Intro

This is less of a review of individual talks and more of a collection of notes from said talks that I tried to put in some sort of coherent prose (rather than the bullet points that I have in my notebook). I haven't always succeeded - in some places the notes didn't have much context, but where possible I put them in a paragraph. In places, I comment on the talk's content - there, I use first person narrative to make it clear it's my thoughts and not the speaker's. Talk blurbs and speakers' Twitter links are available on the ffconf's website. I will add the links to the recordings, once they are available.

"Designing as we want, to create the experiences that we need"
by Florence Okoye

"The Principles of Sociotechnical Design" by Albert Cherns. I only jotted down the title and the author during the talk, but was able to look up a little bit more information off that. The full article can be found here - it's free to read, but it's a collection of scans. I also found a decent bite-size explanation in this blog post. I've taken the liberty to write out the content of the image that summarises the Principles on said blog post (given that the image doesn't have alt text). All the Principles are interlinked, but I'll use the numbering they've been given by Cherns in the original article.

1. Compatibility: Design process should be compatible to the objectives
2. Minimal critical specification: Telling the workers what to do and not how to do it
3. Socio-technical criterion: Variance should be dealt with by the concerned worker
4. The multifunctional principle: Individuals should be able to perform multiple tasks
5. Boundary location: Boundaries should still permit the sharing of knowledge and skills between each group
6. Information flow: Information systems should be properly directed for action
7. Support congruence: Social support should be provided to the workers
8. Design and human values: Criteria for a satisfying job
9. Incompletion: Design is an ongoing process

Design Justice Framework is a visual tool to help explore the relationships between actors in the thing we are designing. To start with, we list the actors in three groups: those who the thing is for, those who might be harmed or excluded by the thing, and those who will benefit or profit from the thing. We then map the relationships between those groups which then allows us to ask ourselves how we can change the relationships to make them more balanced.

On a separate note, it is a good idea to map values and actors on the same page. This allows us to see if one team's goals and values clash with those of another team's in the same organisation.

Florence also mentioned a talk by Alisan Atvur she saw at UX Cambridge in 2018. "The Facilitation Kata" explored how kata (a choreographed sequence of movements prevalent in martial arts) can be used to facilitate various design projects. I was able to locate the slides but unfortunately not the recording.

If you don't have a relationship with a community you can't effectively design for that community. The point Florence made here is that very often we have cishet white people designing for communities they know nothing or very little about. It made me think about how many conferences are filled with "male, pale, and stale" speakers - I know in tech the organisers are trying to ensure some representation, but unfortunately academia lags behind. I also heard about the confusion of people who set out to create Grindr for heterosexual people. Now, the main point of Grindr is that the users disclose their location and solicit intimate encounters and the target demographic is gay men. And this is where Florence's point about relationships with the community is clear - the creators of this new app couldn't comprehend why they didn't have any women signing up to their service… It didn't occur to them that maybe, just maybe, women don't like giving out their location.

Overall, my take away from this talk is to think about the relationships between various actors of the projects I work on; try to balance skewed relationships; and to make sure to seek out project contributors from the communities the project is aimed at.

"Capitalism, The Web, And You"
by Heydon Pickering

Heydon's talk's main subject was a heavy one: capitalism. Every so often there was an intermission in the form of a light and funny slide or a CSS tip. That's why my first note reads "Crabitalism" by Horx Myxln. The CSS tips included accent-color property which adds styling to forms, and the gap property for Flexbox.

We all know that capitalism is about money, loads of it. The point that Heydon makes with the help of baked goods, is that if you have a pie, you can't ~enlarge~ said pie. Capitalism is the idea that we can always make more money. In reality, money is just like the pie, we can't magically make more of it. In a closed system, you can't grow more resources, you can only redistribute them. Gain in one place means loss elsewhere. Making a profit means either underpaying the workers or inflating the value of the goods. Good examples of this are companies who publicise record profit but refuse to give raises to their minimum wage employees and the insulin producers in the US respectively.

Heydon pointed out that burnout wouldn't exist without capitalism because there is a clash between a desire to do good and someone else's desire to make a profit.

Heydon also talks about how heavily the web these days is influenced by ads. We are tracked and our on-line personas are exploited for targeted advertising. Funnily enough, research shows that targeted ads performed slightly worse than assigning the ads randomly.

The talk made an excellent point: feeding the trolls generally makes people money or makes an influencer out of said troll. It's the controversial opinions and culture wars that drive the engagement which gives the opportunity for advertisers to piggyback on the fame of a tweet. So even though we know that the ads on-line rarely work, that doesn't stop said ads being omni-present - maybe giving companies an impression they are reaching way beyond their current client base.

In "The age of surveillance capitalism", Shoshana Zuboff introduces the concept of "The Last Virgin Wood" - that the last thing that capitalism can take from us is ourselves.

Heydon points out that these days British schools forbid sharing anti-capitalist materials and teaching about the disadvantages of capitalism. This made me think of hi-jacking Karl Marx's quote - that these days it's not religion, it's the money that's the opiate of the masses: we centre our lives around making money and spending money, stuck in a vicious circle that benefits very few, in a system that punishes other-thought.

I really like that ffconf makes space for talks like this one, where the primary subject has very little directly to do with tech. This talk was a great food for thought, exposing the weak points of this system. However, it feels incomplete, somehow. Both my colleagues and myself wished that there was something we could DO to make things better, but as things stand, most of the attendees don't have the power to redistribute their companies' profits. Overall, somewhat disheartening, but much needed talk.

"Design for Developers"
by Lex Lofthouse

Lex made clear that the way the talk is structured may be viewed as controversial - she lists the four core principles and four core elements, and as with many of these things, depending who you ask, they might disagree with you and offer a different list.

The four principles of design covered in this talk:

1. Hierarchy - things are arranged by their importance
2. Proximity - using distance to convey the relationship between elements on the page
3. Contrast - visual opposites for emphasis
4. Balance - distribution of visual "weight" - symmetrical versus asymmetrical

The four core elements covered in this talk:

1. Typography. This includes distinction between headings and paragraphs and what font pairings are used. Font should embody or compliment the brand (I think a good example of wrong choice would be a law firm using Comic Sans). In general, it's better to play safe! Spacing between characters and paragraphs also falls under this design element. Resources include fontjoy.com and type-scale.com.
2. Colour. Start with one main colour and build on it. Then think about colour palettes: main, supporting, and neutral. Resources include colour.adobe.com and coolors.co.
3. Imagery. Consider what the primary focus of the image is and how it interacts with the background - where is the eye drawn to. Consider using a special effect to unify photos taken in various places - for example black and white, or colour-pop. unsplash.com is a service where you can find free stock images. Iconography is part of imagery. Don't assume users understand what an icon means. Some icons are universal - a bin to delete and a magnifying glass to search. But others, such as a heart, are much more ambiguous. It's best to label your icons to avoid misunderstandings. fontawesome.com is a great resource for icons.
4. Layout. It's hard to go wrong with a 12 column grid - it divides by two, three, four and six offering great flexibility.

Lex also talked about the atomic design as a framework to build reusable components. The main idea is that you start with atoms - your buttons and images. You put those atoms together to create a molecule - what I'd often refer to as a "card" - which holds the information and actions pertaining to a single item. And then you build an organism out of the molecules - a shop page would contain a card for each of its products. You can find more information about this on atomicdesign.bradfrost.com.

I found this talk to be extremely well structured and full of clear information that I can fall back on if ever need to contribute to designing a website. And I'd like to point out that despite my earlier dig at Comic Sans, it's a font that I loved as a child, and it has a special place in my heart. It is a joke in the typography world, but its spacing and letter shapes make it one of the more legible and accessible fonts that are available for all to use.

"This Talk is Under Construction: a love letter to the personal website"
by Sophie Koonin

Sophie urges us to rediscover the purpose of a personal website. She points out that many of the websites these days look kinda the same these days. A vast majority of them offer services - a stark contrast to the nineties where most of the websites were about personal interests. These days websites are built to gain money and rarely for pure pleasure. And although the websites may offer different things, many are owned by a handful of big-name companies (I think a good example of this is Amazon, who at the time of writing own Audible, Goodreads, IMDB, Twitch, and several dozen other websites).

Even social media platforms are less about social aspects these days; you see less of your friends' content (if they are still active on the platform) and more adverts. Ultimately, we are the product and our data is sold.

Modern websites are for consumers and not the creators. The latter need to figure out how to please their audience on social media, to create engaging content, so that they get more likes and more money.

Many small businesses these days don't bother with an independent website and rely on a facebook or instagram page to show their menu or their goods. Unfortunately, this excludes people who for one reason or another don't use social media, as you can't view the page content without an account. This made me think of a one-page web comic that must be about a decade old at this point: Matthew Inman's "What I want from a restaurant website". I guess, despite a fairly clear minimum viable product, restaurant owners strive for looking impressive rather than giving the users what they want in a clear format.

Sophie mentioned that the inspiration for this talk was the "Keep the internet weird" talk by Rachel White from JSConf EU 2017.

Facebook and Twitter made it much easier to interact with other people. Also, you no longer needed specialist knowledge to build a blog with a guest book - you could post all your woes and victories on a platform that is much easier to use and much more "common" in the sense that it has many more users. This caused a decline in personal blogging for fun, and those who continued to blog did so with the hope of becoming influencers and gaining engagement.

Sophie shows off her current personal website (as well as a number of old ones). She tells us how she uses the webmentions protocol to build a de-facto comment section on her posts. She admits that building an independent website does rely on being able to own a domain, which is out-of-reach of many.

Sophie's message is clear - building a website is a radical act, and a great place to experiment. When writing a personal website, it's OK to commit bad code and ship it; worst case scenario, rolling back is easy. Personal websites are a lost art that should be brought back and rediscovered with joy.

Sophie mentions a number of resources throughout her talk. indieweb.org is a movement to create people-focused websites. yesterweb.org is a modern website with a retro feel. https://personalsit.es is a directory of people's personal websites; you can add yours to the list by making a pull request on GitHub.

Oh, boy, this talk hit some really sore spots for me. I used to blog using a probably now defunct platform the name of which I can't remember. This was in the early noughties. Long story short, the experience ended up with me being a victim of on-line bullying. These days many of my coworkers post Weeknotes - one blog post per week describing what they've done that week. Or they share the ups and lows of their lives on Twitter. I used to do that too, throughout university, using Facebook statuses or Tweets - probably as a means of procrastination, given how many of those were tracking the progress of the word count for various essays. My current Twitter is full of re-tweets of various causes I believe in, and I can't remember the last time I posted anything on Facebook. I find it really difficult to reconcile the desire to share my life with others - with the expectation of interaction, and the knowledge that the web these days no longer affords the anonymity it did in the nineties. I am also much more security conscious - sharing my birth date exposes a crucial piece of personal data; and with enough random information it would be possible over a long term to build a profile of me - and I'm wary about possible consequences. If I go on holiday and see something cool, I'd like to have a back and forth conversation with someone about this, rather than sending a photo into the void of the internet. I find direct messaging friends, or using small discord channels much better for this than Twitter - it also allows me to cultivate better relationships.

I think I would really struggle with finding the purpose for a personal website. If I were to put together, for example, a recipe site, I would like for others to tell me how their attempt has gone - help them troubleshoot any issues and use their comments to improve my recipes. Without that, I might stick with my paper recipe cards - it's much easier to write them up that way, take a photo to have a backup on the cloud. It's also much easier to stick a recipe card to the fridge, and a recipe card doesn't go to sleep as you're kneading! You might say - you put together this very website, surely it's the same! I don't think it is - the main reason for this website is that I want to own my content, and websites such as Medium can randomly put any content behind the pay-wall. If I post something that allows others to learn, I want it to be free and accessible to all. I guess you can say that the existence of this website is spurned by hate rather than by love… Build websites out of spite?

Off the top of my head, I know one website that Sophie would love - it's both very much a personal website, and is beautiful in its simplicity. I'm talking about sheldonbrown.com - the go-to page for bike maintenance and knowledge. Sheldon passed away in 2008, but his glorious website has been kept alive and updated by various entities throughout the years.

On a final note, I would like to thank Sophie for the reassurance that HTML and CSS are enough. Despite my coding skills paying the bills, I get overwhelmed by all the options and possibilities. This website is hand-coded HTML with a tiny sprinkle of JavaScript. It very much does the job.

"Programming with Yarn"
by Lily Madar

In her talk Lily lists all the similarities between writing code and crochet. Note, she picked crochet because that's her yarn craft of preference, but the same ideas apply to most, if not all, fabric and yarn crafts.

Both have

• strong communities
• knowledge sharing
• concept of deliberate mistakes (Persian rugs to avoid being perfect and "it's not a bug, it's a feature, I promise")
• online centralisation (Ravelry and Github)
• language / region specific naming (crochet UK vs US terminology)
• scalability
• repeats (loop recursions)
• reverse-engineering

In the early days, computers used punch cards - patterns - to perform actions.

Designing fibre items is akin to Machine Learning / Artificial Intelligence.

I have surprisingly few notes from this talk. I think a big reason for that is because I've been doing fibre crafts for just shy of two decades and a lot of what Lily mentioned are things that I know and therefore don't need to write down.

I think the crucial take away from this talk is not the similarities, but the difference between software development and fibre arts. As Lily pointed out, the practice of actually writing down patterns is relatively new - yarn companies wanted the buyers to be able to replicate their patterns without needing to have a person teach them. Because in the past, the patterns lived in people's heads. I can't imagine coding without having a search engine at hand, even when it comes to stuff that maybe I should know off by heart - such as JavaScript's string or array methods. I recently saw this video about Tablet Weaving and how 2500 years ago, weavers made very complex patterns - from what we understand, these were done from memory! It's interesting how the availability of information made the need to remember everything fade away.

"Working towards a greener world from behind the keyboard"
by Natalia Waniczek

Natalia talked about a very important subject that many in software development either don't think about or gloss over: that of the impact our work has on climate change. She recommended OMG Climate community that concentrates on bringing to light the numbers involved.

Music streaming is a huge pollutant. Listening to an album twenty seven times is comparable to the environmental cost of manufacturing the CD. I mentioned this stat to my Husband who then asked - does this include just the CD itself? What about the case, the insert, getting it to the shops, etc? I am a huge proponent of owning physical CDs (as well as books and DVDs), but it feels to me that oftentimes streaming is a better solution - if my Random Playlist has one hundred songs, each from a different album, it seems like streaming is the more eco-friendly choice.

The internet accounts for 3.7 percent of (presumably yearly) greenhouse emissions, which is comparable to the whole of Germany, or the aviation industry. After all, the internet is somebody else's computer. The Data Centres, Data Transfer, and the energy needed to charge and power all the devices all add up. Even websites are "heavier" than they used to be.

Although many organisations boast about carbon offsetting, in practice it's merely greenwashing and doesn't have as big of an impact as we'd like to think.

Sustainable Software Principles aim to build a resilient web, one that works for everyone, including those on older devices, slower internet speeds, and unreliable connections. Using SVGs and CSS is much better than images. And do we really need autoplay, and do we need videos?

There are plenty of resources to help us out: the Website Carbon Calculator, Climate Action .tech community and Principles of Green Software Engineering.

I found this talk interesting because it touches on a very important subject. But the same way Heydon's talk did, this one also left me uneasy about the current state of affairs, but without clear actions we as developers can take. Sure, we can host our personal websites on green servers, but it may not be easy to convince the stakeholders of the companies we work in to spend more money on hosting infrastructure, or abandon the swanky videos. I feel this talk encourages us - the individual consumers - to act responsibly, when in reality, we are a drop in the ocean compared to the huge international companies, or the cryptocurrency mining (which, nota bene, I don't recall being mentioned in the talk), when it comes to digital pollution. The same way that countries pass ableist laws such as plastic straw bans in the name of cutting waste and yet big companies don't face any consequences of their actions (oil spills, discharging chemicals into rivers). The talk also assumes privilege to make greener choices; many individuals host their websites on GitHub, which is not green, but is free. Popular blogging services, LiveJournal and Wordpress fall into the same category. The Website Carbon Calculator can tell you if your website uses green hosting, but if your host isn't green, you're not given clear alternatives. Overall, I agree with the sentiment, but I think this needs to be enacted and enforced top-down.

"Digital exclusion in healthcare & how to change it"
by Sareh Heidari

It is not a big secret that there was a group of people, who massively benefitted from the lockdowns. For the first time disabled people had community gatherings, entertainment shows, education, work, and doctor appointments come to them through the internet. Some faced decades of being told that it's too difficult for them to work from home, or too expensive to stream conferences and events - finally the access was a level playing field.

This also meant that digital stragglers needed to adapt too. They could no longer rely on seeing their GP face to face; they needed to learn to navigate the world of video calls if they needed a video consultation. If they wanted to book a vaccination in the UK, the quickest way to do so was through the website.

Although many these days can't imagine life without the internet, the shift to web-based left a large number of people excluded. Despite government drive, there are still digital deserts in the UK - areas with no, or very patchy internet. People affected may need to go outside to make a video call, and are often left wondering if their on-line submission has gone through (this is why confirmation messages are very important). There is also an assumption that people will have a fixed address; try registering with a GP without one - this affects nomadic communities, as well as already vulnerable groups such as the homeless. And let's not forget, one in five people in the UK (twenty percent of the population) live in poverty. Finally, although many more events were available on-line, people who rely on assistive tech often remained excluded.

The COM model: it takes capability (being able to do something), opportunity, and motivation to trigger behaviour change. The pandemic added more opportunities: with the number of groups migrating on-line it was much easier to find something a person was interested in. It also provided motivation - everyone else was on-line and it was safer to join them. Having said that, motivation can be low if one had a bad experience in the past. How many times does one need to struggle with logging in to on-line banking to abandon it altogether?

So what do we need to consider when building services? What we create should be research led, user centred (and not product centred), and holistic.

Fifteen percent of adults in England have the literacy age of nine to eleven years old. Long sentences and fancy words means those people can be excluded just by a paragraph of text. Jargon makes specialist subjects even less accessible. As such it is important to use plain language and write in an active voice if you want to be inclusive. Many specialist institutions such as the NHS (health service) or Monzo (a bank) have style and content guides to make their literature more accessible. However, such measures are still relatively new. Those subjects have dire consequences if they are poorly understood, and yet forty percent of the population (I think this is in England or the UK, rather than world-wide) struggle with understanding health related communications and as a result don't feel confident making decisions about their health and healthcare.

If we design for marginalised people, we improve the product for everyone!

One way of helping people feel less anxious about navigating on-line is to reduce the number of interactions per page. People who aren't confident tech users worry about breaking things, or doing things the wrong way. And when a mistake means that they may not get the medical help they need, the stakes are high. Single action per page reduces the overwhelm and allows the user to concentrate on one thing at a time. I need to admit that personally I'm not a great fan of forms that have one action per page. I find them frustrating to use, especially when the questions require long written answers. I never know how to answer the question - is there one further down the line which would be a better place to say what I want to say? If I go back to change my reply to an earlier question, will what I wrote on the current question be saved (or do I need to copy paste it into notepad just in case)? Clicking through fifty multiple choice questions is tedious - unless it's important to me, this would make me very likely to abandon filling in the form altogether. And often the "next" and "submit" call to action buttons look the same, or in some cases use the same word - sometimes the "Thank you for your submission" message takes me by surprise.

Digital inclusion means giving users the choice and having options available off-line! Although phone calls are less popular these days, some people feel more comfortable making a call than arranging something through the internet. There is something reassuring about talking to another human. And despite the use of machine learning and artificial intelligence, the help both are often more frustrating and discouraging than helpful.

It is important to learn about lived experiences - make sure to research extensively to help you understand the needs of the target audience, especially if they are by nature less vocal ones. To reach more people, you may need to do so via community groups. Sometimes it may be necessary to change the way you conduct research to reach a wide variety of subjects. You may need to build trust, and that takes time. I think in this case it's worth linking back to Florence's talk - rather than trying to do so yourself, whenever possible, employ someone within the community to work with or for you. And when it comes to your research results - act on the conclusions you came up with.

Sareh points out that when one provides a service or goods, one has a relationship with their users or customers - whether one wants it or not.

The Good Things Foundation runs the National Device Bank and National (Internet) Data Bank - aiming to bridge the access gap. I was very excited to hear about this because of the number of my old devices stuck in boxes, but it looks like they are more aimed at getting donations from larger organisations.

We should do our best to influence policies. If we need the internet to do anything - healthcare, banking, taxes - and when other options take significantly longer and require another person's input (for example a bank teller) - then internet access should be a human right.

I really enjoyed this talk and will definitely share it with my friends who are doctors. I find that medical appointments and banking done online, where possible, are much more convenient. It cuts down on travel time, as well as the waiting around. I need to admit it didn't occur to me that some people are excluded by the same thing I value a lot. Even if some people advocate for libraries (and I love my local library), I can't imagine doing banking or taxes in a public space. Something that this talk hasn't mentioned, but that adds to the exclusion, is that many services these days rely on two factor authentication. It relies on people having access to the same smartphone, and the same phone number. People who experience poverty or are otherwise vulnerable often aren't able to provide that. Even if you have a pay as you go number, you must top up every so often or make a call, otherwise the line gets disconnected. And the money you topped up sometimes expires before you get a chance to use it.

I also want to add that digital exclusion can work in a different way too. For years now, my GP has had Florey Accurx webform set up. It's a webform that asks several questions, allows me to attach a few photos. A GP then triages the requests and can reply with a text message if that's all that's needed. It's a win-win. Granted, the form's only available nine to five, Monday to Friday, but I think it's mostly to manage expectations than anything else. I was surprised to learn that a vast majority of GP practices don't have this set up and rely on ableist tradition forcing ill and stressed people to call on eight thirty on the dot to maybe, just maybe, get an appointment. The webform does need a doctor to do the triage, but they are much better qualified to do that than the receptionists who look after the phone lines. I think the issues here are short staffing and cost cutting, but it's frustrating that something that could massively improve triage isn't used more widely.

"Day Disco"
by Ruth John

This was very much a "how I built a cool thing" talk. The notes I have just mention the existence of

• the Web Audio API
• Roland 50 Studio - an in-browser synth
• "Synthesising Drum Sounds with the Web Audio API" by Chris Lowis article
• "A tale of two clock" by Chris Wilson
• Path2D API
• WebCodecs API

Final thoughts

ffconf has a track record of offering excellent talks, and this year was no different. I personally found the balance of social versus technical versus design talks to be excellent. I also really appreciated Day Disco being the last talk - this was my first conference since the last ffconf, and it was really nice to have a light and fun talk at the end as my brain would have struggled with anything heavier. Looking forward to next year!