Good Advice covering the basic parts of system design. I think I’m moving from the “coding” part of my career into the “system” part of my career, which means I’m going to have to think about these things more and more.
What is system design? In my view, if software design is how you assemble lines of code, system design is how you assemble services. The primitives of software design are variables, functions, classes, and so on. The primitives of system design are app servers, databases, caches, queues, event buses, proxies, and so on.
A lot of the advice seems to boil down to: measure your application and fix the slow parts in the most predictable way you can. Avoid cleverness. Like I said, Good Advice.
There’s been a bit of a trend recently for some companies to move to 4-day workweeks. This is making a decent amount of noise, but the actual number of companies offering this still seems pretty few and far between. It’s not hard to imagine why CEOs might feel this is risky, considering that many don’t even trust people to work from home.
What I don’t see much are 80% jobs, which are 32 hour jobs, at 80% salary. This should be a low-risk proposition that I think a lot of people in the tech industry would take, if it were an option.
I think this idea works pretty well for software development jobs, which have notoriously inflated salaries and where short bursts of Deep Work (tm) are punctuated by meetings and context switching. I think a lot of devs would take a 20% pay cut for an extra day on the weekend. Apparently many workers in the Netherlands already do!
The main downside from the discussion on Hacker News seems to centre on the extra bandwidth that would come from managing 80% employees: fixed per-employee costs like hardware and HR, extra burden on the recruitment team. One enterprising poster complained that the outcome of this kind of thinking is the hourly commoditisation of jobs—everyone working on hourly salaries—and that this would lead to some catastrophic impact on software developers' quality of life, but:
some software developers, like those who work for themselves, already work like this, contracting to clients on an hourly basis, and
there's a world of difference between someone in the service industry making minimum (or slightly-above-minimum) wage, and someone in tech making their salary-equivalent.
Man, there is some money in tech. Developers command six-figure salaries and tech companies trade in hundreds of millions of dollars annually. Fortunes change, stars rise and fall (but mostly rise). And the brightest of those stars attend conferences.
Conferences are events of opportunity where the future of software development is written and where the scions of industry trade follows on Twitter. Web & design conferences—like UX London, coming up in June—are the events that I aspire to, if I intend to make a change in the world, however small, using only my computer and an enterprising spirit (& I do).
They're also very difficult for regular schlubs like me to attend, because of how expensive they are. UX London costs £1,250 for the 3-day event, which is like 4% of my take-home annual salary!
I'm led to believe that a lot of software developers don't buy their own tickets; instead, their employer pays for their attendance, and the developer returns with some bleeding-edge wisdom to practice during the nine-to-five. And the employers that have this sort of disposable income aren't, usually, SME agencies in the North East (which is probably why headline tech conferences are rarely held in Newcastle). The result is that conferences, from the outside, feel like a bit of a tech elite club, accessible to the masses only through videos posted after the event if the organisers are generous.
But maybe the high prices are totally justified: maybe margins here are razor thin and the value that canny conference attendees can extract from from a room of so many sharp minds is worth all of Smaug's bullion & then some. Maybe I'm just not good at turning my network into opportunity. (Maybe I don't have a network because I've only been to like 3 conferences!)
And at any rate, this whole "conferences are expensive" complaint isn't a universal rule: there are affordable conferences. Like Hey!'s All Day Hey in Leeds at the beginning of May, which was only ~£50 if you got an early bird ticket.
Let me finish by saying that I know that excluding folks isn't the intention here. I know for a fact that UX London is organised by smart, thoughtful people, assembling a diverse and talented pool of speakers, building a more inclusive, robust internet for all of us. I don't mean to single UX London out—they're just the most high-profile conference on my radar at the moment.
Always thinking about effectiveness. Maybe the overriding concern of the past 3-4 years or so. Since getting into tech.
Men in tech seem extremely preoccupied by effectiveness. Extension of the general self-help/productivity schtick that we're all obsessed with. Really fine-tuning our whole selves to squeeze the last drop of experience out of everything that we do. Wringing ourselves dry and then blogging about it for clout.
Blog posts about not taking a day off. Blog posts about forgiving yourself for taking a day off. Blog posts about how taking a day off makes you more effective on the days that you don't take off. Blog posts on recognising burnout; blog posts on not being able to recognise burnout; blog posts about how good I feel now that I'm not burned out anymore.
I lie on the sofa and my ineffectiveness eats me alive, piecemeal, squirming the whole way down.
I'm at my best when I'm up on my feet, or in my office chair, working on a problem. At my best with a tech-related podcast in my ears. At my best when I'm moving forward. Stopping fills me with dread and anxiety. Stopping gnaws at me until I'm all raw and ragged around the edges like a well-loved dog toy. (Except for dogs are lovely and anxiety is decidedly unlovely.)
When I think of how old I am, something drops out in the pit of my stomach and my guts make sounds that the ear wasn't designed to hear. When I think of how old I am and how much of my useable effective lifespan I spent not doing the thing that I spend all of my time doing, now. When I think of how old I am and how much of my useable effective lifespan I spent not turning my time into spendable assets, into things that can be enjoyed and taken pictures of and reminsced about with rosy spectacles. Yuck.
When I think of the things I haven't done; the things I don't do; the things I won't do. The things I might do, if I don't have the discipline. The things I'm doing right now, that I oughtn't be; the things I ought to be doing but can't find the wherewithal to just get up and.
The difficult decisions to make, the hard conversations to have, the things I want people to understand but can't find the words or the phrases to express myself.
When I think of the Twitter posts reading "Good work gets noticed." When I think of the thought leaders writing, "Put in the graft; talent will be recognised." The thought leaders writing, "Never finish the week with nothing." When I feel that I'm firing on all cylinders and no one knows who I am. When, maybe I only have 2 cylinders in the first place, so all the hard graft in the world isn’t going to get me off the line faster than a mid-range Volkswagen. When I think that I just don't have the displacement to generate the engagement.
When I think of the navel-gazey stuff I've written, the self-pity I've felt, re-reading paragraphs I've just written. When I write weird prepositional phrases joined end-to-end with full stops: no intervening, anchoring phrasal roots. When I finish a sentence with the word and. When my navel-gazing folds back in on itself as a meta-commentary on the Navel Gaze. When I don't know what I'm trying to say.
Encouraging
Writing like this is encouraging. On my worse days I feel like I've become some sort of weird human content mill, just trying to churn out usefulness without any real concern for consumption. The couple of blog posts I've done about job listings at big companies (well, bigger than I've ever worked for) don't feel useful to me, except for as an exercise in writing about dry content. I don't think there's much to take from those posts that couldn't be presented in a bulletted list, probably with fewer than or equal to 4 bullets. Maybe that's why open job posting have so many bulletted lists.
But that's the process of learning and growing, isn't it? Doing something awful, and then coming back to it some time later with revulsion, and doing something marginally better, sometime later. That's how I get better at things:
Do the thing that I wish someone else had done for me
Realise that the thing didn't solve the original problem, or it did solve the problem but not it in the way I wanted it to
Repeat
It's worked out pretty well for me so far.
It feels like part of getting better at writing code is talking less and less about code. I'm not sure yet what I'm meant to talk about instead. But I can see the trend. First an engineer talks about functions, and then they talk about programming paradigms, and then they start talking about approaches or something. By the time that engineer is a senior at Stripe, who knows what they talk about?
All to say that things trend upward here. I think the concern is less that things aren't trending upward so much as that they're not trending upward fast enough. Am I going to die before I'm good enough to work at Stripe? Am I going to run out of useable brainpower? Am I going to become stuck in my ways and gradually wind down? Should I be investing more seriously in Go? How do I market myself better? How do I attract an audience to my work?
Certainly not with blog posts like this.
I walked across a moor with Sam
Halfway up the moor we climbed over a stile, and I stopped to relieve myself in a furrow on the other side of the wall. Sam continued on a little ways; a mother sheep and her lambs scurried out of the way. I continued on after her.
A light wind picked up, jostling the trees in a plantation nearby, jostling the tall strawish grasses that grow in boggy moorland. Sheep bleated off in the distance. The wind blew. A far-off tractor groaned through a town down below us somewhere. It was very quiet.
Sam walking through open moorland above Westgate in Weardale.
Javascript is constantly evolving. Not only is the language evolving, a new standard published each year, but the ecosystem is as well. This isn't a controversial opinion: Javascript fatigue is real. This is the blog post that most people will refer to as evidence, and this is a screenshot of 64 million results for Javascript fatigue.
64 million results for 'JavaScript fatigue', and also a nice time capsule of what Google looked like
I write Javascript each and every day. Once a month, someone likes my Javascript so much that they give me a bunch of money so I can keep doing it. I'm not a Javascript savant by any measure (my Twitter follower count will tell you this), but I'm well-acquainted enough to feel the way the tides of JS move.
And like any swimmer worth the salt in this particular sea, I'm compelled to write what I think about it.
The good
When it comes to framing the Churn of Javascript as a Good Thing, the takes are remarkably sparse. As a middleweight developer, I'm not the hugest fan of the Churn myself. But it does have its good facets. Let me explain.
For starters, the barrier to entry for Javascript developers is extremely low. On any day in the cursed year 2020, the only thing standing between a technically-minded 17-year-old and a command of the Internet is maybe like 70-90 hours of access to free Javascript learning resources. These are plentiful enough that I'm not even going to try and list them here.
In fact, the Javascript education industry seems larger, sometimes, than the industry of developers actually building Javascript-based products. This can't be true, and certainly isn't true: but it feels true. There's no lack of schmucks with MacBook Pros and an entry-level microphone recording tutorials and writing blog posts introducing you to everything from Axios to Zeit (I mean Vercel).
The practical effect of this is that if you're legally allowed to work somewhere, you could feasibly have a Javascript job in like 2.5 months. Which is a super-short runway for skilled labour, it feels like.
But that's not all! The Churn also means that very few popular technologies have been around for any significant time. There's not much in the way of an old guard. Almost 50% of Javascript developers have less than 5 years of experience. No one really knows what they're doing, which means that there's tremendous opportunity for experimentation and exploration. The Javascript pie is always growing.
The bad
Usually, the Churn is framed as a Very Bad Thing. In fact, this is such an accepted point of view that it's become a meme.
The Churn tends to express itself in individual developers as JS fatigue: a combination of burnout and FOMO happening at the same time. Keeping up with the Javascript ecosystem is a marathon, because it never quite ends; but at the same time it’s a sprint, because it’s all happening so quickly. There’s just no way to keep up with it.
And so we’re seeing the modern web fragment, day to day. I’m talking Coyier’s The Great Divide . I’m talking React devs v. Vue devs. I’m talking function programming advocates in JS-adjacent languages like Reason and Elm , v. OO die-hards clinging to ES6 classes and classical design patterns. I’m talking job postings that ask for competent with Javascript (jQuery, lodash) v. job postings that ask for competent with Javascript (React, Redux). There is no one Javascript Developer anymore: there are React developers and build engineers and Node jockeys and 35-year-olds calling $el.fadeOut() and 22-year-olds shitting on 35-year-olds who call $el.fadeOut(). It’s a mess.
And all of these different developers are writing Medium posts and Dev.to posts and personal blog posts about why their chosen stack is the Right One, which means that no matter what you think you’ll always find someone telling you You’re Right; but which also means that no matter what you think you’ll always find someone telling you that You’re Wrong: which can be confusing in the extreme if you are part of the half of all developers who have been doing this for less than 5 years.
Which, I think that’s what makes the Churn so painful: it’s not that there are lots of options, but that the discussion around those options is exhausting.
The confusing
The result here is that no matter how talented you are, there’s always some new library or framework or paradigm being hawked by educators whose job it is to maintain this anxiety in FOMO-ridden devs sitting in their living rooms, and who always seem to have the most Twitter followers and the most dev clout and the shiniest new MacBooks and the coolest LED-ridden battlestations to aspire to if only you, too, could learn how to write Javascript in this new way: which you can, for $79 if you act within the next 36 hours.
But like, that paragraph right there: that is what Javascript fatigue looks like. Being a little tuckered-out and confused by the way things are moving and the people moving those ways. Let me make this clear: I don't think Javascript educators are not the problem. In fact, everything good I've said above, about how Javascript has such a low-friction uptake, is facilitated by these educators. There's a group of people in the Javascript world that are doing very real, very important work. Wes Bos is the closest thing to a mentor I’ve ever had, in tech or otherwise. Shawn Wang’s perspectives on coding as a career are legitimately indispensable. They, along with a handful of others, generate a powerful and effective signalin the JS community.
But the multiplicity of Javascript Thought Leaders makes it very difficult to separate that good signal from the rest of the bullshit noise, of which there is an absolute ton—including this blog, I worry. It makes it ten times worse when you’re new to the language and you’ve never heard the signal before.
And this promotes burnout and instills in developers the sense that they will never been good enough. That there is always something new and exciting to learn around the corner. I have a real, gnawing sense of dread that if don’t keep moving, my skillset will, over 3 or 4 months, become obsolete: and I will become unemployable.
This is the source of a great deal of Developer Anxiety. At the same time, it keeps the crankshaft of the great JavaScript Tutorial Engine turning, and generates a significant profit for a class made up at once of of ultra-proficient educators and hypebeasts masquerading as ultra-proficient educators, and from where I'm standing it's very hard to tell the difference, and so: turn onwards it does.
In concl
So what does this all look like in practice? It looks like an industry of developers with a tonof breadth and almost no depth. It looks devs with 5 years under their belt: 1 year of experience in 5 different technologies. It looks like a bunch of burnt-out white dudes with beards wandering around trying to see around corners; or maybe a few slightly-less burnt-out white dudes building up corners for everyone else to try and see around.
