Audacious Fox

HTTPS and the Secure Web

Should every page you visit on the Internet be served over HTTPS? For banks and online stores, the answer is an obvious yes. But what about blogs, decades old web archives, and other bland online data? Do these documents deserve secured connections?


However, for the past few weeks, spurred on by Google’s move to mark HTTP-only sites as “Not Secure” in Chrome, Dave Winer has asked a similar question, and argued — among other things — that this is a move by Google to seize control of the web:

So now Google points a gun at the web and says “Do as we say or we’ll tell users your site is not secure.” What they’re saying doesn’t stand up to a basic bullshit-test. There’s nothing insecure about my site. Okay I suppose it’s possible you could get hurt using it, I’ll grant you that. But I could get hurt getting up out of my chair and going into the kitchen to refill my coffee cup. Life is insecure. When Google says my old site is insecure what they really mean is “This is our platform now, and you do as we say or your site won’t work.” I don’t believe for a minute that Google’s motivation is protecting users. They seem to believe they can confuse users (they can) and that means they can do anything to the web they like. I suppose they can do that too. But it doesn’t mean the web will cooperate. Imho, it won’t.

Mr. Winer, again, earlier this month:

The web is not safe. That is correct. We don’t want every place to be safe. So people can be wild and experiment and try out new ideas. It’s why the web has been the proving ground for so much incredible stuff over its history.

Lots of things aren’t safe. Skiing. Bike riding in Manhattan. We do them anyway. You can’t be safe all the time. Life itself isn’t safe.

If Google succeeds in making the web controlled and bland, we’ll just have to reinvent the web outside of Google’s sphere. Let’s save some time, and create the new web out of the web itself.

PS: Of course we want parts of the web to be safe. Banking websites, for example. But my blog archive from 2001? Really there’s no need for special provisions there.

We’ve got two arguments here: 1.) Google’s change in Chrome to display “Not Secure” on sites that don’t have HTTPS is the first in what could be a series of steps that eventually lead to HTTP sites being automatically blocked by Chrome, effectively killing the HTTP protocol; and 2.) the world isn’t safe, HTTPS isn’t a silver bullet, and there are simply some types of content that provide no risk and don’t deserve to be called out as insecure.

There’s an additional argument, tangential and articulated by Nick Heer:

I also agree with Winer on another key point: enforcing a pseudo-mandatory policy on HTTPS makes it that much harder for someone new to this stuff to even begin to understand it. As Frank Chimero recently wrote, building stuff for the web has become vastly more complicated since even five years ago. I’m happy to keep learning new skills and growing my understanding of what the web can do, but I don’t know where to begin on this modern web. I don’t intend to hold myself up as a barometer of the complexities of modern web programming or anything — I just don’t know what’s going on any more. I’ve been doing this stuff for nearly twenty years. I don’t know how someone who is eight years old could start digging into React, or Node.js, or any of the other modern JavaScript-based ways of writing <h1>hello world</h1>.

Which brings us 3.) raising the barrier to entry (e.g. requiring someone understand how to set up HTTPS before they can get a site online) harms the approachability of creating something new online.

I disagree on all three arguments, but I don’t think they come from unfounded places. I also have great respect for Mr. Winer’s contribution to the web. When Mr. Winer writes, which he does a lot, I read.

However, the last few weeks have left me scratching my head. I don’t disagree with Mr. Winer’s general distrust of Google — I’m skeptical of Google’s motives when it comes to Chrome’s ad-filter or the likes of AMP — but his recent articles leave me feeling that we’ve missed the forest for the trees; that we’re overlooking the importance of encryption because we’re hung up on our sites being labeled insecure, which, truthfully, they are.

Regardless of what Chrome, Firefox, or Safari do, HTTPS is good for the web, and more sites should enable it for their content. Another way to put it: HTTPS is like fluoride. Fluoride is a proven, safe chemical that we add to water to help prevent cavities. Do you need it, if you consistently brush and floss twice a day? Ostensibly, no, but if there’s a way to help protect your teeth in spite of what is otherwise entirely reliant on your own self discipline and understanding of the risks, why wouldn’t you take advantage of it? The World Wide Web is different today than it was when Mr. Winer first created the content he now is struggling to find reason to provide over HTTPS, but that’s not the visitor’s fault. It’s not even his — yet.

Unfortunately, the world wide landscape today desperately calls for us to encrypt what we can. We, as creators on the web, are obliged to help protect the privacy and security of our readers. Enabling HTTPS on a domain doesn’t hurt existing content, but it does provide your visitors with a little more protection, and — critically — it doesn’t require a change in their behavior. They get to keep just using the web.

Not requiring a change in user behavior is important, because most users won’t change. Recently we had some friends going on a mission trip, and we wanted to give them some money to help cover the costs. They sent us the link to the organization’s site, but when I pulled it up and navigated to the donation page, it was still being served over HTTP. Yet, the page had all the trappings of a secure location. Little lock symbols near the form, a NORTON SECURE sticker — everything but the HTTPS. To someone not scrutinizing the location field for the missinghttps://, every other visual indicator suggested that one could safely submit their credit card information. A large “Not Secure” label would have made the actual page security (or lack thereof) immediately apparent.

As for Google’s motives here, this change in Chrome doesn’t set off red flags for me quite yet. They’re doing what their peers are — trying to educate and protect a vulnerable population. I think a more secure web is good for everyone, and if Google wants to start calling out sites that don’t use HTTPS, that’s their prerogative. And unlike Chrome’s built-in ad filtering, Google doesn’t make tens of billions based on whether or not a website uses HTTPS.

The web is a dangerous place to be sure, but in contrast with skiing or bike riding in Manhattan, the consequences of an unsecured web often aren’t immediately felt. If I break my leg while skiing, I’m damn aware that it’s broken — the cause and effect are instantly apparent. But if I’m inadvertently tricked into submitting sensitive content on a site that’s not secure, I may not know about it until months later. Additionally, depending on what sort of information was compromised, it could affect parts of my life not related to the original incident. If someone gets access to my email or collects enough metadata on the content I’m visiting, it could damage me (or others) in ways I can’t even imagine. It’d be like waking up one day, months after a skiing trip to find your ankle is now sprained, but having no idea when or where it happened.

Finally, regarding HTTPS as contributing to the barrier preventing newcomers from getting started on the web — I think that’s a temporary problem. There used to be a time when I wouldn’t have recommended WordPress to someone starting out in web publishing. It took too much time to configure a server, create the database, and manage the updates. But today, you can go to any web host, pay them $5 a month, click one button, and have a WordPress site up and running in minutes. Eventually, enabling HTTPS on a domain could be equally as easy. Some web hosts are already offering free, one-click HTTPS, and with services like Let’s Encrypt, the technology to make HTTPS easy and accessible is rapidly improving. In short, the overhead required to get a site secure is quickly diminishing, and in a few years, it may well be one of the simplest parts of creating your next new thing.

Publishing to the web should be easy, accessible, and extremely affordable. But the content you publish should also be made available through a secure connection, even if you don’t think the content warrants being encrypted in transit. I think providing an HTTPS connection to your content will be as much a moral duty to web developers in the future as making accessible, open, and fast webpages are today. And although the browser vendors need to be kept in check, I don’t think their efforts to call out insecure sites are nefarious — rather, our world has changed, and our experience using world wide web needs to change with it. The more we can help push forward a fully secured web, the faster it will get here, and the easier it will be to maintain.

Sunday, 11 March

Remember how Strava’s anonymized heatmap accidentally revealed the locations and layouts of U.S. military bases? Well, they’ve pushed an update that makes your data’s inclusion in their visualizations more apparent and easier to disable. Good.

Yet, these sorts of programs should really be opt-in, not opt-out. I’m not against using data to do novel and interesting things, but if your company wants to display my data in some extraneous endeavor, it’s on you to convince me why it’s worthwhile, not me to remember to tell you to stop.

Monday, 5 March

Andy Greenberg, Wired:

In the four or so years since it launched, end-to-end encrypted messaging app Signal has become the security community’s gold standard for surveillance-resistant communications. Its creators have built an encryption protocol that companies from WhatsApp to Facebook Messenger to Skypehave all added to their own products to offer truly private conversations to billions of people. And it’s done so as a non-profit with, at any given moment, a tiny staff that includes just two or three full-time coders. […]

On Wednesday, the creators of Signal announced the launch of the Signal Foundation, which will build and maintain Signal and potentially other privacy-focused apps to come, too. WhatsApp co-founder Brian Acton has also joined as the foundation’s executive chairman, his first new role since leaving WhatsApp last fall. And Acton’s not only devoting the next phase of his post-WhatsApp career to Signal, but a fair-sized chunk of his WhatsApp billions, too: He’s personally injecting $50 million into the project.

If you follow the information security crowd, you’ll quickly pick up on a general cynicism towards technology. Who can blame them? Between the Internet of (unsecured) Things and this quarter’s rendition of guess which retailer leaked your credit card, there’s plenty of room for criticism.

However, whenever I see Signal come up, it really does seem to live up to that “gold standard” label. It’s not perfect, and the app has some problems, but the encryption code is peer reviewed and open source, and it’s trusted by some of the biggest public targets in the world. The United States Senate uses it, and, famously, Signal is Edward Snowden’s preferred messaging app.

Sidestepping my own cynicism that comes out whenever “loved app X takes large investment from vc/company Y,” it’s heartening to see the Signal team get a little structure and financial breathing room. I hope the money goes to fund further development and stability of the service, while avoiding the distractions or gimmicky features — like stories — that every messaging app seems to have these days. Signal is not like other messaging apps, and that’s a good thing. The world needs an incredibly secure, focused messaging protocol, and Signal’s now got the resources to continue building just that.

Wednesday, 21 February

Your ‘Lite’ App Should Be Your Only App

I’ve noticed a trend:

Twitter Lite:

Today, we are rolling out Twitter Lite, a new mobile web experience which minimizes data usage, loads quickly on slower connections, is resilient on unreliable mobile networks, and takes up less than 1MB on your device. We also optimized it for speed, with up to 30% faster launch times as well as quicker navigation throughout Twitter.

YouTube Go:

the app is designed to be offline first and improve the experience of watching videos on a slower network; it gives you more control over data usage, by providing choice and transparency into the amount of data spent on streaming or saving videos.

Amazon Kindle Lite:

Kindle Lite is the new lightweight app built specially for a great reading experience even on slow networks and with patchy connectivity. It is less than 2MB, works on slow networks, and occupies less space on your smartphone.

Google News Lite:

In the new Lite mode things look a little different — we keep the headlines and trim the rest of the components down to their essentials so that the app loads more quickly (and uses less than one-third of the data).

Facebook Lite:

Facebook Lite is less than 1MB so it is fast to install and quick to load. It includes Facebook’s core experiences like News Feed, status updates, photos, notifications and more.

Android Oreo (Go Edition)

With our new and reimagined Google apps, we’ve focused on making them not only smaller, but smooth and fast too. For example, Google Go—a new app to find the information you want—optimizes data by up to 40 percent, weighs less than 5MB in size, and makes it faster to find popular and trending information with a simple, tappable interface.

CNN Lite:

In Hurricane #Irma’s path with a weak phone connection? Stay up to date with the text-only version of our website http://lite.cnn.io 

What part of being fast, data conscious, and reliable is exclusive to old devices or those on poor networks? Why does Twitter Lite feel more like Twitter than anything the company’s done with their main website or app over the past few years? Are Facebook, Twitter, and Google truly so married to ads, analytics, and A/B testing frameworks that their only shot at making a reasonably sized, fast app is to start fresh? Will these lite variants actually stay that way, or will the bloat slowly creep back in?

Here’s a thought: the lite version of your app, service, or website should be your only app, service, or website. And if you’re just starting out, build the lite variant first, then stop.

That said, I do think “lite” is the appropriate moniker. Not because it’s the best label for these lightweight alternatives, but because the regular offerings are tragically obese.

Tuesday, 20 February

I remain highly skeptical of Google — who made $27.2 billion in ad revenue last quarter — having any say in what ads Chrome will or won’t display. That said, the initial launch and implementation of Chrome’s native ad filtering seems honest enough, for now.

If, however, the whole thing leaves you feeling a bit icky, Firefox Quantum is a great alternative to Chrome. I’ve been using it for the past few months and have yet to find a reason to switch back.

Saturday, 17 February

Raymond Wainman for the G Suite Developers Blog:

Today, we’re announcing AMP for Email so that emails can be formatted and sent as AMP documents. As a part of this, we’re also kicking off the Gmail Developer Preview of AMP for Email-so once you’ve built your emails, you’ll be able to test them in Gmail.

AMP for Email opens up several new possibilities:

  • Design interactive components for email using a large library of supported AMP components like amp-carousel, amp-form, amp-bind, amp-list and more
  • Help your content stay up-to-date and interactive for your users.
  • Create more engaging and actionable email experiences

Go check out the gif of what an AMP email can do. Basically, it brings the interactivity of a tiny webpage to your email. Devin Coldewey, TechCrunch:

The moat between communication and action is important because it makes it very clear what certain tools are capable of, which in turn lets them be trusted and used properly.

We know that all an email can ever do is say something to you (tracking pixels and read receipts notwithstanding). It doesn’t download anything on its own, it doesn’t run any apps or scripts, attachments are discrete items, unless they’re images in the HTML, which is itself optional. Ultimately the whole package is always just going to be a big , static chunk of text sent to you, with the occasional file riding shotgun. Open it a year or ten from now and it’s the same email. […]

AMP is, to begin with, Google exerting its market power to extend its control over others’ content. Facebook is doing it, so Google has to. Using its privileged position as the means through which people find a great deal of content, Google is attempting to make it so that the content itself must also be part of a system it has defined.

Google being hellbent on slowly, methodically suffocating simple, durable, and universal tools like RSS and email frustrates me. Email thrives in its lack of sophistication and — as anyone who’s accidentally pressed send too early knows — permanence once delivered. This, at times, can be annoying or limiting, but the alternatives would undermine email’s immense usefulness.

This isn’t about innovation, either. AMP critics aren’t against matured technologies becoming better, but you have to do it without bifurcating the core format. Additionally, if Google’s concerned about the user experience of email, they already have a good initiative going: email actions. These are small tags in emails that allow Gmail to extract flight previews, add one-click “track this package” buttons and more to your messages. These are invisible, additive, and — frankly — convenient things to have; and all without fundamentally changing the original email. Extract all you want, but don’t replace the spec.

We haven’t even talked about spam yet, either. Can you imagine interactive spam? Maybe Google’s spam filtering is robust enough to save Gmail users, but if AMP in email becomes as widely used as they intend, they’ll have handed spammers and malicious actors a whole host of new tools to phish and deceive users.

The email experience can certainly be improved, but it needs to be approached as supportive tools around the email message, not replacing the message entirely.

Saturday, 17 February

Natasha Singer, New York Times:

The courses are emerging at a moment when big tech companies have been struggling to handle the side effects — fake news on Facebookfake followers on Twitter, lewd children’s videos on YouTube — of the industry’s build-it-first mind-set. They amount to an open challenge to a common Silicon Valley attitude that has generally dismissed ethics as a hindrance.

“We need to at least teach people that there’s a dark side to the idea that you should move fast and break things,” said Laura Norén, a postdoctoral fellow at the Center for Data Science at New York University who began teaching a new data science ethics course this semester. “You can patch the software, but you can’t patch a person if you, you know, damage someone’s reputation.”

Computer science would benefit from an equivalent to the medical profession’s Hippocratic oath. As the complexities of computer systems — especially A.I. and machine learning — increase, the easier it becomes to disregard or remain ignorant to the damage these tools can inflict. Personally, I’m still unsure where the ethical line should be drawn, or to what degree, say, an open source software maintainer is responsible for the eventual usages of her code. Maybe some? Not at all? This, to me, is where any comparison of the medical field to the computer science field becomes futile; the doctor creates actions, a computer scientist creates tools. While both products can be used unethically, a tool can operate independently and in ways the creator never imagined. So should the tool have never been created in the first place?

I always liked Google’s now-defunct mantra of “don’t be evil,” because even if the motto was only paid lip service during its final years, it served as a reminder that technology can be and is used for evil every day. So while these systems are too large to blame any one developer or computer scientist, it’s on all of us to not only discuss, but also come to an agreement on the boundaries of what technology should do and how wide-ranging its influence should be.

Tuesday, 13 February

Angela Guzman, retelling the story of her 2008 summer in Cupertino, where she and fellow designer Raymond created several hundred of Apple’s original emojis:

My first emoji was the engagement ring, and I chose it because it had challenging textures like metal and a faceted gem, tricky to render for a beginner. The metal ring alone took me an entire day. Pretty soon, however, I could do two a day, then three, and so forth. Regardless of how fast I could crank one out, I constantly checked the details: the direction of the woodgrain, how freckles appeared on apples and eggplants, how leaf veins ran on a hibiscus, how leather was stitched on a football, the details were neverending. I tried really hard to capture all this in every pixel, zooming in and zooming out, because every detail mattered. And for three months I stared at hundreds of emoji on my screen. […]

Sometimes our emoji turned out more comical than intended and some have a backstory. For example, Raymond reused his happy poop swirl as the top of the ice cream cone. Now that you know, bet you’ll never forget. No one else who discovered this little detail did either.

Apple’s visual approach to emoji is not only beautiful, but also fascinating when you consider how flat-looking iOS and MacOS are today. In fact, if you put a designer in front of iOS for a few hours and then had them draw up a few emoji concepts, you’d probably get images with far fewer textures, no gloss, and little to no depth. But that’s not what we have, and I’m glad. Additionally, I’ve always liked how Apple’s emoji feel like a distillation of and tribute to the original Mac OS X interface style, Aqua. I don’t necessarily miss all the realistic leather patterns and pill-filled buttons, but sometimes a little skeuomorphism goes a long way, and the current emoji feel just right.

Monday, 12 February

Marissa Grunes, The Paris Review:

Primitive Technology was created two years ago by a man in Queensland, Australia, who builds huts, weapons, and tools using only naturally occurring materials. In all of his five- to ten-minute videos, the man wears only navy blue shorts, rarely looks at the camera, and never speaks.

It’s a niche concept, to be sure. The channel does not focus on historically accurate building techniques. It does not offer explanatory tutorials. It will not even help you survive in the wilderness: the “fire sticks” with which he ignites tinder require at least twenty-four hours to prepare and look fiendishly hard to use. So why have the videos attracted millions of viewers? And what do viewers like myself seek when we watch the channel on loop? What do we get from it?

One answer is often floated. Amid the online flood of glossy DIY demonstrations, the paranoiac alarums of super-wealthy “preppers” (people preparing for an apocalyptic event), and the cynical commentary of survivalists, Primitive Technology offers something different: quiet. A few minutes of the channel can make you feel as though you are out in the Australian forest, breathing the sun-steeped, eucalyptus-tinged air, washed clean by rain. The slow precision with which the man undertakes each step of his projects—from finding materials to shaping his tools to assembling his finished structures—lends the videos a soothing sense of purpose. On the Internet, where lunacy sometimes seems to prevail, these videos bring a kind of meditative calm.

Monday, 12 February

The Privacy We Give Away

Last November, Strava — the “social network for athletes” — released their annual global heatmap of user activity or “a direct visualization of Strava’s global network of athletes.” The report consists of 1 billion activities, 3 trillion latitude/longitude points, and over 10 terabytes of raw data. In short, it’s a staggering amount of personal data, anonymized and aggregated, and overlaid on a map.

For two months, the report made little fanfare. But this week, Nathan Ruser, an Australian university student studying the Middle East and security, pointed out on Twitter that Strava’s heatmap revealed more than just popular jogging paths. Alex Hern, The Guardian:

In locations like Afghanistan, Djibouti and Syria, the users of Strava seem to be almost exclusively foreign military personnel, meaning that bases stand out brightly. In Helmand province, Afghanistan, for instance, the locations of forward operating bases can be clearly seen, glowing white against the black map.

Zooming in on one of the larger bases clearly reveals its internal layout, as mapped out by the tracked jogging routes of numerous soldiers. The base itself is not visible on the satellite views of commercial providers such as Google Maps or Apple’s Maps, yet it can be clearly seen through Strava.

Tim Cushing, Techdirt:

Strava does allow users to geofence “private” areas to prevent tracking in those areas. But it’s not a default option. If you don’t want to share every movement with Strava, you have to opt out. Most users don’t. And most users are seemingly unaware of how much data they’re leaving behind.

This “metadata” – something our government refers to as harmless when gathered in bulk – can result in real-world security issues.

And Jeffrey Lewis, The Daily Beast:

No one is really at fault here, other than individual users who may have violated security procedures. What the heat map does illustrate, though, is that we’re living in a very different age than the one where we developed a lot of our ideas about deterrence and strategic stability.

The amount of data the average smartphone user generates on a day-to-day basis is tremendous. Even when that data is anonymized and presented in aggregate, the results can reveal patterns and routines we might otherwise think are private. It’s fair to ask whether Strava should have attempted to scrub the more sensitive data from their results, but the longterm solution is to educate friends, family, and our military about the personal information we’re passively giving away.

This whole story reminded me of back in 2010, when website Please Rob Me used public Twitter and Foursquare checkins to demonstrate how easy it was to know when someone was away from home. Please Rob Me was the first social-networking PSA I remember where freely shared, public data was used to illustrate opportunity for malicious intent. Since 2010, the issue has only become more widespread, as hundreds of millions more smartphones have started cataloging the edges of everything we do.

Following this explosion of mobile devices, an inexorable side-effect is that we now live in a sort of reverse herd immunity when it comes to privacy. Meaning that even if I don’t have any social media accounts or smart devices, my face (and voice) can still end up in the background of an Instagram photo or video, to be later analyzed by Facebook’s image processing A.I. and added to some database of faces — along with the time and place the data was captured. All of this just by walking in the park, going out to eat, or doing any number of public and private activities. I’m not saying it’s reasonable to expect complete anonymity when you’re out in public — that’s never been the case. Rather, when you consider how our devices are not only exposing our own routines and habits, but also filling in the metadata portraits of those around us, it’s easier to see how important the next decade will be when it comes to personal digital privacy and any laws that support or strip away those rights.

Given that companies won’t stop trying to learn more about their customers, smartphones won’t become less capable at recording our surroundings, and we the people won’t suddenly become any less lax about clicking through “I agree” prompts, the situation can seemingly only be improved by our laws or the device manufacturers. Writing privacy into the law takes time, but the European Union has already taken steps towards this with the European Data Protection Directive, and I expect (hope) we’ll see similar efforts or echoes of it from other countries in the future.

However, for now, the fastest road to broadly available increased security and privacy protection lies in the hands of the smartphone/speaker/Internet-connected-device manufacturers. They control the hardware and (to an extent) underlying operating systems. Implementing things like end-to-end encryption, differential privacy, or even making it visually apparent when apps are using your location are examples of ways to help educate, inform, and protect us while we wait for more comprehensive, enforceable protections to be written into law. It’s not an ideal situation, but it’s all we have. In the meantime, take some time to poke around the privacy settings of your most used apps. You’ll probably be surprised at the data you’ve been giving away.

Tuesday, 30 January

DuckDuckGo Founder and CEO, Gabriel Weinberg:

Today we’re taking a major step to simplify online privacy with the launch of fully revamped versions of our browser extension and mobile app, now with built-in tracker network blocking, smarter encryption, and, of course, private search – all designed to operate seamlessly together while you search and browse the web.

The DuckDuckGo browser extension and mobile app will also now show you a Privacy Grade rating (A-F) when you visit a website. This rating lets you see at a glance how protected you are, dig into the details to see who we caught trying to track you, and learn how we enhanced the underlying website’s privacy measures. The Privacy Grade is scored automatically based on the prevalence of hidden tracker networks, encryption availability, and website privacy practices.

Simple, fast, and powered by a company that cares about your privacy. On today’s web, that’s pretty much all you can ask for.

That said, and although the browser extensions work well, the new iOS app is a little rough. I have no doubt it’s functional, but there are a few UI oddities that come off as tone deaf to how iOS apps should behave. For example, the app mimics Safari in that when you’re scrolling down a webpage, the browser UI recedes so you can see as much page content as possible. However, unlike Safari, when you begin scrolling back up the page, DuckDuckGo’s app doesn’t immediately unhide the UI. Sometimes it’ll take an extra downward thumb flick to get the controls back. Here’s where it gets weirder though. Start by scrolling down so the browser UI recedes, scroll up half a screen (in most cases the browser chrome will still be hidden), and then scroll down. Suddenly, the UI controls are back. It’s baffling.1

  1. Okay, now that I’m done beating up their app, let me pitch why you should be using DuckDuckGo as your default search engine. I’ve been using their web search for the past few years (desktop and mobile), and it’s great. I rarely, if ever, need to run something through Google, and for times I do, I simply append !g to my query and DuckDuckGo runs the search for me.2 I wholeheartedly recommend you give DuckDuckGo a try if you haven’t already. ↩︎
  2. DuckDuckGo has a whole host of other “bangs” that work in the same way as !g. Some of my most used are !yt for YouTube, !a for Amazon, and !nyt for when I’m trying to find an obscure usage of the New York Times Manual of Style. ↩︎

Friday, 26 January

Founder of Baremetrics, Josh Pigford:

I realized Medium is really great about surfacing content, but it removes the face of it. It neutralizes all content to basically be author-agnostic. It’s like Walmart or Amazon in that you can buy from thousands of different brands, but you rarely actually know what brand you’re buying…you just know “I got it from Amazon.”

Same with content on Medium. Sure, you can see who the author is or what publication it’s on, but ultimately your takeaway is “I read this article on Medium”, and that’s not what I wanted. I wanted to get back to people saying “I read this article on Baremetrics”.

Baremetrics is still going to use Medium, but only to mirror the posts from their company blog after a two-week delay.

I’m not a fan of Medium-the-platform, because I’m not a fan of forfeiting design and content control to a company that (a) treats every article the same and (b) has yet to show they can turn a profit. Maybe that’s me being a snob who knows how to run his own site and there are lots of people who just want a simple place to write and publish, but I’d argue that — today — self-publishing a blog is a weekend project for most, and the benefits of owning your content outweigh whatever fleeting readership Medium might send your way.

Regardless the platform, you should own the domain and design of your words. Medium as a second-hand distribution network or social graph is fine, but I think people who are all-in on Medium without a backup plan are nuts.

Friday, 26 January

From the Minecraft: Education Edition product blog:

We are thrilled to bring chemistry into Minecraft with a new update for Minecraft: Education Edition. With input from chemistry teachers, students and Minecraft Global Mentors, the Chemistry Update offers educators and learners a fun, accessible way to explore chemistry within the immersive world of Minecraft. […]

Science education is driven by hands-on learning, but only half of fourth graders in the U.S. do hands-on science once a week. In low income schools, the numbers are even lower, as students have less access to labs and equipment. Chemistry in Minecraft allows teachers to introduce chemistry concepts without the costs of lab equipment in the engaging Minecraft world that will inspire more girls and boys to explore the subject.

I looked through the Chemistry Update documentation and was struck by how naturally a chemistry focus will fit into the build-everything-from-scratch, blocky world. It makes perfect sense. Additionally, knowing a few scientists, I can vouch for how expensive lab equipment is. Doing lightweight chemistry in Minecraft will sacrifice the safety and hands-on learning a lab provides, but if it affords more kids an opportunity to explore the sciences, there’ll be plenty of time for the lab later.

Sidestepping the fun of creating Helium so I lift pixelated farm animals into the sky with balloons, I’m fascinated with and curious how far Microsoft can bring Minecraft into the classroom and what, if any, advantage it will bring them. Between Apple with iPads, Google with Chromebooks, and (recently) Microsoft with cheap and rugged Windows 10 laptops, three of the largest technology companies are waging a war for the next generation of users, and it starts in the classroom. So far, it seems Google has the early advantage, having been quick to the scene with easy-to-manage, cloud-based, and, importantly, cheap hardware.1

Microsoft can’t (yet, ever?) out-integrate the Chromebook/G Suite combination, but they can position Minecraft — which, shocker, doesn’t run on Chromebooks — as a much-adored, interactive classroom necessity. Now that Microsoft has a line of cheap hardware options, it might actually be a route some schools are willing to try.

It remains to be seen whether Minecraft will or can be the reason any school chooses one vendor over another, but it’s certainly off to a strong start. Last November, Microsoft announced that — one year after release — Minecraft: Education Edition had hit two million licensed users across 115 countries. Meanwhile, Minecraft (the standard edition), recently sold over 120 million copies, with 55 million monthly players. Minecraft is one of the most popular video games in the world.2

Give it a few more years, and I wouldn’t be surprised if we look back at Microsoft’s 2014, $2.5 billion acquisition of Mojang, Minecraft’s parent company, as one of the better industry investments of the past decade. This Chemistry Update will certainly strengthen the Education Edition offering, but Minecraft’s true strength lies in its brand and LEGO-like universal appeal. I got into Macs because I loved Mac software.3 I could see Minecraft playing a similar role for Microsoft — laying a foundation of goodwill that leads new consumers right to Windows, the Xbox, and whatever else comes next.

  1. And sure, Chrome OS is a garbage fire, but it’s tightly integrated with Google’s host of online G Suite (Docs, Gmail, Classroom), which many institutions can get for free; it’s a compelling package. ↩︎
  2. Anecdotally, I’ve not found a child who (a) didn’t know what Minecraft was or (b) didn’t immediately launch into an explanation of how they strategically organized their wheat fields for maximum yield. Now, go tell those same kids that their school is considering bringing Minecraft into the classroom. They’ll go nuts. ↩︎
  3. As of today, you can buy Minecraft on PlayStation and MacOS. However, both of these systems missed out on last year’s Better Together update, which brought cross-play to Xbox, mobile, Windows 10, and VR. ↩︎

Monday, 22 January

I’m not going to mock an interface that was obviously done by someone who might have a few other important things on their plate, but this is an incredibly misleading design. Thank God you can only send mobile alerts from here and not, you know, actual missiles.

A practical, real example for why you should start your interface designs by focusing on the words.

Aside: Although the correction took too long to get out, I thought the message on Hawaii’s digital highway signs, “missile alert / in error / there is no threat” was particularly terse and efficient.

Correction: The State of Hawaii is saying the original screenshot shared with the media was only meant as an example. The actual screen can’t be shared due to security concerns, but the state emergency agency has since provided a “more accurate” example, which you can see in Marcel Honore’s piece for the Honolulu Civil Beat. Looking at both images, and assuming they’re close to the original interface, I think my comment of “incredibly misleading” still stands.

Tuesday, 16 January

iA, a design and consulting agency, in a ruminative entry on their company website where they hypothesize about a future where A.I. is left unchecked:

As crazy as this may sound, all of this is not Science Fiction. It is happening right now. Machines already filter, sort and choose the information we base our decisions upon. They count our votes. They sort the tasks we spend our time on, they choose the people we talk to and meet. More and more key aspects of our lives are decided by information technology. And things go wrong. Machines are made by humans. As long as we make mistakes, our machines make mistakes.

When things go wrong, both parties—those who use machines and those who build, manage and own information technology—decline responsibility. Users hide behind their lack of power, owners hide behind “the algorithm”. They sell artificial intelligence as Deus ex Machina and when it fails they blame the machine as a mere machine.

The question “Who serves whom?” is not a topic for experts in 2047. It is a key question for all of us, today, right here and now. Whether or not machines can be intelligent is not just technically or scientifically relevant, it is existential.

Imagine if Facebook decided who you could marry, because the network knows more about you and your tastes than you do yourself. What if machines made better politicians? iA raises these and other scenarios, as well as several suggestions for safeguards we can put in place to help keep more distinct lines between human- and machine-generated content.

When most people think of world-ruining A.I., they might think of a robotic uprising. Legions of drones, smart light bulbs, and other sentient machines overthrow their creators. That’s how the movies paint it, at least.

In reality, damaging, world-shaping A.I. will look less exciting, and we’re already getting glimpses of it. Consider the vast influence and reach that A.I. and machine learning have over the algorithms that feed our social networks, populate our search results, and prompt us with contextual lifestyle suggestions. How much of your life is guided by a machine, and to what degree does that influence sway your decisions? Now consider that these same algorithms and neural networks are proprietary code, often being written and improved by engineers who rarely fully understand the complexities of the whole system. At what point does the situation become a problem, dangerous, or irreversible? iA:

We need to know who runs these robots. And we need to know how they work. Bots have no right to anonymity. Algorithms that influence human existence on the deepest level shouldn’t be trade secrets.

I don’t think we have any clear answers yet, but that doesn’t mean we shouldn’t be asking the questions. If anything, these topics need to be more common in our daily lives.

A common example of the dilemma we’re in goes like this: what should happen if a self-driving vehicle suddenly needed to decide between killing a group of pedestrians or swerving off the road and into a ditch, putting the passengers at risk? Should it consider the passengers’ health? What about the pedestrians’ apparent age? Regardless, in either of those outcomes, we should want to know everything we can about the driver. Will we hold machines and their makers to the same standard?

Monday, 15 January

Adam O’Fallon Price:

Periods, commas, colons, semi-colons: in their use or non-use and in their order and placement, can represent elaboration, conjecture, doubt, finality. And in aggregate, over the course of a text, the rhythms of punctuation advance an author’s worldview and personality as surely as any plot or theme. Patterns of punctuation usage are the writerly equivalent of an athlete’s go-to moves, or a singer’s peculiar timbre and range—those little dots and squiggles, in a sense, encode your voice. Anthony Powell’s colon (pardon the inadvertent image) is as signature as Kyrie Irving’s crossover or Rihanna’s throaty cry.

For me, there is no punctuation mark as versatile and appealing as the em dash. I love the em dash in a way that is difficult to explain, which is, probably, the motivation of this essay. And my love for it is emphasized by the fact that many writers never, or rarely, use it—even disdain it. It is not, so to speak, an essential punctuation mark, the same way commas or periods are essential. You can get along without it and most people do.

I wholly, unequivocally love the em dash. Less pretentious than a semicolon, more beautiful than a comma. The perfect punctuation mark—you know?

Monday, 8 January

Start With Words

I start most of my interface designs in TextEdit. Why? Because it forces me to focus on the words. Words are like stock in a delicious stew. If you don’t have good stock, the whole meal feels off.

TextEdit-the-design-tool is surprisingly robust. Let me give you an example.

Say we’ve got to design a rather standard looking confirmation modal with two options for the user. Open TextEdit and start writing:

Are you sure you want to cancel?

Your account will remain active until 2/1/2018.

- If you resubscribe by 3/10/2018 you will retain your data.
- If you resubscribe after 3/10/2018 you'll have create a new account.
- You may download your data until 3/10/2018.

[ No ] [ Yes ] 

You can almost see the modal, can’t you? You can tell that “Are you sure you want to cancel?” is some sort of header. You also know that [ No ] and [ Yes ] are buttons. If you’re feeling fancy, you can even bump up the border radius:

( No ) ( Yes )


Now here’s where the real magic happens — the editing. Before opening Sketch or a new HTML document, continue to iterate on the language. Sometimes it helps to pretend like I have to actually speak these words to someone who’s sitting right in front of me. How would I write this for another human to read?

Are you sure you want to cancel?

Your account will remain active until 2/1/2018.

Come back anytime before 3/15/2018 to reactivate your account or request a copy of your data. After 3/15, your account and data will be permanently deleted.

[ Nevermind ] [ Cancel my account ]

Better. Fewer bullets, more readable.

Constraints help create better designs. TextEdit forces me to focus on the smallest (yet arguably most important) part of any design: the words. Because there’s no functional UI yet, my copy can’t hide behind appealing aesthetic, and I’m less prone to overwriting.

TextEdit (plain text mode, or CMD + Shift + T) is what I use, but pen and paper, Google Docs, or — yes — Sketch work equally as well. Whatever you use, start with words, and then build your design from there.

Thursday, 4 January

Remember how AlphaGo — Google’s Go-playing AI — went 5-0 against the reigning, 3-time European Go champion? Well, Google’s DeepMind team is back with AlphaZero, and it’s just crushed the game of chess. Mike Klein, writing for the Chess.com company blog:

A little more than a year after AlphaGo sensationally won against the top Go player, the artificial-intelligence program AlphaZero has obliterated the highest-rated chess engine.

Stockfish, which for most top players is their go-to preparation tool, and which won the 2016 TCEC Championship and the 2017 Chess.com Computer Chess Championship, didn’t stand a chance. AlphaZero won the closed-door, 100-game match with 28 wins, 72 draws, and zero losses.

Oh, and it took AlphaZero only four hours to “learn” chess. Sorry humans, you had a good run.

Take a minute to notice what’s interesting about this narrative. It’s not that a computer defeated a human, like with AlphaGo, it’s that one computer defeated another computer. At what point will it be more entertaining to watch two AIs compete instead of two humans? While you’re chewing on that future scenario, let me point out why AlphaZero’s accomplishment is being seen as more impressive than AlphaGo’s, despite Go being significantly more complex than chess.

With AlphaGo, Google’s researches, “trained the policy network on 30 million moves from games played by human experts”; essentially, “watch how the best do it, and use that information to choose your moves”. However, AlphaZero wasn’t trained on some huge dataset of chess matches. In fact, AlphaZero wasn’t “trained” at all. Back to Mr. Klein:

This would be akin to a robot being given access to thousands of metal bits and parts, but no knowledge of a combustion engine, then it experiments numerous times with every combination possible until it builds a Ferrari.

At least we still have physical sports. For now.

Sunday, 17 December

The style guide itself is good, but what drew my attention was the visual representation of tone ranges. Personally, I find the visual a far more clear example of what the Firefox UI copy should read like than a bullet list of examples would have conveyed. I also appreciate how a range implies flexibility for the writer to approach the language from a few different angles. A good reminder that product design writing isn’t about how it’s written, it’s about how it reads.

Wednesday, 6 December

Thanks for reading. Check out more in the archive.