I’m very excit­ed to come here and speak here. I’ve been try­ing to fol­low the work that is com­ing out of here, and am very excit­ed about see­ing very fresh ideas that are rel­e­vant to the work that we’re doing. I’m teach­ing at Shenkar College in Ramat-Gan, which is just out­side Tel Aviv, and vol­un­teer­ing with an NGO called The Public Knowledge Workshop that is doing work on civic engage­ment and gov­ern­ment trans­paren­cy in Israel. I’m slight­ly depressed, but hope­ful­ly it won’t show too much. (Politically only.) Anyway.

What I’m going to talk about today is a theme I’ve been work­ing with for the last decade or so, and devel­oped through my years in New York. I spent five years in New York between 2005 and 2010, and this is kind of a theme that as a design­er I’ve been work­ing with for many years now.

I’m going to dis­cuss the theme of inter­face and maybe the pol­i­tics of inter­face through the com­mu­ni­ca­tion cycle, the idea of pro­to­cols, and then I’ll try to sug­gest inter­faces for resis­tance, and ways for that to be addressed as well.

When we talk about life online we have all of the euphemism, it’s dis­trib­uted, it’s open, it’s inter­ac­tive, par­tic­i­pa­to­ry, demo­c­ra­t­ic, social, eman­ci­pa­to­ry. At the same time, when we’re talk­ing about online life, we’re talk­ing about destruc­t­ing, con­trol­ling, intru­sive, abu­sive, repres­sive, shal­low. All of these terms, all of these hopes, all of these con­cerns when it comes to online life, we meet them through the inter­face. So the inter­face is this point of inter­face between us and all of this tech­nol­o­gy. It’s at the heart of the debate. So what is interface?

There are many def­i­n­i­tions. The one that I find inter­est­ing is ​“a com­mon bound­ary or inter­con­nec­tion between sys­tems, equip­ments, con­cepts, and human beings.” This is one of many def­i­n­i­tions, but I think one part of it that I think makes sense and com­mu­ni­cates specif­i­cal­ly is this idea of com­mon bound­ary and inter­con­nec­tion. This idea of the inter­face being some­thing com­mon, some­thing that we share, some­thing that we con­nect through implies some kind of rela­tion­ship between these sys­tems, equip­ments, con­cepts, human beings. And it does­n’t imply nec­es­sar­i­ly a lev­el of con­trol, it does­n’t imply that one side should be stronger than the oth­er or so on. This idea of com­mon­al­i­ty is some­thing that has been dis­cussed through media stud­ies, and before for many years, this idea of mutu­al­i­ty and interconnectedness. 

In the com­mu­ni­ca­tion cycle, we have—it’s like in Sesame Street. I have an idea, I talk to you, you have the idea. You have an idea, you talk to me, we both have [the] idea. Now the two of us have two ideas, how great. In 1980 Stuart Hall wrote a paper called ​“Encoding, Decoding” and at the heart of his paper was this cri­tique of this com­mu­ni­ca­tion cycle. He was say­ing that this is actu­al­ly not the way it works. He said this is basi­cal­ly too sim­plis­tic. What is hap­pen­ing there is a bit more com­plex. He called this ​“tex­tu­al deter­min­ism,” this idea that one side can plant ideas, kind of dragged and dropped into the mind of anoth­er per­son, that’s not how com­mu­ni­ca­tion actu­al­ly works. He tried to focus on the process of how com­mu­ni­ca­tion hap­pens, and he said first of all there’s encod­ing. I have an idea, I encode it into a dif­fer­ent form (in the case of this illus­tra­tion into speech) so the idea is turned into sounds. So this body of knowl­edge is encod­ed into sound, and the sound is def­i­nite­ly not the body of knowl­edge. And as long as we have mean­ing­ful dis­course, as long as you can hear me and under­stand my fun­ny accent and so on, if you’re in the back and you can’t hear me well, then we don’t have mean­ing­ful dis­course and there’s room in the front. But the point is that this is assum­ing you can actu­al­ly hear me, and under­stand me to an accept­able level.

Then there’s the process of decod­ing, decod­ing sound through lis­ten­ing in this case, into a new body of knowl­edge. So if in Sesame Street it was like I have a tri­an­gle and now you have a tri­an­gle, so I have a tri­an­gle and what you have is not a tri­an­gle any­more. It’s dif­fer­ent. I don’t know what it is, you don’t real­ly know what I had in my brain. This process, both the encod­ing process and the decod­ing process are cre­ative process­es, process­es where the infor­ma­tion changes.

I was real­ly inter­est­ed in that Stuart Hall was writ­ing about tele­vi­sion, specif­i­cal­ly, in 1980. That was the medi­um that was being researched, and there were a lot of con­cerns about what does tele­vi­sion do. He referred to three types of code. He was say­ing one way of decod­ing is through a dom­i­nant or hege­mon­ic code. The recip­i­ent decodes the mes­sage using the same code it was encod­ed in, as is basi­cal­ly. Think a weath­er or sports broad­cast, or a cer­tain type of radio host, that what­ev­er they say is being accept­ed as is. So when I’m lis­ten­ing to the sports fore­cast, for exam­ple, I don’t ask myself, ​“Well they said it was a touch­down, but is it real­ly a touch­down?” It’s not real­ly rel­e­vant. In some cas­es it is rel­e­vant to ques­tion the mes­sage, but the hege­mon­ic code would be kind of accept­ing every­thing that is there.

The nego­ti­at­ed code is more crit­i­cal yet not com­plete­ly dis­mis­sive of the read­ing. Think NPR. NPR claims it does­n’t have a lib­er­al bias, you know it has a lib­er­al bias, but you see your­self as intel­li­gent enough to spot it. But it’s still use­ful, it’s still use­ful to lis­ten to it. You take what you can get.

Then there’s the oppo­si­tion­al code. In the oppo­si­tion­al code, the mes­sage becomes an oppor­tu­ni­ty to decon­struct the code. Think a fem­i­nist read­ing of ​“Little Red Riding Hood” or The Sex Pistols’ ​“God Save the Queen” or the ​“Hitler Gets Angry” meme. We know that what the orig­i­nal mes­sage was sup­posed to be, but we refer to orig­i­nal encod­ing in a com­plete­ly dif­fer­ent way. Most memes actu­al­ly are some­what oppositional. 

For me, I’m real­ly inter­est­ed in what hap­pens when it’s not con­ver­sa­tion or tele­vi­sion. What hap­pens when com­mu­ni­ca­tion is struc­tured through inter­face, and specif­i­cal­ly web inter­faces. When the mes­sage is encod­ed, when we see a mes­sage on the web, it’s encod­ed through inter­face. We all cel­e­brate and enjoy how expres­sive the Internet is, and how expres­sive the tools of the Internet are, but this expres­sion is at the lev­el of cre­at­ing these web pages. When it comes to the decod­ing of these mes­sages, it’s basi­cal­ly the same idea. You get the mes­sage, you decode it, you under­stand what you under­stand from it. So far, pret­ty much the same model. 

The thing is what hap­pens when we’re try­ing to encode our mes­sage to the ​“Internet” or to a web page. When we respond to the sys­tem or the web site, it is always through the sys­tem’s dom­i­nant inter­face. If inter­face is an impor­tant part of how we com­mu­ni­cate online, we don’t actu­al­ly get to choose what inter­face to use. We basi­cal­ly work with­ing the con­struct of the inter­face that we were giv­en. I’ll give you an exam­ple to make it a bit sim­pler to understand.

This is from a friend of mine Leila Haddad, she’s a Palestinian blog­ger and jour­nal­ist. She tried to order flight tick­ets on British Airways, and you’ve all prob­a­bly come across this drop­down, ​“What is your nation­al­i­ty?” She’s look­ing down the list to find her nation­al­i­ty, and under ​“P” there’s no Palestinian nation­al­i­ty. So the inter­face demands this lev­el of obe­di­ence. This is who you can be. But the inter­face also teach­es obe­di­ence. If you come into a prob­lem like that, that’s where you call the com­pa­ny sup­port, right? So she called the sup­port of British Airways, and the per­son on the oth­er side was say­ing, ​“Okay I’m look­ing in my sys­tem.” and in that per­son­’s sys­tem there was anoth­er drop­down menu with no Palestinian iden­ti­ty or nation­al­i­ty or pass­ports or doc­u­ments or what­ev­er. So even when you talk to a human, that con­ver­sa­tion is mit­i­gat­ed through the inter­face. The same dom­i­nant inter­face was there as well.

Lisa Nakamura, who’s research­ing race online, she calls it ​“menu-driven iden­ti­ties,” Interfaces that lim­it and define what we are allowed to be. Think how many years it took for Facebook to accept dif­fer­ent types of gen­der def­i­n­i­tions. It’s the same thing. Why is that a drop­down menu? Why can I only be one of these things?

Let’s talk about pro­to­col. To under­stand the role of user inter­face we need to dive deep­er into the com­mu­ni­ca­tion pro­to­col. This work is inspired by the work of Alex Galloway and oth­er thinkers on this top­ic. Specifically I’m try­ing to think what is dif­fer­ent about the web in that sense. When we’re talk­ing about tra­di­tion­al media, we’re talk­ing about cen­tral­ized pro­to­col, broad­cast, one-to-many. That’s a con­sis­tent pro­to­col. Television is always one-to-many. Same with news­pa­pers and broad­cast radio. The fact that it’s con­sis­tent means that the ques­tion of the inter­face in the tele­vi­sion case is not that big of an issue. When you under­stand the inter­face of the tele­vi­sion, when you under­stand it’s con­sis­tent, you can move on. Same with an inter­ac­tive pro­to­col like tele­phone. This is one-to-one, it’s inter­ac­tive, but it’s still a con­sis­tent pro­to­col. It’s a pro­to­col that does­n’t change every time I pick up the phone.

When we go into dig­i­tal media, we have dif­fer­ent kinds of pro­to­cols. We have email, we have chat, we have net­worked games, we have voice over IP, and all of them actu­al­ly have a con­sis­tent pro­to­col. When I’m send­ing an email, I always need an address, it should be @-something and so on. I’m not get­ting every email try­ing to ask myself what am I sup­posed to do now? When talk­ing about pro­to­cols, there’s the idea of the sev­en lay­ers of the OSI mod­el. The OSI mod­el is kind of tech­ni­cal, but the things we should care about—this is the pro­to­col of the Internet, the dif­fer­ent lay­ers of the pro­to­col of the Internet. We have the Internet Protocol lay­er, that’s where TCP/IP lives, that’s basi­cal­ly how pack­et switch­ing hap­pens. Basically every­thing that we know about the Internet being net­worked and so on hap­pens there. 

But the top lay­er of the OSI mod­el is the appli­ca­tion lay­er. That’s where we have our dif­fer­ent appli­ca­tions like email clients, voice over IP, net­work games, and so on, and also a brows­er. When we open the brows­er, I think some­thing slight­ly dif­fer­ent hap­pens, because the browser’s pro­to­col is flex­i­ble. On the web, the pro­to­col is basi­cal­ly a ques­tion of inter­face. Every web page basi­cal­ly adds anoth­er lay­er of pro­to­col, anoth­er lay­er of pos­si­ble con­trol, and I would sug­gest that the OSI mod­el needs to be updat­ed with anoth­er lay­er when we’re talk­ing about the web, a lay­er of inter­face. Specifically user inter­face. If in the appli­ca­tion lay­er we are using a brows­er, then we can get dif­fer­ent kinds of appli­ca­tions that real­ly change the way we should think about con­trol and com­mu­ni­ca­tion online. We have to take into account the inter­face ques­tion is a gov­er­nance ques­tion. It’s a ques­tion of who sets the rules, who’s sup­posed to be abid­ing by these rules, what is the process of mak­ing rules?

You’ve prob­a­bly noticed that this is kind of a media studies-inspired thing, and I tried but I could­n’t resist includ­ing Marshall McLuhan.

We are in the mid­dle of a tremen­dous clash between the old and the new. The medi­um does things to peo­ple and they’re always com­plete­ly unaware of this. They don’t real­ly notice the new medi­um that is wrap­ping them up. They think of the old medi­um because the old medi­um is always the con­tent of the new medi­um, as movies tend to be the con­tent of TV, and as books used to be the con­tent, nov­els used to be the con­tent of movies. And so every time a new medi­um arrives, the new medi­um is the con­tent and it is high­ly observ­able, high­ly notice­able. But the real rough­ing up and mas­sag­ing is done by the new medi­um, and it is ignored.
Marshal McLuhan, tele­vi­sion interview

That’s his clas­sic spiel, the old medi­um is the con­tent of the new medi­um, so the books have become the con­tent of film, film has become the con­tent of TV, and then TV has become the con­tent of the web, right? Well, not only. There’s a lot of things going on on the web, not just the old medi­um. In a sense we can say this is what McLuhan meant by ​“the medi­um is the mes­sage.” But that’s only true as long as the medi­um has a con­sis­tent pro­to­col, because I would say the medi­um is the mes­sage when we’re talk­ing about the kind of media that we have today. We can think about it as inter­face deter­min­ism. It kind of deter­mines that by point­ing at a medi­um, we are also point­ing at the inter­face. Maybe the inter­face is the mes­sage. Maybe that’s what McLuhan actu­al­ly meant, he just nev­er thought of media hav­ing dif­fer­ent inter­faces, which is what we have right now. Maybe the mes­sage and the rules that gov­ern it become ambigu­ous when the inter­face changes every time. Adopting from McLuhan can only go so far because I don’t real­ly share his techno-determinism, but this is kind of a response to him.

I’d like to bring up anoth­er impor­tant and influ­en­tial thinker, Donald Rumsfeld. Trying to make the case for the war in Iraq and refer­ring to the ques­tion of WMDs, Rumsfeld went on an inter­est­ing philo­soph­i­cal state­ment that you prob­a­bly remem­ber, ​“there are known knowns, there are things we know that we known, there are known unknowns, there are things that we know that we don’t know, but there are also unknown unknowns.” There are things that we don’t know that we don’t know, there are things that are so far from what we can even imag­ine, and that’s a good rea­son to go to war with­out proof of WMDs.

When Slavoj Žižek came across this, it was too tasty for him to ignore and he enjoyed the rhetoric, but he said basi­cal­ly this is great, but Rumsfeld for­got the fourth option: there are unknown knowns. What about the things that we don’t know that we know. This is the real rea­son for the war. The sub­con­scious, that’s where the ide­ol­o­gy hap­pens. And Žižek actu­al­ly writes about the fact that that’s where design hap­pens, that’s how design works. Design uses the unknown known, the things that we don’t know that we know, things that work on us with­out us knowing. 

I would add that’s how inter­face works. When it comes to user inter­face design and usabil­i­ty, I think Steve Krug would agree, but the dis­course in inter­ac­tion design is try­ing to keep the known, unknown. So the fact that we don’t know that we know is kind of good for inter­ac­tion design, at least in the pro­fes­sion­al debate. Steve Krug wrote one of the bibles of user inter­face, and this its title Don’t Make Me Think. I’m sure some of you have come across it. It was pub­lished in 2000 and wide­ly quot­ed and vast­ly trans­lat­ed. It has framed a lot of the dis­course about the role of inter­ac­tion design as low­er­ing the cog­ni­tive load to a min­i­mum. And let’s put things in per­spec­tive: that is true, to a cer­tain degree. One of the terms that we speak about in inter­ac­tion design is affor­dance. Affordance is how inter­faces or how tools com­mu­ni­cate what you are expect­ed to do with them.

But then the ques­tion that should be asked is what atten­tion do we afford or, in the case of the web, how is our atten­tion being afford­ed? Because when atten­tion becomes the most rare resource in the Internet, the atten­tion is becom­ing some­thing that we both use, but our atten­tion is being used. It’s being afford­ed, it’s becom­ing a cur­ren­cy on the web. So we’ve estab­lished that there are quite a few polit­i­cal issues involved, and when we’re com­ing to a site, we’re not too much in a posi­tion of affect­ing them. So how can we try to think of inter­faces for resis­tance, or resist­ing the par­a­digm of the inter­face? I would sug­gest a divi­sion of tac­ti­cal resis­tance, strate­gic resis­tance, and logis­tic resistance. 

Around the so-called ​“Web 2.0” around ten years ago, there was a lot of excite­ment about user-generated con­tent. They’re ask­ing us what we think, how amaz­ing is that? It’s like author­ship, all of a sud­den, can hap­pen because all of a sud­den the con­tent of the web is not just con­tent, it’s also inter­faces that would allow us to cre­ate con­tent. That’s when the inter­face is kind enough to allow us to express our­selves. I would call that the first depth of authorship. 

The sec­ond depth would be user-generated con­text. Think about mashups, embeds. This is built into the web pro­to­col. Think of tak­ing a Craigslist post­ings and putting them on a Google Map. That was a big deal when it hap­pened for the first time. Now cre­at­ing the con­text of tak­ing this and that and putting them togeth­er is a new type of author­ship. This kind of author­ship is built into the web’s pro­to­col, the idea that I can take one resource and anoth­er resource and put them togeth­er, that’s how the web works. It’s the same flex­i­bil­i­ty that also gives us track­ers and ad net­works. So it’s not nec­es­sar­i­ly used on our behalf.

The third option I would sug­gest would be user-generated inter­faces. This is also built into the pro­to­col of the Internet. Having users change what the site allows them to do, usu­al­ly by mod­i­fy­ing the brows­er, main­ly with brows­er exten­sions. The idea that you can take your brows­er and you can change it and rede­fine what the brows­er can do is some­thing that can extend what we might want to call inter­face lit­er­a­cy, this idea that it’s not only about how do I read the inter­face in front of me, or how do I low­er cog­ni­tive load, but how do I mit­i­gate this polit­i­cal ques­tion of gov­er­nance that hap­pens on the web.

One way of address­ing that is through tac­ti­cal resis­tance. The tac­ti­cal resis­tance would hap­pen on the inter­face lay­er. This is a sub­ver­sion of the inter­face’s use. I think that a text­book exam­ple for that would be Google bomb­ing. This was around the time of the attack at the begin­ning of the war in Iraq. This was a par­o­dy of not find­ing weapons of mass destruc­tion. When you searched the web for weapons of mass destruc­tion, you would get that as the first result on Google. That hap­pened by a con­cert­ed effort from dif­fer­ent blog­gers that opposed the war. They all wrote the words ​“weapons of mass destruc­tion” and linked them to this page, which raised the Google rank for this page and made it the first result for ​“weapons of mass destruc­tion,” so turn­ing PageRank into an inter­face. PageRank is not sup­posed to be an inter­face, or a user inter­face. What this kind of tac­ti­cal resis­tance rep­re­sents is a hit and run. It’s lim­it­ed in scope, for bet­ter or worse. It’s a lim­it­ed invest­ment, for bet­ter or worse. A lot of hack­ing and culture-jamming hap­pens on this lev­el of resis­tance. It’s about learn­ing how the sys­tem works and find­ing its vul­ner­a­bil­i­ties, tak­ing advan­tage of the inter­face’s open­ness. So the inter­face is open to a cer­tain degree; this is an oppor­tu­ni­ty. Kind of teach­ing the sys­tem some­thing it did­n’t know about itself. Most of these cas­es are sym­bol­ic ges­tures of resis­tance, so protest action and polit­i­cal art is main­ly tac­ti­cal in that sense.

The sec­ond type of resis­tance would be strate­gic resis­tance, and I would sug­gest that that’s the kind of resis­tance that hap­pens in the appli­ca­tion lay­er. That is basi­cal­ly say­ing yes we’re sup­posed to be brows­ing the web through these soft­wares, let’s try to see how can we change that, how do we change the brows­er itself? I delib­er­ate­ly chose an exam­ple that is not polit­i­cal but com­mer­cial. This is Book Burro. It’s not active any­more but there are quite a few sim­i­lar ones. What Book Burro does is some­thing that Amazon would nev­er do: tell you on each book page where you can find the book a bet­ter price online, and where you can find it in pub­lic libraries based on how far they are from where you are right now. We look at that and say, ​“Oh. Yeah, I need that.” None of us would even think in the direc­tion of cre­at­ing an inter­face like that, because we’re com­plete­ly sub­merged with­in the pol­i­tics of inter­face that we are not a part of— that we are not sup­posed to have agency there.

Does any­body know what the most pop­u­lar exten­sion both on Firefox and Chrome? Adblock Plus. As soon as exten­sions became a thing on Firefox, this hap­pened. Because we don’t nec­es­sar­i­ly want to see ads, and in some cas­es we don’t nec­es­sar­i­ly want to be tracked by ads. This is arguably one of the rea­sons for the Google Chrome project. Adblock Plus was one of the biggest hur­dles for Google AdWords from extend­ing and Google had to be in a posi­tion of lever­age in front of Adblock Plus. It was obvi­ous that with all of the euphemism around Google being a great sup­port­er of open source, Chrome could not have start­ed with no exten­sions, and could not ignore Adblock Plus and say, ​“This is a brows­er with no Adblock Plus.” But what they did, because they’re smart, they cre­at­ed lever­age of Adblock Plus’ destruc­tion. Google pays Adblock Plus. How many of you know that? Most of you don’t know that. Google pays Adblock Plus to change one thing about its inter­face. Adblock Plus added a check­box that says ​“Allow inob­tru­sive ads.” [whis­pered:] Google ads. 

So, redefin­ing the ques­tion of why do we not want ads. Do we not want ads because we don’t want ban­ners, or do we not want to be tracked? It’s com­plete­ly rede­fined through Adblock Plus now, and this is one of the rea­sons there are a lot of forks to the Adblock Plus project that don’t have this check­box. This is also one of the things that makes the claim that Adblock Plus is black­mail­ing ad net­works true.

Audience: How much mon­ey? Do we know?

Mushon: Undisclosed.

I can talk about it lat­er, but one of the exam­ples I can bring up is my project togeth­er with Helen Nissenbaum and Daniel Howe, AdNauseam. We basi­cal­ly cre­at­ed an exten­sion that works with your ad block­er. Every ad blocked by your block­er is clicked by AdNauseam. We click all of the ads on the web, basi­cal­ly flood­ing the pro­files that the ad net­works are con­struct­ing and plant­i­ng a lot of mis­trust between adver­tis­ers and ad networks.

This approach is more of a ​“hit and stay” approach. It has larg­er scope, for bet­ter or worse, larg­er invest­ment, for bet­ter or worse. It’s about devel­op­ing alter­na­tive ser­vices. It’s a deep­er and less tra­di­tion­al read­ing of the pro­to­col. The polit­i­cal val­ue here goes beyond the sym­bol­ic, it’s more than a ges­ture of resis­tance. It’s a sol­id resis­tance to the inter­face’s demand­ed obe­di­ence, a con­crete polit­i­cal pow­er to be reck­oned with. (And it’s hard­er to make an art career out of that. Just FYI.)

And the third lay­er is logis­tic resis­tance. Now we’re talk­ing about resist­ing the Internet Protocol, chang­ing the deep tech­ni­cal pro­to­col like TCP/IP either through slow bureau­crat­ic process­es or through rev­o­lu­tion, that’s how big pro­to­cols change. If you get rid of a hege­mon­ic pro­to­col like TCP/IP, you basi­cal­ly need to cre­ate a new hege­mo­ny to replace it. It’s a big polit­i­cal and pow­er­ful and often con­ser­v­a­tive process through stan­dards com­mit­tees, and when one stan­dard is replaced with a new one, it breaks every­thing above it. So this is some­thing that is kind of tough, and the exam­ple for attempts to chal­lenge the TCP/IP pro­to­col is this: I haven’t seen any. We basi­cal­ly haven’t seen seri­ous attempts to say we can go beyond TCP/IP. You are the crowd who can tell me, ​“No, you’re wrong. There’s this and this and that.” and I’m wait­ing to hear that after the talk. I’ve seen a lot of attempts for mesh net­works and oth­er things that are actu­al­ly built on top of TCP/IP, and none of the approach­es that basi­cal­ly say what­ev­er was invent­ed half a cen­tu­ry ago is not the way to go.

Evgeny Morozov, who can be seen as both an impor­tant and provoca­tive thinker, and an arro­gant and vio­lent troll (a bit of both, I guess), defines it as tech­no­log­i­cal defeatism, this belief that since a giv­en tech­nol­o­gy is here to stay there’s noth­ing we can do about it oth­er than get on with it and sim­ply adjust our norms. I think this cri­tique is actu­al­ly very valu­able. There are things that we don’t like about the way pro­to­cols are work­ing, but we don’t see enough seri­ous attempts to chal­lenge them. It’s like this con­cept that has been with us for fifty years, and it’s as if this is the way we’re sup­posed to work from now on.

All of this cri­tique of things that have to do with the Internet, it’s only online, it’s not real. Why do we need to wor­ry about that so much? But just try to think how many hours we spend in front of these screens. Most of our work days are in front of these screens, and then we spend more and more time in front of gov­ern­ing inter­faces, so most of our dai­ly expe­ri­ence in con­tem­plat­ing author­i­ty and new rule sys­tems is in front of web inter­faces. Because when we talk about the web demand­ing obe­di­ence, it’s ²⁴⁄₇. Every web page that I open has the sub­lim­i­nal ques­tion of ​“What am I allowed to do here?” And I don’t even ask myself—this is the unknown known. The fact that I’m con­stant­ly think­ing about it, con­stant­ly made not to think about it, that is the most pro­found expe­ri­ence of obe­di­ence that we have on a dai­ly basis, on a page by page basis. And not only online. Away from key­board, these dom­i­nant inter­faces of con­trol are not new. To put it in con­text, we can see some exam­ples for resis­tance to inter­faces of con­trol that are not tech­no­log­i­cal or not digital.

For tac­ti­cal resis­tance, we can look at This is a response to the hege­mon­ic code of the Iranian gov­ern­ment as part of the Iranian upris­ing in 2009. The bills were used as a form of protest, with writ­ings like ​“death to the dic­ta­tor” and ​“for free­dom.” It has become a real prob­lem. More and more bills are chang­ing hands with polit­i­cal state­ments that are against so-called ​“counter-revolutionary” or counter-counter-counter-revolutionary. It gets com­pli­cat­ed with Iran. Money’s not sup­posed to be an inter­face for protest, but they made it into an inter­face of protest and nobody’s going to throw that in the garbage, even though Ahmadinejad tried to say that every bill that had a sign like that would not be accept­ed, but the phe­nom­e­non was so wide that they could­n’t do that. 

We can think of strate­gic resis­tance. My exam­ple here is Rolling Jubilee, which is a very strong hit and stay approach. What they do is they’re basi­cal­ly bail­ing out the poor. The way debt works, is if there are high-risk debts they are trad­ed between dif­fer­ent banks and com­pa­nies, and they’re trad­ed for much less than the deb itself. What Rolling Jubilee does is they fundraise to buy debts and strike them. When this screen­shot was take, just a bit more than half a mil­lion dol­lars abol­ished almost twelve mil­lion dol­lars in debts. So this is read­ing the inter­face for the finan­cial sys­tem and say­ing, ​“We are going to play that game, but not in the way that is expect­ed.” And this is chang­ing things on the ground. These are peo­ple who are get­ting their lives back.

And the third lay­er, the logis­tic resis­tance is best rep­re­sent­ed by what’s been hap­pen­ing in the Middle East. Sometimes it looks like a very hope­ful attempt to say we’ll plant some­thing new in Tahrir Square and things will be dif­fer­ent. We’ll get rid of that pro­to­col and we have ideas for a new pro­to­col. Sometimes it looks a bit more grue­some, and we’ve all seen much much more grue­some pic­tures. My neigh­bors around the Middle East have had enough of the pro­to­cols that gov­ern them. Tunis, Egypt, Libya, Yemen, and oth­ers got rid of their pre­vi­ous pro­to­col, and what we’re see­ing now can be seen as a pro­to­col war, redefin­ing the rules of play.

To con­clude, even though the con­flict is unavoid­able, I find it very inter­est­ing and quite impor­tant to con­front. I address it in my aca­d­e­m­ic and cre­ative work through dif­fer­ent tac­tics and strate­gies. Specifically I find the ques­tion of inter­con­nect­ed­ness to be crit­i­cal, espe­cial­ly from my geopo­lit­i­cal perspective. 

Thank you.

Audience 1: You’ve paint­ed this almost tiered or lay­ered approach to think­ing about the tac­ti­cal, strate­gic and logis­tic forms of resis­tance to the obe­di­ence that you’ve laid out is demand­ed. I’m won­der­ing whether, and you’ve sort of made a prompt at least in the online sense or with the dig­i­tal tech­nolo­gies, to say that the logis­tic you don’t have good exam­ples of, or are look­ing for the good exam­ples of it. I want to push you to think about why that is. Is one of them eas­i­er to do? Is one of them eas­i­er to under­stand? Is it a tech­nol­o­gy affor­dance, or a sort of philo­soph­i­cal affordance?

Mushon: I think it’s both and beyond that. It’s the way the tech­no­log­i­cal world is con­struct­ed. The fact that we are led into these new polit­i­cal set­tings by com­pa­nies who have their own inter­ests and are the biggest play­ers is a big prob­lem. It basi­cal­ly puts us in a point of dis­ad­van­tage. I also think you’re lucky enough to be in the US, because the US con­trols the Internet, and from my per­spec­tive not being an American cit­i­zen, I’m not in a posi­tion of influ­ence as you guys are. We’re see­ing that with the whole debate about Snowden is they’re fol­low­ing American cit­i­zens. Is it okay to fol­low me? The answer is yes because I’m clos­er to ter­ror­ism, potentially. 

Audience: The Constitution only pro­tects the citizens.

Definitely not online. Definitely not the tech­no­log­i­cal gov­er­nance that we’ve built that is built from a cer­tain ide­ol­o­gy, to serve cer­tain ide­ol­o­gy. There’s so much pow­er involved. But the rea­son that I referred back to on the ground phys­i­cal activism or the finan­cial sys­tem is because my con­cern is that if we don’t resist the inter­faces online, then we don’t expect to resist any­thing, because that is where we learn how to inter­act, that’s where we learn what it expect­ed from us on a civic lev­el. And that is concerning.

I can say that in the case of the way tech­no­log­i­cal gov­er­nance works, we’ve come across it with AdNauseam.

Audience 2: This is more just a resource ques­tion if you have any thoughts. I’m inter­est­ed to know how the enforce­ment or the set­ting in of the demand for obe­di­ence tracks with things like decline in health, advance­ment of prob­lems with the envi­ron­ment, advance­ment of things like dia­betes and cer­tain oth­er kinds of ill­ness­es that are more… I’m look­ing at the epi­demi­ol­o­gy. I’m inter­est­ed to know if there’s any [resources?] that you know of to track on a very gran­u­lar, incre­men­tal lev­el, how the advance­ment of this kind of demand­ing, obe­di­ence tracks with oth­er things. Do you have any sense of [crosstalk]

Mushon: It’s not my field, so I would­n’t know how to research that, exact­ly. The state­ment I’m mak­ing is about polit­i­cal imag­i­na­tion, to a cer­tain degree that is lim­it­ed by our dai­ly expe­ri­ence of obe­di­ence. So I’m con­cerned about our abil­i­ty to imag­ine dif­fer­ent rela­tion­ships but I don’t know how to quan­ti­fy that.

Audience 2: Any sense of resources where I might be able to…

Mushon: Maybe some­one else in the room. I don’t know.

Audience 2: Okay. Thank you.

Audience 3: Maybe I missed this but you had this tan­ta­liz­ing lead at the end­ing about inter­con­nec­tion, and I kind of missed the con­nec­tion of that con­cept with—

Mushon: With interface.

Audience 3: Yeah.

Mushon: Basically the def­i­n­i­tion of inter­face has to do with this con­nec­tion between sys­tems, peo­ple. We’ve expe­ri­enced inter­face for so long now and so intense­ly through user inter­face that we for­got that. We for­got that there is no implied hier­ar­chy in the con­cept of inter­face. There could’ve been a web where the pow­er set­tings of site own­ers ver­sus site users would’ve not been set as hier­ar­chi­cal. So in its core, inter­face is about rela­tion­ships, about interconnectedness.

[Next ques­tion repeat­ed­ly drowned out by pho­to­copi­er; may con­tain some errors.]

Audience 4: I’d like to pull on this thread of what you say is a lack of sub­ver­sion of the IP pro­to­col. First, that’s set­ting a very high bar. Even the engi­neer­ing task force that wants to replace IPv4 and IPv6 has been strug­gling for a decade. Even the peo­ple who con­trol the mech­a­nisms of tech­no­log­i­cal deter­min­ism can’t change it when they want to. But I would argue that there’ve actu­al­ly have been a large num­ber of attempts suc­cess­ful to sub­vert it. I would argue that any con­tent fire­wall that does con­tent fil­ter­ing is sub­vert­ing TCP/IP because it’s not let­ting pack­ets that con­tain a cer­tain mes­sage go through. I would argue that dis­trib­uted denial of ser­vice attacks that use char­ac­ter­is­tics of the IP pro­to­col, or Anonymous’ Low Orbit Ion Cannon. [China and dis­trib­uted denial of ser­vice attacks] That’s sub­vert­ing the IP pro­to­cols. Things like DNS cache poi­son­ing that will take a web site our of domain name res­o­lu­tion, and peo­ple would talk about things like SOPA and PIPA, which are legal efforts to get sites tak­en out of the domain name sys­tem [are] a legal effort to sub­vert the way the IP pro­to­col is sup­posed to work.

Mushon: I under­stand your point. I just think, at least the way I’m see­ing it, this is using the TCP/IP pro­to­col. This is what the TCP/IP pro­to­col allows. That is maybe not the way we would’ve want­ed or cer­tain bod­ies would’ve want­ed to use it, but this is a use of that pro­to­col. This is a part of the prob­lem with a pro­to­col. I would say the fact that pri­va­cy is not embed­ded into the pro­to­col is anoth­er prob­lem with that pro­to­col, beyond the prob­lem of dis­trib­uted denial of ser­vice attacks or pack­et sniff­ing. This is the pro­to­col, this is what it allows. If this is done with­in that con­text, this is not chal­leng­ing that pro­to­col by say­ing here’s anoth­er pro­to­col. It’s maybe expos­ing the vul­ner­a­bil­i­ties of that pro­to­col, but not nec­es­sar­i­ly direct­ing anoth­er way for going beyond what the pro­to­col allows or does not allow.

Audience 4: [Mostly inaudi­ble] oppo­si­tion­al the same way […] fem­i­nist reading

Mushon: I would just put it in one of the lay­ers above the TCP/IP lay­er. But I think we under­stand each other.

Audience 5: The ques­tion I have has to do with your speak of inter­faces and affor­dance. I do a lot of US and I do a lot of dis­abil­i­ty pro­fes­sion­al­ly for patients and peo­ple in health­care. These are, in my expe­ri­ence, and this is the ques­tion for you, is you said these inter­faces ask us to obey, they push you in a cer­tain direc­tion. Well, just like in the DDoS case, I almost want­ed some kind of pro­to­col to save me from these prob­lems, but I did­n’t have any because I have ​“free­dom.” But now I find the inter­faces that are bad or that don’t allow you to real­ly do what you meant to do or accom­plish the task you meant to accom­plish, which is what cre­ates that rela­tion­ship. I feel like a lot of that dis­con­nect is slop­pi­ness and not an agen­da. So the ques­tion for you, I deal with the US gov­ern­ment right now, who is forc­ing me to make it very uni­ver­sal­ly usable. Great, right? It sounds fan­tas­tic, but it comes from the gov­ern­ment as a man­date, a set of rule ​“I have to meet” or some­how prove to have met. But it’s kind of oppo­site to my nat­ur­al incli­na­tion, being a cyber-child, to say no, ​“No, I don’t need you to tell me what to do. I will sur­pass your require­ments, I will do some­thing very dif­fer­ent.” So I just…a ques­tion of slop­py ver­sus mak­ing it bet­ter for every­body. Individual choice ver­sus a mandate.

Mushon: A ver­sion of this ques­tion was made to me yes­ter­day. I gave a talk at MassArt about dis­in­for­ma­tion visu­al­iza­tion, or how to lie with dataviz. My argu­ment there was that we should not look at visu­al­iza­tion like Edward Tufte claims we should look at it as beau­ti­ful evi­dence, we should look at it as beau­ti­ful arguments. 

I think it’s about the argu­ment. It’s not about ​“inter­faces are there to grab us all.” And I think the Don’t Make Me Think book is actu­al­ly good in the sense that there is a prob­lem with cog­ni­tive load. It’s a chal­lenge. Cognitive load is a chal­lenge for design, but when we are so obsessed about direct­ing atten­tion and mak­ing our users not thing, we also for­get that we are rep­re­sent­ing dif­fer­ent inter­ests. There are the inter­ests of the site own­ers and there are the inter­ests of the users and I think the exam­ple of Book Burro, that’s why I like it so much, because the dif­fer­ence of inter­ests there is so obvi­ous. When we are in the posi­tion of ​“design­ing” these user inter­faces, we are doing polit­i­cal work. We are doing work that should be poli­ticized. So I’m not going to politi­cize bad inter­faces. It’s not about mak­ing you think about things that are not polit­i­cal. I don’t think peo­ple should make crap­py user expe­ri­ences just so you think about the face there’s a crap­py UX design­er behind it. But I do think that if what’s embed­ded within a cer­tain inter­face is some­thing that should be poli­ticized, we should change the way we think about what we are allowed to do online. That’s more of the cri­tique that I’m making.

Audience 7: This is a follow-up to that. This seems par­tic­u­lar­ly impor­tant now ver­sus say, in 2002 or some­thing because of the con­sol­i­da­tion of atten­tion on the Internet. So it seems like this is par­tic­u­lar­ly rel­e­vant when you’re talk­ing about a site like ama​zon​.com. Politicizing ama​zon​.com is very dif­fer­ent than politi­ciz­ing my per­son­al blog or some­thing like that because so much of the atten­tion, so much of the per­son­al data that is cap­tured and extract­ed on the Internet flows to Amazon and Google and Facebook and Twitter and so on. So do you feel that that is part of this pol­i­tics, this cur­rent geog­ra­phy of atten­tion and data on the Internet?

Mushon: It makes a lot of dif­fer­ence but the play­ing with con­text on the Internet has changed dra­mat­i­cal­ly as well. Out of the 100 most pop­u­lar web sites on the Internet, 97 of them report back to Google on every­thing that you’re doing, and it’s not like 97 out of the 100 most pop­u­lar sites are Google sites. So even your blog is exposed to these ques­tions. And they don’t even expose a user inter­face, it’s just an implic­it inter­face that by brows­ing the web you’re report­ing back to Google, because peo­ple are using Google Analytics or YouTube or Google Maps or what­ev­er. So I think we can’t actu­al­ly say these are only prob­lems in the big sites.

Audience 7: I was just think­ing your argu­ment about inter­faces sort of demand­ing obe­di­ence would be in a sense less prob­lem­at­ic if they were— It seemed like you were say­ing they would be less prob­lem­at­ic if there were more of us pro­duc­ing those inter­faces. If the task of who’s own­ing the sites and who’s using the sites is more equal­ly dis­trib­uted amongst peo­ple. Is that not—

Mushon: I don’t think that if all of us build our WordPress blogs, every­thing would be fine. Which is hap­pen­ing less and less, by the way.

Audience 7: And also mobile mobile. There’s noth­ing user [search—?] [crosstalk]

Man: And then there’s Internet of Things, which brings inter­face to a whole new—

Mushon: Yeah, and if we con­tin­ue along the pre­vi­ous medi­um being the con­tent of the new, think about mobile apps and how the web is becom­ing the con­tent of them. I think it has to do with lit­er­a­cy. And I actu­al­ly think it has to do with polit­i­cal imag­i­na­tion. Can we imag­ine a dif­fer­ent rela­tion­ship? I don’t think we can do that because we haven’t thought of chal­leng­ing the TCP/IP pro­to­col. I think there’s a lot to do even before we chal­lenge TCP/IP. And I’m not try­ing to chal­lenge TCP/IP, I’m far from being in these lev­els of action or tech­no­log­i­cal lev­el. I think there’s a lot to do with both tac­tics and strate­gies that are inspir­ing polit­i­cal­ly as well, and tech­no­log­i­cal­ly, and UX-ly and so on.

Audience 5: An exam­ple that real­ly dri­ves me insane in terms of mod­ern devel­op­ment of design is, we’ve all known BBC Online. We’ve used it maybe, for news, resources, etc. Now any­body here has a mobile app. It is like back to TV land. I feel even less engaged, less pow­er­ful, less able to learn. For exam­ple, you can read the arti­cle but there’s nev­er relat­ed con­tent or links to the embassy of Nicaragua and the embassy of the US who are hav­ing an issue. There’s no com­ments or even a name of the per­son or peo­ple who wrote it. And I feel like here we are after the 90s and the first ten years of 2000 after we’ve seen this cul­ture explode, seen this con­tent being cre­at­ed by all us, see some peo­ple per­co­late up and being new types of jour­nal­ists, and now I feel like eas­i­ly reversed—

Mushon: I have an answer for you that might actu­al­ly address what you—the answer that I haven’t giv­en you. [Audience 7] 

I’m work­ing on a client project. This is pub­lic radio in Israel. When I was talk­ing with the com­pa­ny that built the pre­vi­ous site that we’re now redesign­ing, I said we will need an API as part of the new redesign. The same com­pa­ny was sup­posed to build the new site as well and said, ​“We don’t want to build an API for them. If we build them an API, they’ll be able to get rid of us.”

I was like, ​“What? I can’t believe you even allow your­selves to say that.” (The good news is that this com­pa­ny is not going to build it.) But there is a future that we can get to, that is not very far, where we all instead of pro­vid­ing web sites, we would pro­vide APIs. So your blog would be expos­ing an API. It actu­al­ly was like that, that was RSS. I think when RSS hap­pened, and since it kind of died, we did­n’t under­stand it yet because tech­nol­o­gy was­n’t there yet. Ajax was—the ​“X” was for XML; it came after. I think this is still valid, and I think it’s becom­ing valid again ten years or so after. Now there is this need for us to expose APIs, and to allow oth­er peo­ple to do oth­er things with dif­fer­ent inter­faces over dif­fer­ent data. I can see a trend towards that, and that would be a very inter­est­ing place both for data lit­er­a­cy and inter­face literacy.

Audience 9: To fol­low on that. When you talk a lot about the lit­er­a­cy, one approach to the— There are assump­tions or deci­sions or just slop­pi­ness in these inter­faces that restrict what I can do with it. Obviously the sort of per­son­al response is build toolk­its to allow peo­ple to build their own inter­faces, which you’ve done. Is that pri­mar­i­ly an argu­ment or response about lit­er­a­cy, or an argu­ment or response to say hey, I was able to make my own inter­faces so now I’m lib­er­at­ed from at least the top lay­er of restric­tions that were imposed on me. Is this sort of an argu­ment about util­i­ty or a meta-argument about ​“Oh, now I know how to build these things, so I’ve become empow­ered in gen­er­al to think about the assump­tions in what I’m allowed or not allowed to do…”

Mushon: I think the answer is bo— You know the answer is both, but I think the only way to get to the first assump­tion, the one about lit­er­a­cy, is through util­i­ty. Because this might be an inter­est­ing debate to have, but until it becomes an actu­al— That’s what I like about Book Burro. I need that. I use that. The fact that Book Burro not only rec­om­mend­ed oth­er places to buy but actu­al­ly sent me to my pub­lic library is super pow­er­ful. I was in a con­sumerist mode and it sent me to a civic mode. And there was val­ue, finan­cial val­ue and civic val­ue for me there. Same with Adblock Plus, in a dif­fer­ent way, obvi­ous­ly. Civic val­ue in Adblock plus, you would­n’t find, but it’s solv­ing a prob­lem. And while doing that it expos­es the dif­fer­ences of inter­est between me and the site itself

So I think there should be more ini­tia­tives that do think about ser­vices in that sense and are not only about lit­er­a­cy or polit­i­cal state­ments about infor­ma­tion, but also about what do peo­ple want.

Audience 9: And this is a dif­fer­ence between your AdNauseam exam­ple and Adblock Plus

Mushon: True.

Audience 9: Adblock Plus is pro­vid­ing util­i­ty for me, and has none of these sort of activist end­points where­as your AdNauseam is pro­vid­ing util­i­ty if it does the block­ing or pig­gy­backs on the block­ing, but then it’s mak­ing this aggres­sive stance to say, ​“Hey, I’m also going to try to dis­rupt the sys­tem that I’m check­ing out of rather than just check­ing out of it” I think a num­ber of the tools that you men­tion or that I can think of off of my head are more of the ​“I’m check­ing out of this” rather than the ​“I’m check­ing out and I’m try­ing to mess it up for every­one else.”

Mushon: That’s a prob­lem that I have with the pri­va­cy debate. It’s super depress­ing. We did­n’t go online to hide. We go online to com­mu­ni­cate. We go online express our­selves. We enjoy that. It’s fun, it’s social. The whole pri­va­cy debate, it gets peo­ple to either hope for the best or know that you’re being fucked but that’s just how it is. Or you’re tech-savvy enough to encrypt every lit­tle thing that you do and to use exact­ly the right set of tools to cir­cum­vent any attempt of data being col­lect­ed on you by some­one who you don’t know. There’s some­thing so mate­ri­al­is­tic about this, and it’s super depress­ing. It’s no fun. 

I’ve nev­er enjoyed pro­tect­ing my data. Until we’ve start­ed to go the com­plete oth­er way with AdNauseam. The whole idea is that we want to be expres­sive, let’s be super expres­sive. Let’s cel­e­brate it, let’s per­form it. If you want me to tell you that it’s a sol­id way of avoid­ing data col­lec­tion? No, it’s not. It’s not going to com­pete with the lat­est encryp­tion algo­rithm in the next data secu­ri­ty con­fer­ence. But it is try­ing to com­pete with this idea that secu­ri­ty is some­thing for extreme activists or ter­ror­ists or uber-geeks or what­ev­er. We need to change the lan­guage around that, and come from that point where the web peo­ple want to be, that expres­sive point that is play­ful. It’s not only pro­tect­ing myself, it’s actu­al­ly inflict­ing pain and get­ting back to the point of lever­age, which is what we’re lack­ing there. Because the back­ground of AdNauseam is the fail­ure of the Do Not Track stan­dards com­mit­tee. This is exact­ly the polit­i­cal process that did not give us a valu­able or valid solu­tion, while the solu­tion is there. 

You can basi­cal­ly already on Firefox you can already check a check­box, even in Internet Explorer. You can check a check­box that says you do not want web sites to build a pro­file on you. The only prob­lem is that they don’t respect that. That’s what the com­mit­tee was built do set. To set the rules of how this check­box is going to be respect­ed. But appar­ent­ly, three years after it start­ed, the com­pa­nies sit­ting on that com­mit­tee are there to make sure it nev­er leaves the ground.