Rose to green?

Market sentiment is tricky, particularly when it turns. You could argue that Tech (capital T, whatever that means) had a good run, but it has certainly stumbled in recent years, as the halo of promise made way for a shadow of uncertainty.

Responses to that Tweet raised some good tipping points, including a trend-leading article from the New Yorker, as well as reminders of some of most high-profile stories (Fowler and Uber, Cambridge Analytica and Facebook around the 2016 Election). More recently there has been controversy at YouTube around their treatment of complaints by Vox’s Carlos Maza, Twitter is accused of providing a platform for white nationalists, while Facebook’s approach to altered videos has been scrutinized. Meanwhile fear of Deepfakes abounds, and Elizabeth Warren wants to break up Big Tech. Everything indicates that the rosy glasses have finally been removed, and may even have been replaced by green tinted spectacles*, revealing an unhealthy pallor of unforeseen consequences.

And yet I write mid-way through a banner year for tech IPOs, as stalwarts continue their strong 2019 and new direct-listing Slack closes its second day of trading to general acclaim.

One could argue that there are larger macro factors at play. But regardless, the financial markets at least are out of step with the popular narrative on Tech, and it will be interesting to see whether that divergence continues. There is a line of thinking that the industry still hasn’t taken ownership of the problems it faces; I wonder if a reckoning in the stock market would be required for Tech executives to take notice.

* I don’t know whether this phrase exists so please indulge me.

An ordinary legacy

I saw a Tweet by Eugene Wei a while back about words that can mean both one thing and almost the opposite thing.

I responded with an old favourite of mine, “quite”:

I was at a session today where I came across another in “legacy”. Legacy has a general meaning, “something left or handed down by a predecessor”, which can obviously go one way or the other. But it also has some slightly more specific definitions.

In legal terms, a legacy is a gift of property, especially personal property, as money, by will; a bequest. Which is generally considered a good thing, both to leave and to receive.

However when you are talking about software or computing (which was the topic today), if something is described as legacy (“a legacy system”), that almost certainly means it is bad, but somehow still in use. That meaning is also invoked in more general settings (“legacy banks” are a particularly popular target in Fintech London-town).

It might be tempting to attribute this to simple neophilia, out with the old, in with the new. But I think it is actually that fundamentally human trait Jeff Bezos describes in a recent shareholder letter, describing human beings as…

“divinely discontent. Their expectations are never static – they go up. It’s human nature. We didn’t ascend from our hunter-gatherer days by being satisfied. People have a voracious appetite for a better way, and yesterday’s ‘wow’ quickly becomes today’s ‘ordinary’.”

We don’t appreciate the work and the sweat and the “of its time” imagination of these legacy things – they have been normalised, they are ordinary. And surely we can improve on ordinary.

It is humbling that the best case scenario for technologists is to create something that is ultimately considered ordinary. But maybe it should also give us a renewed appreciation for things we think of as “legacy” right now.

Conference encounters

It is London Tech Week this week apparently. It was also CogX from Monday to Wednesday, which I think is unaffiliated. It all feels a little like a technology Valentine’s day, fabricated and promoted by conference organisers to extract money from startups and interested people.

I found myself in possession of some free CogX tickets, so went along to a couple of talks on Tuesday and had a bit of a wander about.

Being there with no specific agenda, I wondered what it was that my fellow attendees were hoping to get out of the conference.

As an attendee, a conference is ultimately about encounters, whether those encounters are with ideas, individuals or organisations. These encounters can be formal (i.e. part of the agenda of the conference), or they can be informal (enabled by the structure of the conference, but not part of the agenda). They can be anticipated and planned by attendees in advance (“I really want to go to that talk”, “I want to meet that person”), or they can be serendipitous (“I wonder what’s happening in that tent”, “what brings you to the conference?”). Then the outcome of those encounters can be learning, selling (yourself or your business), or just enjoyment (god forbid).

Reflecting on it, I was at CogX to encounter some new ideas via the formal talks, not really feeling like networking that day (shudders). There was one talk I wanted to attend as I knew all the speakers, but the others I picked as the mood took me, so I suppose I erred on the side of serendipity. And my intended outcome was somewhere between learning and enjoyment. By contrast, I had breakfast with someone who was there largely to meet individuals informally and serendipitously, to recruit people for their business (so a “selling” outcome).

Looking around the event and peering at some name badges, the attendees were a very diverse bunch, and no doubt had a wide range of goals (and therefore requirements for the conference). And even two attendees with the same generic goal might have very different requirements. For example if two people are there to learn about artificial intelligence, and one is a secondary school student while the other is a machine learning engineer, the same keynote presentation will struggle to serve them both.

I am therefore interested in how conference organisers think about maximising the utility for their attendees, as they run the risk of trying to please everyone, and ending up pleasing no one. And similarly how do conferences balance ambitions to grow attendance (and revenue) against the likely diminishing and potentially diluting returns as they expand beyond their initial audience.

What’s in it?

When you buy food from the supermarket, you expect to be able to see what is in it. You don’t necessarily want to to check every item you buy for its percentage of sugar or unsaturated fat (or whatever), but you want to have the option. What if you are allergic to something? What if you are watching your weight? You might want to know what you are putting into your body before you eat it.

We don’t have the same expectation when we are choosing digital products. Sure, there are terms and conditions and privacy policies, but nobody reads them, and in any case do they even contain the salient information? We don’t know if we are exposing ourselves to ad-based monetization, or an infinite scroll attention-vortex, or addictive gamification. We don’t know if our activity is being mined for future recommendations, or how our data is being stored or shared, actually. But we wilfully let these applications into our minds and our lives, even though we don’t know what’s in them.

I’m not saying regulation is required. But I do think technologists should think carefully about what they are putting into their products, outside one-eyed KPIs. And I think as users we should be thoughtful about what we eat. So to speak.

Normalhuman

A foundational premise of this blog post is that jokes are better when you explain them in excruciating detail, so can we all get on board with that please.

If you are denizen of tech Twitter, you have probably heard of Superhuman. For those of you fortunate not to spend time on the Hell website: it is an email client, self-proclaimed “the fastest email experience ever made”. It has achieved its Twitter fame through a combination of invite-only exclusivity, high profile advocates and admirers (mostly proclaiming via Twitter), some noteworthy quirks (one-on-one onboarding, a $30 per month pricetag, beautiful pictures when you hit inbox zero), as well as what is apparently a genuinely transformative email experience.

I say apparently because I wouldn’t know – I am an Android user for my sins, so even though I managed to wangle a coveted invite (I know right), I was rebuffed at the final hurdle: there isn’t an Android app yet. My email signature will not indicate that I am Superhuman, at least not for a while.

In a fit of pique, I wondered whether maybe, just maybe, $30 was a bit much for just, like, sending email? Did I even want to see nice pictures at inbox zero? Why would I want an onboarding where someone else critiqued my emails? Was it actually all just about getting a golden ticket, and telling the world about it?

And thus, NormalHuman was born: the email experience you already have, just with a sprinkling of exclusivity.

I know what you are thinking: this is kind of weird, why are you explaining this quite shoddy attempt at satire? Cool your jets, I’m getting there. Because this post is actually an ode to the internet.

Everyone knows that the internet has revolutionised and democratised access to information blah blah blah. All manner of entities, from individual hackers to not-for-profits to big corporates have brought us products and services that are almost like magic, achieving a mix of fortune and fame along the way.

What I think is less well understood is how easy it is in 2019 for a person with an idea to bring that to the world at speed, with internet scale (even if that idea is a half-baked internet parody), thanks to a mix of tools, community and a little bit of luck. Let me break this thing down for you.

The tools

I’m not suggesting that Normalhuman.net is a groundbreaking thing. It is actually quite simple: it is a single page website, with a sign up form, some nice pictures and transitions, a $5 card payment option, and some analytics tracking in the background. I guess what is significant is that ten years ago, that would have been quite an undertaking, requiring a chunk of time and at least some custom development. Today, it took a couple of hours of messing around with some user-friendly point and click tools.

The hub of the whole thing was built on Carrd. For those who don’t know it, Carrd is the world’s simplest one-page website builder, by one-man powerhouse AJ. I have been using it for a couple of years, and I have huge respect for how the platform has developed in that time – it has added some great new functionality, but it has done that while remaining incredibly simple, staying true to its core use-case (easy to make single-page sites). If you have never made a website before, and want to put something of your own on the internet, I can’t recommend a better place to start – it was my first port of call, and I was off the mark.

I wrote some slightly irreverent copy. But a few word jokes does not a good satirical website make. I needed to beef that sucker up.

Superhuman has an access request form, so Normalhuman had to have one too. The act of a couple of minutes in Carrd, linking up a dusty old Mailchimp account I once set up. The internet doesn’t need another wax lyrical about bootstrapped Mailchimp, so I will simply say: it was easy to do.

Given the wonderful aesthetics of Superhuman, I had to make sure my stock photo game was up to scratch. I hadn’t come across Unsplash before, but their photos were beautiful so I snagged a few for Normalhuman, plus their simple embedding API was just what I needed. I got some nice transitions going, and the whole thing started to look kind of professional. We were cooking on gas.

I wasn’t sure what to do about the endorsements from high profile people, a compulsory component of every wannabe unicorn website – I didn’t feel like I could put real people’s faces up there. Fortunately I was reminded of the slightly chilling thispersondoesnotexist.com, an endless stream of AI-generated faces. I could put a face to a quote with impunity (sorry computers).

I talked to a friend of mine, who said I should add a card payment option in there. Carrd of course has a seamless payment integration with Stripe. I thought why not? And added a $5 widget (proceeds to go to Wikimedia). It was my first time using Stripe. Having worked on a couple of payment integrations for bank wires and credit cards in a previous job, it was kind of uncanny how easy it was to go from no Stripe account, to accepting payments. Truly a wonderful user experience.

Then, in a moment of optimism, I added Google Analytics tracking. Needless to say this was also very easy to do in Carrd.

Not going to lie, I kind of gave up when I got to designing the logo. It is the number eight. Don’t ask me why.

The community

Having made a joke website, it could have gone the way of some of my previous satirical efforts – URLs passed on to like-minded friends, otherwise largely forgotten. But I thought I might as well prod it out into the world, so I did two things. Firstly, I created a Twitter account, and secondly I posted it on Product Hunt.

I’m by no means an active Product Hunt user, but I think it is an interesting community which surfaces some cool stuff (I read their daily email). I’d never “hunted” anything before, but thought it might get some attention if I was lucky – the VC Starter Kit (a masterpiece) had done pretty well on Product Hunt, so there was some precedent for joke websites. I Hunted it, pushing it out into the void.

Back to Twitter, where I have only recently started engaging and posting things on my personal account, having spent a decent chunk of time lurking. While I have some issues with the platform (not for now), it does spark some really interesting conversations, and I have really appreciated the fact that you can engage directly with people a world away. It was in the hope that I tweeted my Product Hunt page, and tagged @Superhuman.

A little bit of luck

So NormalHuman got lucky a couple of times. The Superhuman team saw my tweet, and they seemed to get it. A couple of them followed my Normalhuman account, and retweeted or liked or whatever it is one does on Twitter. The site got some traffic over a few days (~4K people!), some people gave me their email address to apply for access. I got a nice feeling that some people had seen the thing I did, and thought it was ok. Which is about all one can reasonably hope for. I was pretty happy with how it had all gone, and was trying to explain to my family what it was (“so if you follow Tech Twitter…”), when I checked Google Analytics – about 1,000 people were concurrently on the site.

Interesting.

To Twitter!

It became apparent that Benedict Evans had included the link on his weekly newsletter, which goes out to >100K people, which explained the sudden influx of traffic. As a reader of said weekly newsletter, it was kind of cool to see my joke website included. I emailed Benedict, who informed me that it was one of the most-clicked links that week. Still not sure what to do with that one.

But I am sure that it totally obliterate my previous metrics, with a halo lasting for a little while as some other people picked it up. As of writing, >17K people have visited the site, >2.5K people have applied for access, and, in the most miraculous outcome of all, 18 people paid $5 for lifetime membership, with all profits to go to the Wikimedia foundation.

Spot the moment the newsletter landed

So that was all good fun.

Some miscellaneous thoughts I may unpack a bit more later: Twitter is an incredible place to spread information (or disinformation). Newsletters with large readerships are a powerful tool. People are pretty willing to give out their email address. People are mostly up for a laugh.

To people who visited the site: I hope it made you smile (I know it’s not like ha-ha funny, we’re in wry smile territory here and that’s ok).

To people who applied for access: I am sorry that I don’t (currently?) have an email client to offer you. GDPR mandates that I let you delete that data if you want, just hit me up if that is the case @GetNormalhuman.

To the people who paid $5: you are absolute angels. I will be making a payment to Wikimedia forthwith.

To team Superhuman: kudos for being cool about it. If you will still have me, I’d love to give the service a try, even just on my old Macbook Pro.

To Benedict Evans: thanks for the boost, that was fun.

To AJ and all the people who made the tech I used: thank you for the leg up. Keep doing what you do.

I am conscious of course that this kind of wide-eyed one-eyed optimism has lead to all sorts of issues for technology companies big and small. There is much more nuanced and thoughtful reflection to be done. But I had an enjoyable micro-experience of the positive power of the internet, and I wanted to share it here.

Time is money

Time is money. Kind of.

Spending time is like spending money. You should think carefully about what you spend it on, particularly if you don’t have much.

Saving time is like saving money. It gives you a little more to spend in the future (please do remember to spend it).

Investing time is like investing money. You do it in the hope of future benefits (it doesn’t always pay off, and that is OK).

You can’t borrow time like you can borrow money, though. So do make the most of what you have right now.

Don’t go passing time, you wouldn’t do that with money. Don’t waste it. Spend it.

Books the remix

Andy Matuschak put out a very interesting piece recently, “Why books don’t work“. While the title is slightly clickbait-y, the content is a substantive and thought-provoking argument that non-fiction books are not actually good tools for conveying knowledge. That the central premise of book-based learning – “people absorb knowledge by reading sentences” – is not generally true*. Instead, people depend on “metacognition” (thinking about what they are reading) to actually absorb and internalize what they have read, which is not something that comes easily. And even those books which appear explicitly designed to facilitate learning (e.g. textbooks) depend on context to actually act as useful learning tools, whether that context is a course, a classroom or a discussion group.

The prescribed antidote is fortunately not an accepting shrug, but instead that this is an opportunity to “make new forms… it is possible to design new mediums which embody specific ideas”. To improve on the book medium “so that its “grain” bends in line with how people think and learn”. Matuschak acknowledges that this is a tall order, but proffers Quantum Computing, an initial attempt with his collaborator Michael Nielsen.

Matt Clifford included the piece in his weekly Matt’s Thoughts In Between newsletter (subscribe!), with the salient observation that this is a bet on human design in the face of the evolved book medium (which is often a brave bet to make). But books are a technology that haven’t changed in millenia (even e-readers try to mimic paper as much as possible), while our means of information transmission have transformed and proliferated unimaginably. Now seems as good a time as any to reconsider this cornerstone of human knowledge. And while Matuschak’s piece says “it doesn’t even necessarily mean abandoning paper”, I do think if something replaces the medium, it will likely be dynamic and tech enabled.

With these thoughts fresh in my mind I was interested to come across Going Critical, by Kevin Simler, which weaves interactive simulations into a very interesting piece on diffusion through a network. The piece is interspersed by illustrative examples that deepen understanding of quite complex phenomena in a very intuitive way. I recommend you give it a read.

This particular approach worked very well, and was very memorable. But I think what struck me was how appropriate it was to the specific topic, but that it wouldn’t necessarily be generalisable. Which lead me to a general observation: I think that if something does replace books, it won’t be another monolithic medium, but a rich mixture of forms. Kevin shared a range of his inspirations. I look forward to seeing others’ explorations, and may even break out of the box myself.

*In a perfect and slightly ironic illustration of this phenomenon, I found myself having to revisit the original article quite a bit to remember the details, having read it a few days ago.