Usability = Abusability?

Tom Coates talks about using an inverse-learning curve (or learning-cliff-with-an-overhang) to defeat spammers and other abusers of internet systems:

“Essentially what many systems online need – the systems that are prone to abuse that is – are graphs of ‘difficulty of use’ that are exponential – start almost flat and then escalate heavily afterwards.”

Cf. with notions of “seamful design” for beautiful executions of this concept? Any existing examples?

13 thoughts on “Usability = Abusability?”

  1. “Now there are a variety of reasons why (for example) you’d want to send lots (thousands or tens of thousands) of e-mails, but the main reason will probably be advertising or spam – kinds of e-mail which are generally perceived as an abuse. In fact I would hazard a guess that it becomes increasingly likely that someone is abusing the system the more e-mails they send.”

    This is why every one of these proposals that I’ve seen so far net out with a system where the very wealthy and powerful can afford mass communication, and the poor cannot.

    The ability to send alerts out to very large groups of people is the ability to have a functional democracy.

    EFF, Greenpeace, ACLU, CDR, NTK… There are thousands of good and worthy advocacy and information resources that send out tens of thousands of messages at a go, using the same tools spammers use.

    Popular speech never needs defending. When mass communication is given back to solely the rich and powerful, the Internet’s promise of samizdata, of Journalism 3.0, of real liberation, is eroded.

  2. Unfortunately, I can’t reply to you on your blog, Tom, so here’s my response:

    Whether we’re talkign about money or effort or difficulty is irrelevant. A computational tax (hash-cache) or an effort tax both translate into an advantage for those who can lay hands on additional resources (i.e., the rich) at the expense of those who can’t.

    There is no scarcity here. This is an amazing and unprecedented occassion in human history: plenty. A commons that is nonzerosum. We need new tools for managing such a commons, but tariffs that create scarcity where none exists are like “fixing” the problem of high-speed travel by putting horseshoes on railroad engines.

  3. I’m not really sure how to respond to your post here – too many issues and resonances are pulling at me at the same time for me to be sure that I’m going to cover them all and at the same time establish a clarity that I’ve clearly failed to manage so far…

    I’m going to start at the end with your final metaphor which seems to slightly spurious to me – I’m trying to determine what the analogous elements are between high-speed travel and web-site functionality / user-interactions and I’m at a bit of a loss. Even in terms of a direct comparison with e-mail I’m a little confused. I can think of no potential abuse of the system of high-speed travel at all, whereas spam constitutes a clear case of e-mail abuse (even if one feels at that point that is innappropriate to do anything about it, which would seem to be your point).

    The closest I can come to a workable analogy here would be if those high-speed train-tracks carried more trains/e-mails than the stations/users could legitimately handle or process. In the absence of being able to upgrade or build new users, one option might be to find some way of restricting the number of trains on the track…

    Which brings me to your concept of scarcity. In fact there is one very finite resource online and that is the individual persons ability to process the information that is thrown at them – this in itself has financial implications (and money is scarce) and time implications (and time is scarce).

    Which is where I think your decision to refer to a badly-scaling curve for user-effort as a ‘tax’ is quite interesting and actually surprisingly apposite.

    You say that such a tax would mean that only the rich could perform certain acts, but who is ‘effort-rich’? What individual exists who has more than twenty-four hours a day to do things in? There’s a natural upper boundary there to the resources available. And in the meantime these are processes that place a small burden on OTHER people’s 24 hours…

    In fact, one could quite argue directly in opposition to your point and say that at the moment the technologically (and web-hosted) ‘rich’ (ie. the people with the ability to send spams and the like) are currently using their power indiscriminately on the technologically poor – and that what a the system needs is a kind of PROGRESSIVE (ie. with a curve that points upwards) taxation that re-democriticises an otherwise hideously iniquitous/unequal system in which the poor pay (in time and bandwidth) for services needed to create a tamed environment for the rich.

    All of this being interesting speculation in and of itself. But let’s extend it for a moment into my specific area of expertise – online community spaces. We often talk about structuring user experiences, making spaces that are most conducive to helping groups of people accomplish a goal (cf. Clay’s recent mostly-useful ship-building metaphor). We accept that community spaces in real-life will have types of ‘functionality’ inbuilt – we can ignore people, walk away from them, recluster elsewhere, we can recognise regular trouble-makers immediately by their appearance (and guess – sometimes prejudicially – whether someone new is likely to be a trouble-maker or not). Many of these are based around us accepting the likelihood that people have stable identities that we can recognise visually. This is not something we can guarantee online – people can have as many identities as they have time to register e-mail addresses. And while that has certain advantages, it also rips a large amount of the ways in which human beings make sense of each other completely out of the picture.

    In a sense, the lack of scarcity of avatars is a consistent problem for community spaces – and if we can build in some kind of scarcity that gives them value, then that can help individuals relate to one another and form meaningful and useful communities. One of the ways I’ve been thinking of building in this scarcity is to find a way to make the process of registering and maintaining a new identity scale badly – ie. it’s relatively easy for one person to maintain one identity, but it becomes much more difficult each time you try to run another one.

    I know that’s the second time I’ve run through that model, but I think it’s an interesting one, because quite apart from suggesting a replacement or new means of governance for a canonical communication tool like e-mail, a discussion forum amounts to separately designed and competing entities online. If a system that restricts abuse creates better online communities then it will be used more and will become a standard whether or not it fulfills your personal criteria for acceptable online interaction or not…

  4. The ability to send mass-email is nonscarce — or close enough thereto. In other words, email gives us the ability to blast tens of thousands, millions, billions of messages at a very low cost (and generally at close-to-zero incremental cost).

    When the phone was invented, it created enormous social problems. Here’s a quote from Sterling’s Hacker Crackdown:

    “In Edwardian Britain, issues of class and privacy were a ball-and-chain for telephonic progress. It was considered outrageous that anyone — any wild fool off the street — could simply barge bellowing into one’s office or home, preceded only by the ringing of a telephone bell. In Britain, phones were tolerated for the use of business, but private phones tended be stuffed away into closets, smoking rooms, or servants’ quarters. Telephone operators were resented in Britain because they did not seem to “know their place.” And no one of breeding would print a telephone number on a business card; this seemed a crass attempt to make the acquaintance of strangers.”
    Lower barriers to entry are lower for all parties — good and ill alike. But while high barriers to entry are easily overcome by the wealthy — who have never lacked for access to mass media since the days of town criers — the low barriers are a boon to both the poor and the unsophisticated criminal (sophisticated criminals are rich).

    The telephone, railroad, telegraph, printing press and bicycle all created new opportunities for mass action and communication at the expense of people’s “finite attention.” Each one has been met with calls to legislate some kind of politeness that would scale back change to accomodate the poor have-nots who are stuck coping with all the newfangled mishegas and gimcrackery.

    The point is that the wealthy and powerful already know how to cope with the status quo, and democratizing communication shears away some of that advantage. Holding the line on the status quo leaves the powerful intact, and robs the powerless of the opportunity to upset the apple-cart.

    Of *course* the rich have more than 24h in a day! They have [number of employees] * [number of hours worked/employee/day] in a day. Think of Violet Beauregarde in Charlie and the Chocolate Factory whose wealthy father had all of his peanut-shellers devoted to the task of unwrapping Wonka bars until the Golden Ticket was discovered.
    All distruptive technologies disrupt. They create new problems for everyone. But disruptive technologies can upend the social order — think of Martin Luther — and give have-nots a break they can never receive under the status quo.

    “Protecting” have-nots by limiting the disruptiveness of new technologies is just a way of ensuring that the status quo remains intact.

  5. Ok – two points. You don’t address the idea that if the rich already have a substantial advantage in spamming, then some kind of ‘progressive’ taxation might be appropriate. Nor do you address the issue of having 99% of the population pay for the 1/3rd of all e-mail on the planet that is categorised as spam – spam sent on the whole by a very very much smaller part of the population who are eager to make money at direct cost to individuals and business… How is that a democratised medium? Most importantly, you don’t move the issue (as I was attempting to in my very first post on the subject) from Joel’s concept of pay-for e-mail towards other kinds of user/site interactions particularly in online community spaces.

    But even your argument on e-mail seems to be imlicitly based on the assumption that the ‘purest’ form of a technology is necessarily the best. It is possible for a new technology to have tremendous positive affects along with a few negative ones – negative ones that might on occasion need to be addressed legislatively or technologically. Every single one of the technologies that you mention has changed and developed since it was created – both technically and legally. Legally, we as users of these technologies have asserted our rights not to be harrassed by telephone marketers by being able to say “Not me, please” either individually or centrally. Cars can only be driven by responsible people of a certain age. Technologically, these technologies do not arrive fully formed – to bastardise your analogy from earlier – I’m sure that early trains didn’t have particularly good brakes – but the simple fact that high speed trains might go faster without them isn’t reason enough to argue that brakes shouldn’t exist…

  6. “Protecting” have-nots by limiting the disruptiveness of new technologies is just a way of ensuring that the status quo remains intact.

    This seems to me to be a ridiculous statement – I’m sure we can all think of new technologies that started off disruptively or even dangerously and had to be worked on to be effective, usable, non-invasive, less financially damaging, less time-consuming and more intuitive – whether or not that simultaneously limited specific kinds of uses. The analogies you’re providing just don’t seem to be holding water to me. Maybe you should be demonstrating why the mass e-mail we’re bombarded with isn’t turning into a completely undemocratic and one-to-many broadcast media?

  7. “You don’t address the idea that if the rich already have a substantial advantage in spamming, then some kind of ‘progressive’ taxation might be appropriate.”

    You’re right, I dismiss it. I don’t believe it’s true that the rich have an advantage in sending spam. If that were true, all of the spam would be generated by rich people.

    “Nor do you address the issue of having 99% of the population pay for the 1/3rd of all e-mail on the planet that is categorised as spam”

    Socialized costs always have disproportionate benefits. 99% of the taxes collected to support public schools is collected from childless people, people with grown children and people who could afford private education.

    99% of the trips taken on the public transit that our tax dollars support are taken for causes that we wouldn’t — going to see movies we hate, going to buy ugly clothes, shopping for .

    99% of all the speech protected under the First Amendment is odious horseshit.

    If your standard for eliminating a category of speech that is currently both libre and gratis by taking away the gratis part is that some substantial slice of it is “bad,” then it’s time for progressive taxes on newsprint, book distribution, television and so forth.

    “Legally, we as users of these technologies have asserted our rights not to be harrassed by telephone marketers by being able to say “Not me, please” either individually or centrally. Cars can only be driven by responsible people of a certain age.”

    The phone network isn’t end-to-end. Services are designed and deployed by a centralized authority. There’s a “daddy” you can appeal to if you don’t like the way the network is being used.

    The Internet is decentralized. Regulating the network to create “progessive” limits on mass communication would require substantial redesign of the Internet — redesign that would break end-to-end by causing different packets to be treated differently at different points on the net.

    It seems like you’re saying, “Free speech has other channels that can support it — how does regulating mass email limit it?” This is the argument that’s raised every time someone attempts to regulate speech in some medium — textbooks, videos, libraries, radio, television, message boards, etc — and thankfully, the judicial interpretation of the First Amendment is clear and uniform.

    Regulation of speeech in one medium is not excused by the unregulated speech of another medium.

  8. Before I start again I want to state that in terms of the e-mail debate, I’m quite prepared to state straight off that the concepts I was exploring are not necessarily practical. I don’t agree with many of your reasons for arguing against them (which I will explore further below), but I’m probably fairly close to agreeing that in and of itself such a proceedure would not operate effectively or appropriately. More importantly, I think you’ve consistently missed my point that I’m interested in those occasions where an increasing marginal cost in effort or usability can have useful and beneficial effects – particualrly in the maintenance of community structures. Such mechanisms limit the potential damage that any individual can cause, while maximising the need and value in collaborative enterprise.

    That’s the stuff that I’m interested in, and have been thinking about outside this debate for quite a while. In fact it’s a point that I’ve found expressed very clearly by Clay Shirky recently when he talked at a recent social software seminar about how individuals needed to be protected from themselves in community spaces, and directed towards accomplishing something rather than just given a space to operate within.

    If you read my original piece again, the point was that Joel’s concept argued that spam would be reduced (ignoring side-effects for a moment) by changing the graph of cost/interactions from this to this. His final point (that if the money went to the person the e-mail was sent to then there would be little or no cost for normal users) was gesturally in the direction of my point, which was that with the aim of making it harder to spam than to write legitimate e-mail it would be better to have a graph like this. That’s all this conversation is about – my suggestion that there might be circumstances where something scaling badly or having a growing marginal cost, a diseconomy of scale – might be a good thing. This example – e-mail – is just what got me onto the subject (and I think within the remit of the discussion that was taking place was a completely plausible move to make), but the subject as far as I’m concerned is this escalating graph – not the example of e-mail itself.

    Having said that, there are some things that I’m finding very uncomfortable about your rhetoric, as well as a lot of things that I simply don’t agree with…

    I’m kind of stunned by your usurpation of issues like free speech in this debate. I don’t believe there is any necessary correlation between people having freedom of expression and not having any way of stopping people doing intensely dangerous or destructive things with a medium – no matter who else it affects, destroys, compromises or costs. People can still sue for defamation/libel of things said on the net for example – and libel seems to be a completely legimate limitation on free speech. And your argument keeps circling around this idea that it is some kind of sin to limit the powers that a new technology provides you with. A denial of service attack is something that is possible within the system – it’s an abuse of scale that could be undertaken by people with the very best of intentions, and just like a huge excess of spam it could bring a part of the network down. We don’t celebrate or countenance the rights of the people who undertake those kind of activities, and why? Because eventually selfish abuses of a system damage the ability of the system to empower citizens and fuel legitimate protest and activism – as well as damaging social networks, basic human interactions and business…

    At which point I feel like I should probably argue with your use of the word ‘democratic’ here as well. Yet again it’s one of those high-impact, rhetorically satisfying terms that throws all arguments into ‘good’ and ‘bad’ sides of “The free” versus “The oppressed”, but it gets debased when it’s cheapened like this.

    In fact the internet is profoundly NOT democratic, and neither is e-mail. Most people on the internet have no influence on the way it is governed – nor does their individual empowerment have any direct relationship with government or civic responsibility. Or in another sense, it’s not available to all citizens by rights – there’s a financial cost involved. In fact, I can think of few (if any) definitions of the word ‘democratic’ that encompass people’s usage of the internet.

    In fact, what you seem to be talking about (and I completely agree with) is that the internet (and e-mail) provide tools that allow people the most potential empowerment to act and fight for the things they think are important. I would think it debatable that a medium swamped with spam e-mail aids that empowerment or radically hinders it.

    To end on a historical point, but I think an important one, I think your decision to reference the First Amendment as if it was a universal tenet of human existence that had a fundamental right to governance on the internet demonstrates a particularly parochial, culturally colonialist and American-centric view of issues of what is right and proper for the internet. Clearly I agree in principle with people having a freedom of expression, but your particular brand of it doesn’t happen to cover me nationalistically, nor is it necessarily one that I would want to. For good or ill, it also doesn’t cover 5.8 billion other people across the world. In my country – for example – inciting racial hatred is considered illegal – and I believe rightly so. Citing it as an cornerstone or foundation in a debate about a world-wide network and communictaions medium makes as little sense and is as useful as citing one of the ten commandments.

  9. “That’s all this conversation is about – my suggestion that there might be circumstances where something scaling badly or having a growing marginal cost, a diseconomy of scale – might be a good thing. This example – e-mail – is just what got me onto the subject (and I think within the remit of the discussion that was taking place was a completely plausible move to make), but the subject as far as I’m concerned is this escalating graph – not the example of e-mail itself.”

    Yup, I understand that this is what we’re discussing. My point is that the cure of ever-increasing costs for mass-communication (and the concommitant damage to freedom of expression) is worse than the disease.

    “People can still sue for defamation/libel of things said on the net for example – and libel seems to be a completely legimate limitation on free speech.”

    Suing effectively for libel in the US is basically impossible. Why? Because the First Amendment trumps claims of libel again and again. In countries where strong libel statutes prevail, such as the UK, libel claims are often used to shut down legitimate expression.

    “And your argument keeps circling around this idea that it is some kind of sin to limit the powers that a new technology provides you with.”

    No. Mass communication at near-zero cost is a SPECIFIC capability brought to the world by email. It is a new form of expression available to the rich and poor alike — where “poor” includes people who own computers. New forms of expression are protected constitutionally, and it is a sin and a social ill to limit them.

    “A denial of service attack is something that is possible within the system”

    SYN flooding is not a form of expression.

    “it’s an abuse of scale that could be undertaken by people with the very best of intentions, and just like a huge excess of spam it could bring a part of the network down.”

    I’m not opposed to stopping spam. I am opposed to stopping spam at the expense of other forms of mass communication that is machine-indistinguishable from spam.

    “We don’t celebrate or countenance the rights of the people who undertake those kind of activities, and why? Because eventually selfish abuses of a system damage the ability of the system to empower citizens and fuel legitimate protest and activism – as well as damaging social networks, basic human interactions and business…”

    In fact, we do celebrate and countenance the rights of:

    * Street protestors

    * Picketers

    * Religious canvassers

    * Sit-down strikers

    “At which point I feel like I should probably argue with your use of the word ‘democratic’ here as well. Yet again it’s one of those high-impact, rhetorically satisfying terms that throws all arguments into ‘good’ and ‘bad’ sides of “The free” versus “The oppressed”, but it gets debased when it’s cheapened like this.”

    Tom, we’re arguing about values. Citing a value in support of a position is not “loaded” or dishonest.

    “In fact the internet is profoundly NOT democratic, and neither is e-mail. Most people on the internet have no influence on the way it is governed”

    The Internet is an end-to-end network. ALL services are defined at the edge. That is PROFOUNDLY DEMOCRATIC. Contrast this with every other form of mediated communication in history. It is the very first medium in which all services and protocols, all message handling — in short, all governance — is defined at the edge, in the computer on your desk.

    “I can think of few (if any) definitions of the word ‘democratic’ that encompass people’s usage of the internet.”

    Really? You don’t consider the ability to publish preely without intervention from censors a democratic use? Or the ability to assemble in arbitrarily large groups? Or the ability to use strong crypto to make yourself safe from unreasonable search and seizure? Killfiles — the ability to selectively ignore anyone?

    “To end on a historical point, but I think an important one, I think your decision to reference the First Amendment as if it was a universal tenet of human existence…”

    Tom, I’m not even an American. Language similar to the First Amendment is embedded in the UN Decaration of Human Rights, the Canadian Charter of Rights, etc etc etc. When I cite it, I cite it because I believe that it is a social good — universally. Freedom to speak, without fear or limitation, is a value that I support.

    The Internet is an American medium. It was invented in the US. 70+% of the traffic on the Internet is routed over US soil. The lingua franca of the Intenret is American English. The rhetorical tactic of asserting the Internet’s lack of borders is in fact very dangerous, since it perpetuate sthe notion that the Internet cannot be regulated.

    In fact, even minor forms of Internet regulation, as with Panama’s UDP filtering against VoIP, can substantially affect the technological possibilities available around the world.

    Policing at MAE East and West could certainly and subtantially regulate the Internet effectively.

    It is not an accident or a technologically determined event that has kept the Internet as free as it is. Were 70% of the Internet’s traffic routed through the Great Firewall of China, or even throught he UK with its strong, libel-based limitations on speech, the Internet would be a very different place indeed.

    The Internet’s freedom can be accounted for by the Bill of Rights of the country through which the vast majority of the packets sent and received passes. The Steve Jackson Games case, which established the need for warrants for Internet wiretaps, shaped and protected the discourse of the Internet.

    Bernstein, the case that legalized strong crypto and hence privacy, shaped and protected Internet discourse.

    A thousand Doe and Roe cases, which safeguarded anonymity on the Internet, shaped and protected Internet discourse.

    As with “democratic,” citing a value in a discussion of values is absolutely fair game. This isn’t reflexive parochialism — it’s a piece of the debate.

  10. On democracy – an interesting set of comments from Robert Putnam:

    Anonymity and the absence of social cues inhibit social control – that is, after all, why we have the secret ballot – and thus cyberspace seems in some respects more democratic … Research has shown that on-line discussions tend to be more frank and egalitarian than face-to-face meetings … Some of the allegedly greater democracy in cyberspace is based more on hope and hype than in careful research. The political culture of the Internet, at least in its early stages, is astringently liberatarian, and in some respects cyberspace represents a Hobbesian state of nature, not a Lockean one. As Peter Kollock and marc Smith, two of the more thoughtful observers of community on the internet, observe, “It is widely believed and hoped that the ease of communicating and interacting online will lead to a flourishing of democratic institutions, heralding a new and vital arena of public discource. But to date, most online groups have the structure of either an anarchy [if unmoderated] or a dictatorship [if moderated]”…

    These is I think are mistakes you are making – (1) confusing your own legitimate investment in libertarian politics and a sense of active and generally positive anarchy with “democratic practice”, and (2) continually focusing only on the subject of e-mail – ignoring the fact that there is scope for new types of network-wide communication (as AIM, Usenet etc have emerged) and/or local online group-practices or web-based communities.

    You say, “My point is that the cure of ever-increasing costs for mass-communication (and the concommitant damage to freedom of expression) is worse than the disease” and that’s completely fair enough. I still don’t agree with the value systems you’re putting into play here, nor am I prepared to agree with your sentiment that the way individuals interact via software online is something inherently “American”, but I’m quite prepared to accept that in the particular case of e-mail at this particular point in time techniques like Joel’s (and my half a step beyond it) wouldn’t be particularly practical.

    I believe if you recast your argument as saying that, “In the case of e-mail alone I believe that the spirit of anarchy and libertarian politics that I believe is fundamental to the spirit of the internet would be compromised by such an enterprise” then you would get no arguments from me… I might even agree with you. In return I’d like you to accept that there might be other circumstances where a reintroduction of scarcity by individuals building communities might have tremendously beneficial effects…

  11. Pingback: plasticbag.org
  12. Pingback: plasticbag.org

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.