invis­ible in these cycle-slash-data routes. So the net­work of nodes and con­nec­tions, prob­ably unsur­pris­ingly, cor­res­pond to where wealth and prosper­ity are accu­mu­lat­ing in the city.

At risk of over­sim­pli­fy­ing things, what I want to say here then is that the mul­tiple sys­tems of count­ing and the mater­i­al infra­struc­tures through which the counts are pro­duced do polit­ic­al work, but, and crit­ic­al to my point here, is they do a work that merely reminds us of what we all already know; to bor­row Donna Haraway’s , Noth­ing”, not even num­bers, come without their worlds”, and these worlds like the ones etched out of the Bor­is bike’s data maps recapit­u­late the kinds of dif­fer­ences we know too well.

Draw­ing heav­ily on Kat’s ever-so art­ful ways of treat­ing the empir­ic­al site — of treat­ing it dare I say with the dis­tain it deserves — my urge here has been to inter­vene, to find new entan­gle­ments that might pro­voke oth­er ways in which dif­fer­ence might be done, that might if you will trouble the trans­mis­sions.

So on one fine autumn day last Octo­ber I took my first ride on a Bor­is Bike, on bike num­ber 2175.

My jour­ney is between two dock­ing sta­tions that lie at the East­ern edge of the cycle scheme’s car­to­graphy of routes and stops. The route, start­ing at a dock­ing sta­tion on Aber­feldy Street leads me fur­ther East (about 5km bey­ond the rent­al bike scheme’s east­ern most dock­ing sta­tion), through a series of neigh­bour­hoods that, des­pite their prox­im­ity to the fin­ic­al dis­trict, Canary Wharf, still feel a long way from London’s ever increas­ing prosper­ity and cycles of gentri­fic­a­tion.

After rid­ing North along the pop­u­lar mar­ket street, Green Street, in Newham I come back on myself, head­ing due West along the Newham borough’s Gre­en­way, an embank­ment of green­ery and con­crete over­lay­ing the 150 year old North­ern Out­fall Sew­er, part of London’s net­work of Vic­tori­an sewage sys­tems.

In total, my jour­ney takes 45 minutes, start­ing at 16:45 and end­ing at 17:30. The aver­age jour­ney time for the 74 rides that began at the same time, across the scheme, was 15 minutes. In the week pre­ced­ing my jour­ney 18 jour­neys began from Aber­feldy Street against a sev­en day total of 139,793 across the entire scheme.

My jour­ney is then an inten­tion­al move to the edges of London’s bike rent­al dock­ing sta­tions and the asso­ci­ated data trails of bike flows. Start­ing with the modes of count­ing that have suc­cess­fully reminded us of what we already know, I’ve sought out some­thing else.

And to mess around with these counts fur­ther, my body is also instru­mented with a range of off-the-shelf bio­sensors or self-monitoring sys­tems, each pur­port to cap­ture in some shape or form indi­vidu­al physiolo­gic­al or bod­ily phe­nom­ena, steps, heart rates, glob­al pos­i­tion, a sequen­tial visu­al memory.

Again, my aim here is to infuse some­thing dif­fer­ent into the mix­ture of seem­ingly famil­i­ar counts. Intro­du­cing pecu­li­ar jux­ta­pos­i­tions and instabil­it­ies between counts, it is an attempt to sur­face oth­er kinds of flows and con­nec­tions that might just etch new topo­graph­ies into the city. What I really want to do here is alter how we see life in the city, to trans­mog­ri­fy what counts, in answer to Nigel Thrifts evoc­at­ive call:
We need spaces that graft… We need spaces that don’t line up. We need spaces that breathe dif­fer­ent atmo­spheres. We need new slopes, strips, roads, tracks, ridges, plains, seas... We need room. This is meant as an effort to make room.”

Here, I want to leave as ill defined any ideas for how things could come to count.

What I want to say though is that, what I’m strug­gling with is a sense of count­ing as an appar­at­us of trans­mis­sion for how we might open up the pos­sib­il­it­ies for new rela­tions. From my own exper­i­ment, the fluxes of rates, coordin­ates, steps’ , image sequences, and so on are open ques­tions about how we might sur­face a mix­ture of worlds, ones in which the counts spir­al off the map lit­er­ally and fig­ur­at­ively, ones where we are not sure what might come to count.

Count­ing and being coun­ted here then col­lapse:

Count­ing, becomes a way to inter­vene in the num­bers and to fur­ther entangle the panoply of eco­nom­ic, tech­nic­al, com­pu­ta­tion­al, polit­ic­al, and eth­ic­al modes that make worlds. Count­ing is to shift what it is that counts, and to ask how life wheth­er that be amongst cells or for those of us in liv­ing togeth­er, could be dif­fer­ent.

1. See this paper for a longer account of the work on this mod­el­ling tool: Alex S Taylor, Jas­min Fish­er, Byron Cook, Sam­in Ish­tiaq, Nir Piter­man (2014) Mod­el­ling Bio­logy – work­ing through (in-)stabilities and fric­tions, Com­pu­ta­tion­al Cul­ture 1(3).
See this paper for a longer account of the work on this mod­el­ling tool: Alex S Taylor, Jas­min Fish­er, Byron Cook, Sam­in Ish­tiaq, Nir Piter­man (2014) Mod­el­ling Bio­logy – work­ing through (in-)stabilities and fric­tions, Com­pu­ta­tion­al Cul­ture 1(3).
See Har­away, D., (1997), Modest_Witness@Second_Millennium. FemaleMan©_Meets_OncoMouseTM:
Fem­in­ism and Tech­nos­cience, New York: Rout­ledge. And also see la Bel­lacasa, de, M. P. (2012) ‘“Noth­ing comes without its world”: think­ing with care’, The Soci­olo­gic­al Review, 60(2), 197 – 216.

Leave a comment