Blog

The Cookie Monster

05-09-2021

I'm involved in a project at the moment to ensure compliance with the law surrounding cookies. For those that don't know, a cookie is just a somewhat persistent client-side data storage mechanism used by web browsers to enable a variety of functionality in the modern web. There's actually other more modern technologies for client-side data persistence but the law covers that too. They have become a bit of a distraction in the discussion about privacy.

There's different uses for client-side data storage in developing web technologies. One key one is session management - ensuring that when a fundamentally stateless web browser agent is talking to a server that the server knows it is the same agent between requests. Another is storing customisation/user info locally for a site. There's all kinds of use cases but fundamentally it's a technology that allows some information to be left on your device by a server you communicate with that is then sent back to the server in future.

This simple tool, like all tools, can be used for good or ill (and whether a particular use of a tool is good or ill is an exercise for the reader). Cookies can be used for tracking your use of the web, across multiple sites/domains. 'Third party' cookies are ones set by domains/sites other than the one you are accessing (websites/apps frequently pull in resources from a variety of places). Those third party cookies have traditionally been used as a fingerprinting mechanism to allow advertising and other businesses to track your activity across the Internet.

Making people aware of the impacts that erosion of their privacy has on their lives is a worthy goal, but one which this law fails to really help with. It's a classic case of politicians wanting to appear as if they are doing something - by fixing a very visible effect of a problem - rather than legislating to bring about systemic change and address the actual issue.

Currently every website that wishes to be compliant must implement its' own banner or notification that serves to inform users of how the site uses client-side persistence technologies. This is a vast implementation burden pushed down to the developers and maintainers of websites - and also leads to a very inconsistent user experience. It also (much like TLS/SSL warnings and other repetitive security/privacy theatre, e.g. long-winded privacy policies/terms of use) serves to get users used to clicking 'accept' because who has time for that?

Even in the (rare) cases that a site goes to the trouble of actually providing education as to how these technologies work and their implications, and even if the user reads that and understands it, it still doesn't address the real issue - you don't need client side persistence to track people anymore.

So all this law is currently doing, really, is wasting developers' time, training users to ignore warnings/education about their privacy, and distracting from a very real growing issue around how large companies have a vast amount of latent power over the lives and wellbeing of people - both now and increasingly into an uncertain future.

If I were to be involved in making law around educating users about client-side persistence technologies (and digital privacy more generally) I'd suggest requiring browser makers (and other portals to the internet) to provide actual useful education at point of first use. Ensure there's a consistent experience of consent based on adequate information, and focus on empowering people to navigate and leverage the digital world we are co-creating. Ideally via both modifications to the way that browsers/apps/devices and the Internet work, and via changes to the education system to start preparing people for the future.

UK Age Appropriate Design Code

02-09-2021

Today the UK implements the Age Appropriate Design Code a set of standards aimed at protecting children in their use of the Internet. This is broadly based on the United Nations Convention on the Rights of the Child. This is actually quite a well thought out bit of legislation by the looks of things.

But I am curious, why do these protections end when you turn 18? What exactly about that list of protections ought not to apply to someone at age 19, or 30, or 60? Why is the privacy of an adult worth so much less than the privacy of a child? Yes, this is based on the assumption that adults can consensually engage with their own privacy but that assumption is based on two things; that they are educated and empowered enough to be able to understand the systems they are using, and that those systems actually respect and make it possible (and at least somewhat easy) to actually implement those intentional conscious choices. And given the length and complexity of most services' privacy policies and terms of use, and the complexity of data protection legislation in differing jurisdictions, I think both of those points are laughable.

How about this - replacing the word 'child' with the word 'user' - does this sound like an unreasonable expectation for the design of digital services? I certainly wouldn't mind if all digital systems had to conform to a code like this. (Modified from the text available on ICO website.)

  1. Best interests of the user: The best interests of the user should be a primary consideration when you design and develop online services.
  2. Data protection impact assessments: Undertake a DPIA to assess and mitigate risks to the rights and freedoms of users accessing your service, which arise from your data processing. Take into account differing ages, capacities and development needs and ensure that your DPIA builds in compliance with this code.
  3. Consistent application: Apply the standards in this code to all your users.
  4. Transparency: The privacy information you provide to users, and other published terms, policies and community standards, must be concise, prominent and in clear language. Provide additional specific ‘bite-sized’ explanations about how you use personal data at the point that use is activated.
  5. Detrimental use of data: Do not use users' personal data in ways that have been shown to be detrimental to their wellbeing, or that go against industry codes of practice, other regulatory provisions or Government advice.
  6. Policies and community standards: Uphold your own published terms, policies and community standards (including but not limited to privacy policies, age restriction, behaviour rules and content policies).
  7. Default settings: Settings must be ‘high privacy’ by default (unless you can demonstrate a compelling reason for a different default setting, taking account of the best interests of the user).
  8. Data minimisation: Collect and retain only the minimum amount of personal data you need to provide the elements of your service in which a user is actively and knowingly engaged. Give users separate choices over which elements they wish to activate.
  9. Data sharing: Do not disclose users' data unless you can demonstrate a compelling reason to do so, taking account of the best interests of the user.
  10. Geolocation: Switch geolocation options off by default (unless you can demonstrate a compelling reason for geolocation to be switched on by default, taking account of the best interests of the user). Provide an obvious sign for users when location tracking is active. Options which make a user's location visible to others must default back to ‘off’ at the end of each session.
  11. Parental controls: If you provide controls that allow one user to control the actions of another, give the user appropriate information about this. If your online service allows a user to monitor another user's online activity or track their location, provide an obvious sign to the user when they are being monitored.
  12. Profiling: Switch options which use profiling ‘off’ by default (unless you can demonstrate a compelling reason for profiling to be on by default, taking account of the best interests of the user). Only allow profiling if you have appropriate measures in place to protect the user from any harmful effects (in particular, being fed content that is detrimental to their health or wellbeing).
  13. Nudge techniques: Do not use nudge techniques to lead or encourage users to provide unnecessary personal data or weaken or turn off their privacy protections.
  14. Connected devices: If you provide a connected device ensure you include effective tools to enable conformance to this code.
  15. Online tools: Provide prominent and accessible tools to help users exercise their data protection rights and report concerns.

Of course, this would severely affect the ease with which businesses can exploit their users for profit, so...

The joy of iteration

29-08-2021

I've been making parts for an art project recently, which is part of an ongoing larger project to produce a modular system for creating art. It's interesting conceptually how rapid iteration and applying the principle of 'just enough' can get you toward a goal.

Initially I started out using the technologies and techniques I knew to build what worked. In this case that was a bit of wood with some plug-and-play electronic components crudely screwed to it. It worked and allowed me to focus on the software.

Then came new requirements - it needs to be waterproof and modular. Take what's there and modify it to add in those new things. At the same time my knowledge of how to make stuff is improving, and my awareness of available products and manufacturing techniques likewise. Encapsulating/integrating parts into a cohesive whole is tricky. You need to avoid prematurely optimising the system before you know what it needs to actually do, but also produce test prototypes to validate your assumptions.

This led to questions around modular cabling - more to learn - power supply, which brings with it issues of safety. How do you mitigate the risks associated with high voltage AC while still effectively transmitting power over distances too long for 5V DC to work effectively. That led to compromises which inform the design of the system. Non-functional requirements.

Constraints play a huge part in the design process too. The most fundamental constraints are your own knowledge, apparent physical realities, time space and money. Fortunately we live in a world where a huge amount of work has been done by others to create an ecosystem of freely available high quality software and hardware that can be integrated to produce really magical things. This, combined with some happy happenstance led to my choice of things like the FadeCandy and Raspberry Pi platforms to work with. Later I've been experimenting with optimising for cost/hardware size with the ESP 32 platform and others - but initially using what worked already, and what enabled me to make rapid progress with the novel aspects of what I was doing, was a huge boon.

Producing prototypes, even imperfect ones, to enable you to integrate and test things in the real world and iterate toward your overall goal (or even to inform new goals as you adapt to changing circumstances) is important too. Rapid feedback and agility in development mean you don't waste time building something which just won't work, or if you do you minimise the wastage and learn something. In the end, the learning you get from doing is the most valuable part of the process.

This led to the 'fluxnode' concept, modular boxes containing LED control and power supply hardware, networked together to enable scalable lighting displays. Still a work in progress but that's the joy of iterating. I've recently been learning CAD and 3D printing, which is a transformational technology for producing parts - especially 'glue' to combine other readily available components. This is the Unix tools philosophy in hardware, use modular components that work well and write just enough 'code' to stick them together efficiently. I'm going to be working on uploading .stl files to Thingiverse soon.

Even in this, iteration and prototyping is useful. I didn't worry about creating a perfect mounting for my circuit board, I made one which works and then refined it. The rapid nature of 3D printing enables this and it's a huge game changer for design and development. The thing which has made human progress speed up so much in the past few decades is how rapidly you can change software and when you can tweak a 3D model, print it and see how it works in the space of hours or minutes, it starts to make sense to apply the same methodology to physical things as virtual ones - at least to some extent. The 'compile' time for printing a model, and the waste of actual resources involved are now pretty low but still higher than for code.

Even this blog has the same approach. Write just enough code to get started, and then iterate on it over time. The most important thing is to make small incremental steps toward both your goal, and informing what your goal is.

Hello (again) World

28-08-2021

When I was young, I wrote a blog. The internet was new back then, exciting, different. That was the 90s and the 00s, the world looked pretty good then - things seemed quite hopeful.

Then the world got busy, and social media took over from blogging, and I lost my voice somehow through change and the necessity of life. I started blogging to friends, instead of to the world, spilling my heart in chat rather than online. It's not quite the same. You can post on social media and not know who will hear you - algorithms decide how to best market your voice to manipulate others. I am not content, I am not for sale.

I don't know if anyone will read this. I don't really mind. We live in a post-privacy society now, there's a certain inertia to things. Obviously this isn't my full voice, it's what I think won't come back to bite me, which is sad. But it's better than nothing.

If I keep it going, expect to see musings on philosophy, inevitability, technology, society - who knows. Mostly I just needed a first post to start building a blog engine around. Let's see how this goes.

This blog of More Possibility is licensed under CC BY-NC-SA 4.0