• The Startup Game

    July 16, 2014

    Stories of startups are most often told using superlatives. These are stories of founders dedicating themselves to a single idea, prepared to risk it all for a chance of success. Teams putting in extraordinary effort: their sweat, late nights, and swapping socialising outside of work for obsessively improving their products.

    These are always heroic tales of turning talent and youth into success and money. They are so pervasive that in some circles it is automatically assumed that your goal is to either have your own startup eventually, or be a part of one destined to make you rich.

    Even outside of the tech bubble many believe the hype. The government is happy to claim to have played a part in the success of Silicon Roundabout, whatever that success may be. Young people skilled at programming are hailed “the next Zuckerberg”, their companies “the next Google”.

    The sad truth is that most startups fail, but these stories are always presented as post-mortems: we did this, and it hurt us in that way. We ran out of money. We couldn’t scale.

    As Nikki Durkin points out, “the startup press glorify hardship”. As a founder, you’re expected to put on a brave face.

    Ask any founder how they’re doing and you’ll hear something positive. Whether that’s the truth or not, that’s what we’re trained to say.

    I found postmortems of startups outlining what didn’t work and why the company went under, but I was hard pressed to find anything that talked about the emotional side of failure — how it actually feels to invest many years of your life and your blood, sweat and tears, only for your startup to fall head first off a cliff.

    Of the personal stories of my friends most of them are not the happy ones of hardship, survival and success. Mostly they’re stories of workplaces that in the name of innovation and moving fast throw out healthy working practices. I’ve heard of startups doing “agile” by having three stand ups a day. Ones which insisted they were still too small for HR departments despite complaints. Companies where managers were so scared of losing control that they had to micromanage everyone below them. Places which enforced permanent “crunch time”. Stories of unfair dismissals, unforeseen firings, burn out resulting in months off work.

    Anecdotally at least it seems to me that those stories are at least as common as the success ones, but they’re not told as often, though they should be. That’s why I made the Startup Game (mirror here).

  • Goodbye Flickr

    June 01, 2014

    Last year I felt an itch to start saving my digital records from obliteration next time a service I use gets shut down or sold. For ages I’ve been thinking of dipping the toe in the world of hosting all my own data. paralell-flickr by Aaron Straup Cope sounded like a good project to try.

    paralell-flickr is a really thoughtful piece of software. You still put your photos on Flickr, and keep up with your friends there (or rather, you used to, when your friends were still using it), but on your own server you keep a replica, a copy that keeps track of your contacts, favourites, and permissions. You have to sign into it using your Flickr account and it will know who can see which photos.

    As the readme states:

    parallel-flickr is not a replacement for Flickr. It is an effort to investigate – in working code – what it means to create an archive of a service as a living, breathing “shadow” copy rather than a snapshot frozen in time.

    Aaron wrote up a talk he gave about it, which is worth reading in full.

    I thought that having my own parallel copy of each service I use — Foursquare, Instagram, Flickr, Twitter — would be a good way to preserve my data without hassle, and if I ever decide to delete my account, or for another reason won’t be able to carry on using it, at least I’d already have a complete backup in place.

    The moment when parallel-flickr came in handy happened sooner than I anticipated.

    I created my main account in 2010, but have another two which date back to 2008 and 2006. Flickr lets you personalise your URL, so that’s what I did, every time, and I’ve grown to regret it. The URL on my main account has a name in it that I don’t go by anymore. It annoys me because it denotes gender that doesn’t match mine, so I want to change it.

    But Flickr wasn’t built with that kind of flexibility in mind. You can change your username, and all the other details, but not the URLs. Once you choose one, that’s it, game over. Permanence of the URLs is more important than your comfort, your life, your future choices and changes you cannot anticipate.

    There are a million reasons why you’d want to change the URL other people find you by. Perhaps it was funny 7 years ago when you called yourself grandma_disco_fever but now you no longer do. Maybe mother_of_three was a good moniker a while back, but now you need to increment that number. What if you called yourself wife-of-someone but you’ve since divorced? Maybe these things don’t immediately spring to mind in the first year of the service’s existence, but once you’ve spent many years building an archive, full of connections and relationships, it becomes hard to leave it behind just so you don’t have to be a prisoner of your past URL choice.

    We don’t arrive in the world as fully formed, unchangeable entities. We’re not finished and permanent, and when dealing with URLs of people it seems deliberately harsh to demand permanence. Especially when the HTTP spec accommodates redirection to new ones.

    So, Flickr, I think this is goodbye.

  • Chrome obfuscates the URLs, Google benefits

    May 05, 2014

    Chrome Canary introduced a new feature which obfuscates the website’s URL.

    A member of the Chrome team mentions that “the whole point is to prevent phishing”.

    Here’s a screenshot from Canary in which the feature is enabled: Chrome Canary and its URL hiding feature

    You’re viewing a specific blog post, but you wouldn’t know it.

    The only way to get the URL is to click on the domain name. When you hover over it you get a hint that it’s clickable, but otherwise it’s not obvious.

    Accessing the URL in Chrome Canary

    Clicking the omnibar hides the domain name entirely and prompts you to search for a phrase or enter another URL.

    So who benefits from this?

    It’s not the people using the browser; it’s the search engine makers and big social networks. The ones so well-established that people willingly put their share buttons on their own websites.

    Let me repeat that: the parties benefitting from this change are the search engine vendors and big, monolithic social networks that strive to create walled gardens.

    If this approach becomes the default, people visiting sites will be unaware of where they are. To share interesting things on the web they will have to use the actions provided by the site maker. Often these are reduced to sharing on Twitter and Facebook. Gone are the days when people were provided with more than a handful of options. Any new players on the market, any new bookmarking and sharing tools will simply not stand a chance, as only their size will affect adoption of their own sharing buttons.

    A huge slew of sharing and bookmarking tools

    Beyond options provided by each individual site creator sharing will be limited to power users and those who understand what URLs are (and how to find them). That group will get progressively smaller, as URLs become unfamiliar and rarely encountered in their full form.

    Marketing campaigns already frequently suggest searching for a specific phrase instead (or alongside) providing a URL. If you can’t simply point to a URL, then you have to maintain first position in search results in all dominant search engines to help people find your site. The search engine becomes the necessary—and only—mediator of this interaction, a buffer zone, a lobby, with total control over who gets to access to the thing you made. It’s no surprise really that it was the Google Chrome team who came up with this.

    There’s been much debate about whether the URLs are ‘ugly’ or ‘beautiful’ and whether people really understand them. This debate misses the point.

    The URLs are the cornerstone of the interconnected, decentralised web. Removing the URLs from the browser is an attempt to expand and consolidate centralised power.

    If you really care about usability, there are better ways of highlighting the domain without obfuscating the URL entirely. As Josh Emerson points out, IE has been doing this for a while:

    IE already highlights te domain part of the URL

    This the solution that Remy Sharp is proposing. Highlight the key part of the URL and truncate the rest if necessary, but keep it visible, helping the person viewing the site to see where they really are.

  • Internet of Dependent Things

    March 20, 2014

    In February, Berg unveiled a prototype internet connected washing machine.

    Connectivity holds out promise for something really new, these machines can become something actually different. Cloudwash is an expression of our thoughts on how to make this stuff matter and some steps to something better.

    Promising remote access and a better interface, this convenience comes at a price: lock-in. Lock-in comes in many sizes, big and small.

    Little Printer is certainly small. I’ve had it for a while now, and even though technically I don’t own it (I borrowed it from work) I call it ‘mine’. It’s in my bedroom. Messages and pictures addressed to me print out of it. It’s practically mine.

    But sometimes it reminds me that it isn’t — sometimes it suddenly wears glasses, or a mask. One time it grew a moustache. There’s a setting that allows me to change the appearance of the printer, but it only affects the hairstyle. I have no control over anything else.

    It will say things like “the Little Printer thought I looked cute in glasses” or “the Little Printer thought the glasses were too big and I had to give them back”, which reminds me that I don’t have full control of the device. The ‘Little Printer’ which the device refers to has control, but who is it? There is no explanation. Someone else is pulling the strings, controlling aspects of appearance and even gender expression (which I don’t get to pick).

    I’m assuming it’s designed to give this object an opportunity to appear as if it has a life and identity independent of the owner, to make it more lively, interesting and playful; but for me it just regularly reminds me that someone else is in control.

    Someone can remotely do what they please with a device that has been given access to my bedroom. What the printer can do is limited to printing stuff, of course, but I intensely dislike thinking about the reality: that it’s an expensive piece of single-use hardware that someone out there can make redundant in a heartbeat. Still, the Little Printer is just a toy, so i’m not so worried about the loss of control.

    It’s not so easy to trade off convenience for ownership with other appliances or devices. For my phone, the loss of control is too much to bear. The very idea that someone gets to decide what I’m not allowed to do with my pocket computer is unacceptable. Although i’m not as attached to my washing machine as my phone, the loss of control still scares me.

    It’s not a nightmare scenario, but there are plenty of things that could go wrong:

    • — What was once built in functionality becomes an in-device purchase. You used to just turn on the drying programme, but now you have to subscribe to it.

    • — The cloud service is unprofitable, and is switched off. The module that communicates with the cloud is tightly integrated into the appliance which is now obsolete.

    • — Instead of being sunsetted, it’s sold off. A new agreement, new pricing structures. Sign in with Google+ to enable pre-wash rinse. Like Zanussi on Facebook to get a free quick wash.

    • — Ads start appearing on the digital display. Remove them for $3 a month, or customise the machine with your own messages for $5.

    • — Someone hacks my washing machine and floods my home.

    • — NSA gets another way to monitor what I do and when I’m home.

    Third-party access to my domestic appliance creates a power disparity between the manufacturer (or service owner) and me. They can use their power to generate profit in ways that didn’t exist before, forcing me to pay in ways that go beyond the purchase of the appliance itself.

    I make trade-offs daily about which privacies and freedoms to give up, and in exchange for what. Some are worth it and buy me closer connection with friends, or some useful convenience; others are foisted upon me because I have to make them in order to do my work; but some just to go too far.

    I resent that the meaning of an acceptable trade-off is shifting toward less privacy, less control and towards tipping the balance in favour of for-profit companies and convenience for governments who want to spy on everyone.

    Maciej Cegłowski puts it way better than I can:

    What upsets me isn’t that we created this centralized version of the Internet based on permanent surveillance.

    What upsets me, what really gets my goat, is that we did it because it was the easiest thing to do. There was no design, forethought, or analysis involved. No one said “hey, this sounds like a great world to live in, let’s make it”. It happened because we couldn’t be bothered.

    It doesn’t have to be this way. Open projects could fill in the usefulness of adding connectivity to appliances. They could open-source the design of the hardware (or instructions on how to put it together), and the software it runs on. The owner wouldn’t be reliant on the manufacturer to make improvements, or to create versions that can work with different machines, or give them access from different kinds of devices. Ultimately, they could be in control of the hardware and the software involved.

    Just like I would like to see a trend towards decentralisation of the web, I would like the internet of things to become full of decentralised entities, built on the premises of freedom and empowerment, before it’s entirely normal for marketers and governments to live in my washing machine.

  • 2013 Yearnotes

    January 05, 2014

    Last year began with the realisation that as a Polish citizen it was going to be very hard for me to get a US working visa. I missed out on participating in a great secret project, but was looking forward to taking the whole ofJanuary off to have some space to think about things. In the end I never did, because I got tempted to work with my favourite clients on fun projects. Taking holidays is hard when you’re your own boss.

    In March I decided to leave Brighton for personal reasons. Many of my friends live in London, and that’s where most of my clients were, so it seemed like the time had come to embrace it. I didn’t think I was going to like it, but my wonderful friends have helped me make it my home.

    But why just change one thing at a time? In the spirit of total upheaval I’ve joined Makeshift full-time. I liked the idea of being able to work on a project for an extended period of time, and having the chance to properly immerse myself in it. At Makeshift we work on multiple products simultaneously, so my favourite things about freelance work remain, while I get to work on a project from the beginning and stay with it as it matures. To date I’ve worked on Help Me Write, Attending?, Wrangler and Linkydink.

    Despite getting excited about the idea of ”blogging like it’s 2004” and promising myself that I would write more, in 2013 I didn’t actually do that. Partially because I was busy reorganising my life, but also because I was writing and practicing two conference talks.

    I spoke at the Design and Artists Copyright Society and Scottish Ruby Conference about Thinking through Making (video), which was a compilation of some of the reasons for making stuff. You don’t always make things for the end result; sometimes the exploration of a problem space is just as important.

    I also gave a more serious talk, trying to convince the audience of the importance of considering the consequences of design decisions taken when making software. It was titled “Make world less shit NOW” and I’ve presented it at Nordic Ruby, JS Conf EU, as well as a shortened version at Apps World in London. You can watch the video from JS Conf EU.

    In March and December I ran workshops for BA Design students at Goldsmiths for the second and third time. In four days I taught them about the basics of HTML, CSS and JavaScript so at the end they could play with APIs and use data as their medium.

    In 2013 I had two long-running side projects. I started recording the Amazeballs podcast with my Best Friend Forever, Linda (♥♥♥). I also realised that even though a very small number of people still use Offbott, those who do really care about the service. I’ve been working on making it more maintainable so I can make some small improvements this year.

    It’s been chaotic, but fun, and I have enjoyed the pace.