Our sources report that the underlying reason behind the impressive tech demo for Unreal Engine 5 by Epic Games is to ridicule web developers.
According to the Washington Post, the tech demo includes a new dynamic lighting system and a rendering approach with a much higher geometric detail for both shapes and textures. For
example, a single statue in the demo can be rendered with 33 million triangles, giving it a truly unprecedented level of detail and visual density.
Turns out that the level of computational optimization and sheer power of this incredible technology is meant to make fun of web developers, who struggle to maintain 15fps while
scrolling a single-page application on a $2000 MacBook Pro, while enjoying 800ms delays typing the corresponding code into their Electron-based text editors.
…
Funny but sadly true. However, the Web can be fast. What makes it slow is bloated, kitchen-sink-and-all frontend frameworks, pushing computational effort to the browser with
overcomplicated DOM trees and unnecessarily rich CSS rules, developer
privilege, and blindness to the lower-powered devices that make up most of the browsing world. Oh, and of course embedding a million third-party scripts to get you all the analytics,
advertising, etc. you think you need doesn’t help, either.
The Web will never be as fast as native, for obvious reasons. But it can be fast; blazingly so. It just requires a little thought and consideration. I’ve talked about this recently.
May 27th, 17 years ago, the first release of WordPress was put into the world by Mike Little and myself. It did not have an installer, upgrades, WYSIWYG editor (or hardly any
Javascript), comment spam protection, clean permalinks, caching, widgets, themes, plugins, business model, or any funding.
…
Seventeen years ago, WordPress was first released.
Sixteen years, eleven months ago, I relaunched I relaunched my then-dormant blog. I considered WordPress/b2/cafelog, but went with a
now-dead engine called Flip instead.
Fifteen years, ten months ago, in response to a technical failure on the server I was using, I lost it all and had to recover my posts from backups. Immediately afterwards, I took the opportunity to redesign my blog and switch to WordPress. On the same day, I attended the graduation ceremony for my first degree (but somehow didn’t think this was worth blogging
about).
Fifteen years, nine months ago, Automattic Inc. was founded to provide managed WordPress hosting services. Some time later, I thought to myself: hey, they seem like a cool company, and
I like everything Matt’s done so far. I should perhaps work there someday.
There comes a point where you’ve run out of new lockdown activity ideas, and you just start combining random pairs of activities you’ve already done. This morning’s first
activity was… “Pyjama Party / Water Fight”.
Is it just me, or does “Pyjama Party / Water Fight” sound like a PWL
song title?
When the COVID-19 lockdown forced many offices to close and their staff to work remotely, some of us saw what was unfolding as
an… opportunity in disguise. Instead of the slow-but-steady
decentralisation of work that’s very slowly become possible (technically, administratively, and politically) over the last 50 years, suddenly a torrent of people were discovering that
remote working can work.
The Future is Now
As much as I hate to be part of the “where’s my flying car?” brigade, I wrote ten years ago about my dissatisfaction that remote
working wasn’t yet commonplace, let alone mainstream. I recalled a book I’d read as a child in the 1980s that promised a then-future 2020 of:
near-universal automation of manual labour as machines become capable of an increasing diversity of human endeavours (we’re getting there, but slowly),
a three- or four-day work week becoming typical as efficiency improvements are reinvested in the interests of humans rather than of corporations (we might have lost sight of that
goal along the way, although there’s been some fresh interest
in it lately), and
widespread “teleworking”/”telecommuting”, as white-collar sectors grow and improvements in computing and telecommunications facilitate the “anywhere office”
Of those three dreams, the third soon seemed like it would become the most-immediate. Revolutionary advances in mobile telephony, miniaturisation of computers, and broadband networking
ran way ahead of the developments in AI that might precipitate the first dream… or the sociological shift required for
the second. But still… progress was slow.
At eight years old, I genuinely believed that most of my working life would be spent… wherever I happened to be. So far, most of my working life has been spent in an office, despite
personally working quite hard for that not to be the case!
I started at Automattic six months ago, an entirely distributed company. And so when friends and colleagues found themselves required to work
remotely by the lockdown they came in droves to me for advice about how to do it! I was, of course, happy to help where I could: questions often
covered running meetings and projects, maintaining morale, measuring output, and facilitating communication… and usually I think I gave good answers. Sometimes, though, the
answer was “If you’re going to make that change, you’re going to need a cultural shift and some infrastructure investment first.” Y’know: “Don’t start from here.” If you received that advice from me: sorry!
(Incidentally, if you have a question I haven’t answered yet, try these clever people first for even better
answers!)
More-recently, I was excited to see that many companies have adopted this “new normal” not as a temporary measure,
but as a possible shape of things to come. Facebook, Twitter, Shopify, Square, and Spotify have all announced that they’re going to permit or encourage remote work as standard, even
after the crisis is over.
Obviously tech companies are leading the way, here: not only are they most-likely to have the infrastructure and culture already in place to support this kind of shift. Also, they’re
often competing for the same pool of talent and need to be seen as at-least as progressive as their direct rivals. Matt
Mullenweg observes that:
What’s going to be newsworthy by the end of the year is not technology companies saying they’re embracing distributed work, but those that aren’t.
…some employers trapped in the past will force people to go to offices, but the illusion that the office was about work will be shattered forever, and companies that
hold on to that legacy will be replaced by companies who embrace the antifragile nature of distributed organizations.
Tomorrow’s Challenges
We’re all acutely familiar with the challenges companies are faced with today as they adapt to a remote-first environment. I’m more interested in the challenges that
they might face in the future, as they attempt to continue to use a distributed workforce as the pandemic recedes. It’s easy to make the mistake of assuming that what many
people are doing today is a rehearsal for the future of work, but the future will look different.
Some people, of course, prefer to spend some or all of their work hours in an office environment. Of the companies that went remote-first during the lockdown and now plan to
stay that way indefinitely, some will lose employees who
preferred the “old way”. For this and other reasons, some companies will retain their offices and go remote-optional, allowing flexible teleworking, and this has it’s own
pitfalls:
Some remote-optional offices have an inherent bias towards in-person staff. In some companies with a mixture of in-person and remote staff, remote workers don’t get
included in ad-hoc discussions, or don’t become part of the in-person social circles. They get overlooked for projects or promotions, or treated as second-class citizens. It’s easy to
do this completely by accident and create a two-tiered system, which can lead to a cascade effect that eventually collapses the “optional” aspect of remote-optional; nowhere was this
more visible that in Yahoo!’s backslide against remote-optional working in 2013.
Some remote-optional offices retain an archaic view on presenteeism and “core hours”. Does the routine you keep really matter? Remote-first working demands that
productivity is measured by output, not by attendance, but management-by-attendance is (sadly) easier to implement, and
some high-profile organisations favour this lazy but less-effective approach. It’s easy, but ineffective, for a remote-optional company to simply extend hours-counting performance
metrics to their remote staff. Instead, allowing your staff (insofar as is possible) to work the hours that suit them as individuals opens up your hiring pool to a huge number of
groups whom you might not otherwise reach (like single parents, carers, digital nomads, and international applicants) and helps you to get the best out of every one of them, whether
they’re an early bird, a night owl, or somebody who’s most-productive after their siesta!
Pastoral care doesn’t stop being important after the crisis is over. Many companies that went remote-first for the coronavirus crisis have done an excellent job of
being supportive and caring towards their employees (who, of course, are also victims of the crisis: by now, is there anybody whose life hasn’t been impacted?). But when
these companies later go remote-optional, it’ll be easy for them to regress to their old patterns. They’ll start monitoring the wellbeing only of those right in front of
them. Remote working is already challenging, but it can be made much harder if your company culture makes it hard to take a sick day, seek support on a HR issue, or make small-talk with a colleague.
These are challenges specifically for companies that go permanently remote-optional following a period of remote-first during the coronavirus crisis.
Towards a Post-Lockdown Remote-Optional Workplace
How you face those challenges will vary for every company and industry, but it seems to me that there are five lessons a company can learn as it adapts to remote-optional work in a
post-lockdown world:
Measure impact, not input. You can’t effectively manage a remote team by headcount or closely tracking hours; you need to track outputs (what is produced), not inputs
(person-hours). If your outputs aren’t measurable, make them measurable, to paraphrase probably-not-Galileo. Find metrics you can work with and rely on, keep them transparent and open, and
re-evaluate often. Use the same metrics for in-office and remote workers.
Level the playing field. Learn to spot the biases you create. Do the in-person attendees do all the talking at your semi-remote meetings? Do your remote workers have
to “call in” to access information only stored on-site (including in individual’s heads)? When
they’re small, these biases have a huge impact on productivity and morale. If they get big, they collapse your remote-optional environment.
Always think bigger. You’re already committing to a shakeup, dragging your company from the 2020 of the real world into the 2020 we once dreamed of. Can you go
further? Can you let your staff pick their own hours? Or workdays? Can your staff work in other countries? Can you switch some of your synchronous communications channels (e.g.
meetings) into asynchronous information streams (chat, blogs, etc.)? Which of your telecommunications tools
serve you, and which do you serve?
Remember the human. Your remote workers aren’t faceless (pantsless) interchangeable components in your corporate machine. Foster interpersonal relationships and don’t
let technology sever the interpersonal links between your staff. Encourage and facilitate (optional, but awesome) opportunities for networking and connection. Don’t forget to get together in-person sometimes: we’re a pack animal, and we form tribes more-easily when we
can see one another.
Support people through the change. Remote working requires a particular skillset; provide tools to help your staff adapt to it. Make training and development options
available to in-office staff too: encourage as flexible a working environment as your industry permits. Succeed, and your best staff will pay you back in productivity and loyalty.
Fail, and your best staff will leave you for your competitors.
I’m less-optimistic than Matt that effective distributed
working is the inexorable future of work. But out of the ashes of the coronavirus crisis will come its best chance yet, and I know that there’ll be companies who get left behind in
the dust. What are you doing to make sure your company isn’t one of them?
A new Automattician who started this week looked me up to thank me for the first in my series of blog
posts (2, 3) about starting with @automattic. I feel
useful! 😍
The owner of a pizza restaurant in the US has discovered the DoorDash delivery app has been selling his food cheaper than he does – while still paying him full price for orders.
A pizza for which he charged $24 (£20) was being advertised for $16 on DoorDash – and when he secretly ordered it himself, the app paid his restaurant the full $24 while charging
him $16.
He had not asked to be put on the app.
…
This entire news story is comedy gold.
So it looks like food delivery network DoorDash try to demonstrate demand for their services by providing them even if you didn’t ask, then show
you how popular they were. So if you run a pizza restaurant, they might start selling your pizzas as “deliveries” to customers, then come and pick them up as “collections” and deliver
them. Because they’re trying to drum up support and show how invaluable they would be to you, they might even resell your product at a loss in order to get customers on-board
early. It’s all pretty slimy, but I’m sure that wherever they’re operating (New York, in this case) they’ve had the common sense to make all the legal language line up.
(If you can’t see the problem with this model, remember that the customers will be reasonably assuming that the restaurant is involved, so when their pizza turns up cold they’ll phone the restaurant and complain and ask for their money back [or slate them
in reviews online]. Plus, let’s not forget that this is a strongarm tactic: once a restaurant has been seen to be offering delivery, customers will be upset if you take the option away…
even if you never actually offered it in the first place.)
Anyway: this guy noticed that his restaurant was on DoorDash without his consent, and that they were selling his pizzas for less than he did. So he ordered them from himself:
he paid DoorDash $160 for the pizzas, DoorDash paid him $240 for the pizzas, DoorDash sent somebody around to pick them up from him and deliver them to his neighbour. Free money.
Next time he did it, the restaurateur didn’t even bother to put toppings on the pizzas. After all, he didn’t need to be eating them anyway! He was just paying DoorDash to pay him (more)
to move them from place to place. The restaurateur and his friend pulled off several off these trades and DoorDash never seemed to catch on. With some investigation, they discovered
that it was probably an imperfect scraper that had resulted in the price DoorDash advertised being lower than the price they
would pay the pizzeria, which immediately makes me wonder whether you could honeypot it with deliberate scraper-traps… (Owing to various bits of work I’ve done in the past, I’m pretty
well-versed in offensive and defensive screen scraper techniques.)
And to finish the news article off, we’re reminded about the attitude of Mosayoshi Son, the CEO of DoorDash’s parent company (which incidentally also tried to buy, and then got sued by, WeWork, demonstrating his financially-savvy). Recently,
defending his company’s general trend to attract venture capital and then lose it very quickly, he compared himself first to Jesus, who was also a “high-profile
visionary who was initially misunderstood”, and then to the Beatles, who “did not become a success overnight”.
Comedy gold I tell you. And now I want pizza. (Especially if I can persuade a stupid startup to pay me to make and then eat it myself.)
A s*** font that f***ing censors bad language automatically.
This is pretty beautiful, in a sick-and-wrong way. It’s a font which contains ligatures that can be automatically used by supported software. But instead of ligatures for things like æ
and œ, this font replaces the letters of common swear words with a glyph that looks like a censor bar. So it’s an automatically-self-censoring font.
Better yet, the authors were aware of the Scunthorpe problem and attempted to mitigate it; this also provides the font’s
name. Unfortunately it’s not possible to do so perfectly without adding ligatures for just about every dictionary word individually (now that would be a font) so words
like shitake and cockerel still get censored. And even where the mitigation works, it produces other problems: e.g. the use of the ligature
Scunthorpe means that the word cannot be broken e.g. hyphenated across two lines. Nor will letter counters work properly.
But I don’t think anybody’s suggesting that this font should actually see mainstream use. Right?
As we’ve mentioned in previous blog posts, the $FAMOUS_COMPANY backend has historically been developed in $UNREMARKABLE_LANGUAGE and architected on top of
$PRACTICAL_OPEN_SOURCE_FRAMEWORK. To suit our unique needs, we designed and open-sourced $AN_ENGINEER_TOOK_A_MYTHOLOGY_CLASS, a highly-available, just-in-time compiler for
$UNREMARKABLE_LANGUAGE.
…
Saagar Jha tells the now-familiar story of how a bunch of techbros solved their scaling problems by reinventing the wheel. And then, when that didn’t work out, moved the goalposts of
success. It’s a story as old as time; or at least as old as the modern Web.
(Should’a strangled the code. Or better yet, just refactored what they had.)
The lockdown’s having an obvious huge impact on strippers, whose work is typically in-person, up close, and classed as non-essential. And their work isn’t eligible for US programmes to
support furloughed workers. So Lucky Devil Lounge in Portland decided to adapt their services into one that is classed as essential by providing a drive-through food service.
With strippers.
This is Erika Moen’s comic about the experience of visiting the drive-through. Her comics are awesome and I’ve shared them with you a fewtimesbefore (I even paid for the product she recommended
in the last of those), of course.
I first came to this series in 2014 – seems like a lifetime ago! – but got called away by something (a problem with the beta of Three Rings‘ Milestone: Promethium) before I could find more than the first one. Today, under our recently reduced
lockdown and in an effort to save the kids from crawling to the walls, we came out for a walk along this side of the woods and quickly found this cache. TFTC!
Jay Foreman’s back with a long-awaited tenth episode of Unfinished London. This one follows up on Why does London have
32 boroughs? and looks deeper into the complexities of the partially-devolved local government of London.
Since I accepted a job offer with Automattic last summer I’ve been writing about
my experience on a nice, round 128-day schedule. My first post described my application and recruitment process; my second post covered my induction, my initial
two weeks working alongside the Happiness team (tech support), and my first month in my role. This is the third post, running through to the end of six and a half months as an
Automattician.
Always Be Deploying
One of the things that’s quite striking about working on many of Automattic’s products, compared to places I’ve worked before, is the velocity. Their continuous integration
game is pretty spectacular. We’re not talking “move fast and break things” iteration speeds (thank heavens), but we’re still talking fast.
My team tackles a constant stream of improvements in two-week sprints, with every third sprint being a cool-down period to focus on refactoring, technical debt, quick wins, and the
like. Periodic HACK weeks – where HACK is (since
2018) a backronym for Helpful Acts in Customer Kindness – facilitate focussed efforts on improving our ecosystem and user experiences.
I’m working in a larger immediate team than I had for most of my pre-Automattic career. I’m working alongside nine other developers, typically in groups of two to four depending on the
needs of whatever project I’m on. There’s a great deal of individual autonomy: we’re all part of a greater whole and we’re all pushing in the same direction, but outside of the
requirements of the strategic goals of our division, the team’s tactical operations are very-much devolved and consensus-driven. We work out as a team how to solve the gnarly
(and fun!) problems, how to make best use of our skills, how to share our knowledge, and how to schedule our priorities.
This team-level experience echoes the experience of being an individual at Automattic, too. The level of individual responsibility and autonomy we enjoy is similar to that I’ve seen
only after accruing a couple of years of experience and authority at most other places I’ve worked. It’s amazing to see that you can give a large group of people so much self-controlled
direction… and somehow get order out of the chaos. More than elsewhere, management is more to do with shepherding people into moving in the same direction than it is about dictating how
the ultimate strategic goals might be achieved.
Na na na na na na na na VAT MAN!
Somewhere along the way, I somehow became my team’s live-in expert on tax. You know how it is: you solve a bug with VAT calculation in
Europe… then you help roll out changes to support registration with the GST in Australia… and then one day you find yourself
reading Mexican digital services tax legislation and you can’t remember where the transition
was from being a general full-stack developer to having a specialisation in tax.
Tax isn’t a major part of my work. But it’s definitely reached a point at which I’m a go-to figure. A week or so ago when somebody had a question about the application of sales taxes to
purchases on the WooCommerce.com extensions store, their first thought was “I’ll ask Dan!” There’s
something I wouldn’t have anticipated, six month ago.
Automattic’s culture lends itself to this kind of selective micro-specialisation. The company actively encourages staff to keep learning new things but mostly without providing a specific direction, and this – along with their tendency to
attract folks who, like me, could foster an interest in almost any new topic so long as they’re learning something – means that my colleagues and I always seem to be
developing some new skill or other.
I know off the top of my head who I’d talk to about if I had a question about headless browser automation, or database index performance, or email marketing impact assessment, or queer
representation, or getting the best airline fares, or whatever else. And if I didn’t, I could probably find them. None of their job descriptions mention that aspect of their work.
They’re just the kind of people who, when they see a problem, try to deepen their understanding of it as a whole rather than just solving it for today.
A lack of pigeonholing, coupled with the kind of information management that comes out of being an entirely-distributed company, means that the specialisation of individuals becomes a
Search-Don’t-Sort problem. You don’t necessarily find an internal specialist by their job title: you’re more-likely to find them by looking for previous work on particular
topics. That feels pretty dynamic and exciting… although it does necessarily lead to occasional moments of temporary panic when you discover that something important (but short of
mission-critical) doesn’t actually have anybody directly responsible for it.
Crisis response
No examination of somebody’s first 6+ months at a new company, covering Spring 2020, would be complete without mention of that company’s response to the coronavirus crisis. Because,
let’s face it, that’s what everybody’s talking about everywhere right now.
As the UK’s lockdown (eventually) took hold I found myself
treated within my social circle like some kind of expert on remote working. My inboxes filled up with queries from friends… How do I measure output? How do I run a productive
meeting? How do I maintain morale? I tried to help, but unfortunately some of my answers relied slightly on already having a distributed culture: having the information and
resource management and teleworking infrastructure in-place before the crisis. Still, I’m optimistic that companies will come out of the other side of this situation with a
better idea about how to plan for and execute remote working strategies.
I’ve been quite impressed that even though Automattic’s all sorted for how work carries on through this crisis, we’ve gone a step further and tried to organise (remote) events for
people who might be feeling more-isolated as a result of the various lockdowns around the world. I’ve seen mention of wine tasting events, toddler groups, guided meditation sessions,
yoga clubs, and even a virtual dog park (?), all of which try to leverage the company’s existing distributed infrastructure to support employees who’re affected by the pandemic. That’s
pretty cute.
In summary: Automattic’s still proving to be an adventure, I’m still loving their quirky and chaotic culture and the opportunity to learn something new every week, and
while their response to the coronavirus crisis has been as solid as you’d expect from a fully-distributed company I’ve also been impressed by the company’s efforts to support staff (in
a huge diversity of situations across many different countries) through it.
I think our local red kites were mostly living on roadkill, and the lockdown means they’re not getting fed. They’ve also forgotten how to hunt: this afternoon I watched one get its ass
handed to it by a medium-sized crow.