Infrastructures, Internet and the future?

A colleague of mine asked me the other day to think about how might users be represented in conversations about what the future Internet infrastructure ought to be like? Tricky question on so many levels! What does it mean to think about the users and their needs and desires when we are talking about Internet infrastructure at the level of packets and protocols? I think that talking to people directly and asking them what they think the internet should be like in the future won't give you anything. This does not mean that this conversation and representing the users and their interests is unimportant or impossible though. In fact it is incredibly important because even though there are many things that can be achieved at the level of applications and policy, modifications at the level of infrastructure will always be the ones hardest to circumnavigate. So if there are particular things that are deemed crucial for the future, the need to be built in at the level of infrastructure.

Here is how I've been thinking about this - if you buy the Bowker and Starr's argument, infrastructure is invisible and that's where it needs to be for people to be able to use it effectively. That is, you want your plumbing to be in the walls and to deliver cold and hot water at the right times, offering a range of pricing options that allow you to get more for less (thnx Laura!). However, the shape of the city plumbing infrastructure and it's basic logic matters. There is a big difference between the plumbing infrastructure that is built around cold water delivered to residences and then heated on the spot vs. hot and cold water delivered via separate pipes coming from a central city boiler station. It implies a different set of costs and concerns down the line directly transferred to the end-user. So this is a question of where the boiler gets put and as a result, who is responsible for its maintenance and how many people get affected by outages if you want to push that metaphor further.

What this kind of toy example brings up are two things:
1. Who is the user - individual home owners? businesses? government organizations?
2. What are the tradeoffs? In the first example there is a lot of flexibility at the end-user level but also a lot of costs shifted directly to the user. There are questions of efficiency here as well as questions of ease of innovation and future development (harder to convince many homeowners to buy new boilers because you want to change the system to a more climate-friendly one for example). The latter system is efficient in different ways but also gives a lot more control to the entities controlling the central boiler. So you can turn the hot water off for everyone if necessary or convenient. The maintenance issues are also different - hot-water pipes require more maintenance.

It's the same thing here. The internet architecture is about how packets (bits of information) are moved from point A to point B. There is nothing else here - just the bits of information and their movement. Even here though, we see how values of designers become manifested in the architecture (consider Sandra Braman's recent work). So ideas of transparency, anonymity and control as aspects of the system are very important.

What happens if we ask the 'user' though (assuming for example that we have decided that our notion of 'the user' are private individuals)? Users most likely won't tell you that anonymity is important to them. In fact, users will probably tell you that for the most part it is not important to them or they haven't thought about it. What is anonymity, how it is operationalized and who is anonymous to whom changes with the group you speak to. If you speak to hackers from the relatively "free" Europe, anonymity and ability to be untraceable is probably one of the most important things to them. However, if you consider similar communities in authoritarian countries, they would emphasize identifiability and transparency to the state as crucial for many of their activities. If you conceive of organizations or governments as your users, untraceability might be most unwelcome. You might want to ensure that governments can not limit content so that a government can not just decide to turn off Facebook because they want to stifle protest. Then again there are many cases where such limits may be arguably necessary (consider closed pedophile networks that often reside on social network sites).

The issue at stake here, as I see it, is that if we leave these things out of the architecture level, if packets will be identifiable and then can be made anonymous at the level of applications or policy, then they can easily be made identifiable again. If on the other hand packets are made anonymous (if the base internet architecture actually does not, can not keep track of identity) then identifiability would have to be created at the level of applications making it harder to retain and easier to slip back to anonymous.

Having said that, I think the question you need to answer first is, what CAN we address at the level of packets traveling through a network from point A to point B?

• Access? - how much effort should we make at reaching the last mile through network architecture (flexible, ad hoc configurable, etc)
• Anonymity? - how much identifiable information about nodes and paths will be trackable and can we institute things like requisite data loss to ensure certain types of identifiable information are never available at the network level (so only available if the user gives permission at the application level).
• Transparency? - who can see what, at what point and what kinds of protocols are necessary to read the contents of a packet given security implementations for data integrity.
• End user flexibility vs. ease of broad-scale development?

Given these, how can we identify which of these are important at the user level (having identified what we mean by the user) and why. This last one then is where asking for what people see as broken might be your first clue to what is important. IF you take infrastructure notions at their base, people only really notice infrastructures and their deficiencies when things break, so maybe we need to ask about what doesn't work first? Then, why is that important? Rank things brought up in these lists. Ask about annoyances (evidence that things are awkward or not working as people want, not integrating well into daily practices) and rank these annoyances too. Finally ask about potential outcomes of removing these annoyances - sometimes you can get regular people to follow trains of thought that can bring them to surprising places.

For example, in my work in Russia and Kazakhstan, I asked everyone the following question:
"If you were to wake up tomorrow and find that all social networks sites suddenly disappeared. Just poof and they are gone. How would you react to that?"
This brought up really interesting initial reactions and then explanations and ideas that at times surprised the interviewees themselves.

There are many ways to ask this question and others have asked it differently. I constructed that question with very specific ideas in mind. I wanted to ask about a loss of a service in a way that would imply that everyone lost access rather than the individual in question for two reasons:

1. if it's just you loosing access and everyone else still has it, this is exclusion and ostracism that has severe psychological and social implications in itself.
2. having people imagine such excision of a service without any effect on others allows people to think what other services or things might come in to replace this given the current state of things, what other ways of doing something similar might be available, and also invites them to consider whether this particular activity is really something they want to continue doing. All of this assuming that everyone else would have to reconsider and shift and change too, so it's about creativity or 'going back to the roots' but not about exclusion.

I think the mental exercise of imagining what the world and your life might be like and how they might have to be reconfigured without something as profound and basic as plumbing, refrigeration or phone lines is an interesting one. It can highlight what people think is the most important aspect of their reliance on these technologies - what they do with it even as they do not think of it in their daily practice. It can also help shape our thinking of how these basic infrastructures might be productively reconfigured in the future. Of course, I have no idea if that is going to be the case.