Underrated meta skills when starting out as a software developer
No long winded, whimsical introduction here. The following tips are exactly what the title suggests. Also, I am horrible at introductions and this is TOTALLY not a cope out. Here go my 7 tips:
1. Break down your task into small chunks
You know that feeling when you hit "submit” on that pull request or "close” a github issue? That feeling of fulfillment and rainbows and unicorns and everything wonderful even if short lived?
That my friend, is the feeling of progress and momentum. The power of small wins should not be underestimated. The dopamine hit of checking off an item from your todo list is very real, so leverage that. Break down your tasks — no matter how big — into smaller chunks that can be completed in half a day, or at maximum, a day.
Take designing and implementing a new feature, for example. It might not always be clear from the start how you might implement all the parts. In that case, just break down the first two steps you need to complete, and write down what you would need to figure out afterwards as your step 3.
2. Create small, easy to digest pull requests
This not just ties into the feeling of quick wins and consistent progress as mentioned above, it also gets others to:
review it more quickly and easily. It's less daunting to review a pull request of 10 lines than 1000.
catch any unintended consequences earlier keeping the code quality high.
onboard quicker as less context might be required per pull request.
As others have said already: ship small diffs, how this is an unwritten rule, and how to write one.
3. Get feedback early
Make a pull request with a [WiP] or “Work in Progress” title latest after half a day of work even when your code is far from ready (which should be easy to do if you're following tip 1 already). This not only gives others in your team a sense of where you are at and offer help proactively but it’s also easier to get help when you’re making a decision/debugging. You can simply point others to the relevant line where you’re stuck and get direct inputs.
4. Do a 3 minute reflection after each task
Whenever you have completed a task, take 3 minutes to reflect these 3 questions:
"Where did I get stuck?”
Answering this will help you identify any particular concepts that you might not have understood as well as you thought you did. Could be a specific language or library feature, could be a part of the code base you're working with or could be your mental model being different from what happens in reality.
"What took longer than I expected?”
The answer to this will often lie in
you not being familiar enough with a particular tool/library feature (regex is a notorious example). If you know that you will be using this tool/feature often enough, it might be worth the time to practice some drills to sharpen your skills (10-20 minutes might already do the trick!).
you not using the correct debugging tool. Take the time to create a good setup in this case or at least jot down which tool helped you in the end so you can reuse it the next time.
you uncovering wrong assumptions about the code base. By reflecting on this, you have already updated your mental model. Congratulations!
“Was there anything new that I learnt?”
You might just have learnt something super cool or spent a long time figuring out something that feels trivial in hindsight. By reflecting on it quickly though, you solidify that knowledge in your mind and give yourself a positive boost or a virtual pat on the back for learning something that you didn't know before! Or maybe you didn't learn anything new at all, in which case you might have at least gained some more practice time at a skill.
5. Be strategic when debugging
If you’re debugging something and it’s not working after your third trial, stop and start being more strategic than just random trial and error. Write down:
“What am I trying to solve?”
“What have I learnt so far?”
"What have I tried and what didn't work?”
Then make a tree of potential solutions and cross them of as you progress. Take a second to think “how surprised will I be if this solution doesn’t work?”
This little exercise will help you not get lost in the weeds especially when you're debugging by yourself and trying to avoid falling down the rabbit hole of “Oooh! I wonder what this shiny thing does.”
6. Enable others to help you
Say you are still struggling to debug that nasty bug after what feels like hours. Or, you are supposed to be implementing a new feature but you have stumbled up on a fork in the road (or more like many, to be realistic) and find yourself going in circles delving over the pros and cons.
Fear not! Ask your unsuspecting colleague for help. But for your colleague to help you without having to jump through hoops of confusing puddles of misunderstanding, do a little bit of preparation ahead of time:
For bugs:
jot down what you're trying to solve, what you have tried so far and what's not working — which you should have already if you were following the tip above.
have the relevant parts of the code open on your workspace so you can do a quick demonstration if needed
Many a times, just doing this little exercise can make the inner light bulb glow and help you see the solution that had been hiding in plain sight.
For feature design:
sketch out a quick block diagram of the design alternatives.
jot down your assumptions of the use case explicitly. You'll be surprised how many items you can end up with a simpler design just because one of your initial assumptions turns out to be not as important as you thought it would be!
lay out the trade-offs you're thinking of
make a recommendation while making it clear that you're open to suggestions. It's good to have somewhat of an opinion going in, so your colleague doesn't feel that the responsibility of choosing is being dumped on them unfairly.
Many a times, this can save you tons of wasted hours coding up a feature geared towards the wrong requirement or help you spot a gnarly dependency/constraint earlier than you might have noticed by yourself.
Besides making you efficient, by being a little more prepared when asking for help, you show others that you respect their time and this makes it easier for them to help you again the next time. They might even look forward to it!
7. "How would I implement that?”
If you're having a hard time understanding how a particular piece of code works, take a step back and think about the functionality it enables. Now, put your developer hat on and ask yourself:
if you were to implement this functionality, how would you do it?
what type of input would you need? How would you process it? What would be the output?
what could go wrong with this way of doing it?
Then revisit the piece of code and try to map the components you see to your mental model. Many times, they might even match 1 to 1. The other times, you might figure out either a different way of solving the same problem or learn about the corner cases this piece of code takes care of, that your imagined one might not have.
Either way, you have an easier time understanding the code and you learnt something new!
8. Bonus: Review others’ code
Especially when starting out in a new team, you might undervalue your opinion, and think that because you might not have all the context, you'd make for a bad reviewer. But don't!
Review others’ code as much as you can. Reading others’ code is a fantastic way for you to get started in a new code base, learn more about the unwritten guidelines, think through the implication of the code, and share your fresh view point that others might seriously benefit from too!
Explicitly ask a more experienced dev to review it as well, or even better, pair-review it together. This will not only make you feel more comfortable, but you also get to learn some points you might have missed in your review.
Conclusion
Not all of software development is logic puzzles and solving technical problems that you either just have the capability for or don't. By being a bit more proactive, bit more reflective, you can go a long way in making yourself effective. This way, you also retain much more of what you learn as many of these tips involve some form of active retrieval.
The sense of consistent progress generates an addictive momentum that's hard to get away from too! And finally, you accelerate your learning, problem solving skills and productivity as you put these tips in practice.
You don't have to do it all from the get-go, of course. The advice I often give is: pick out one of these tips, try to apply it in your day-to-day work and see how it works for you. Once it has become a habit, revisit these tips and pick up a new one.
Did any of this help you at all? Do you have some tips of your own you would like to share? Let me know in the comments below, or reach out to me on Twitter where I first wrote a part of this thread.
What I read in 2020
2020 was the year that rekindled my love for science fiction after a long hiatus.
2020 was the year that rekindled my love for science fiction after a long hiatus.
Beginning is everything
First off, I started the year with two fantastic books — QualityLand and Exhalation.
QualityLand
QualityLand by Marc-Uwe Kling (originally written in German) is part satire, part speculative fiction, where the future evokes an eery feeling of being just a step away from our present reality.
On one hand, the book paints a future that is grim, reminiscent of "Black Mirror" themes. A dating service that not just matches you with a partner based on your social rating and an algorithmically determined compatibility score, but even “upgrades” your partner and arranges for secret dates when a new, “more desirable” partner becomes available. There are drones that deliver you items that you hadn't realised you wanted (Ronny Chieng's wishes coming true?). A humanoid robot is in the running for president while his opposition regales the public with "alternative facts". And of course, internet trolling is now a full-time profession.
On the other hand, the book also refuses to take itself too seriously, rooting its plot in a pink dolphin vibrator. This single delivery by the nearly omniscient drone delivery service is pivotal to our initially purposeless protagonist who feels absolutely certain there has been a mistake, while the company insists otherwise.
This combination of an almost-too-near, too uncomfortable-because-plausible future with humour that's serious and silly at the same time (especially highlighted by the multiple adverts from the future the book is lined by) makes the book a potent mix whether you're looking for a dose of existential crisis with your morning coffee or a way to start a conversation over dinner without coming across as preachy.
Exhalation
Exhalation by Ted Chiang is a short story collection — one that reads more like a series of thought experiments written exquisitely and thoroughly grounded in humanity, even when the story itself is set in a universe that's as alien as imaginable:
In ancient Baghdad, a fabric seller tries to grapple with past mistakes as he comes across a portal that might allow him second chances at his past.
An alien scientist in a faraway world dissects his own brain to understand what makes him, him.
What happens when there is no free will? Would life lose all meaning?
What are the ramifications for the society of becoming literate? What happens when you lose this skill?
“Science Fiction offers a way to talk about things, issues that are relevant to us, metaphorically.”
You can read more about the author's thought process here.
Pandemic hits
Alongside the strong beginning, one other thing happened in 2020 that made me seek refuge in science fiction — the event that shall not be named a.k.a "The Pandemic”a.k.a "COVID-19 crisis" (so much for the not naming).
Given that I spent much of the time (like most people), working from home, through extended lock-down(s) with many combinations of restrictions — science fiction became a refuge through not just mere escapism but also as a portal to new discoveries full of imaginary landscapes.
I have never truly enjoyed space operas — a series of events describing "how a hero saves the world” — as much as stories that make you wonder “what happens if…" , or even better, “assume these conditions to be true, what happens then?".
Asimov's work has been a long time favourite of mine in this regard, while I have had mixed reactions to Clarke's. In 2020, I finally got my first taste of Heinlein with "The Moon is a Harsh Mistress" thereby completing my essential reading of the "big three” of science fiction.
My personal "big three", though has to be the trio of Ted Chiang, Liu Cixin, and Ken Liu. While I first got acquainted with all three authors in 2018 (a post for another time, perhaps), 2020 deepened my enjoyment and sense of wonder when reading their works.
Ted Chiang doesn't write much — but when he does, both the ideas and their execution are a thing of beauty. You might also know of the author through the movie “Arrival", which is based on a novella of his ("Story of Your Life"). Liu Cixin is of the "Three Body Problem” fame, with Ken Liu as the translator for the first and third book of the trilogy. However, Ken Liu is also an author in his own right, and I have immensely enjoyed both his own short story collections, as well as the ones he has curated and translated from Chinese.
“I don’t get a lot of ideas that interest me enough to write about them. Writing is hard for me and I have to be really interested in something in order to put myself through it.”
And with that, I'm wrapping up my first proper post on this blog here. My full reading list from 2020 follows:
The 2020 Reading List
Format: title by author - rating - month read
HBR Guide to Delivering Feedback - 3/5 - Jan
QualityLand by Marc-Uwe Kling - 5/5 - Jan
Exhalation by Ted Chiang - 5/5 - Jan
Ball Lightning by Liu Cixin - 3/5 - Mar
The Left Hand of Darkness by Ursula Le Guin - 3/5 - May
The Martian by Andy Weir - 3/5 - May
Broken Stars by Ken Liu - 4/5 - Jun
The Hidden Girl and Other Stories by Ken Liu - 4/5 - Jul
Of Ants and Dinosaurs by Liu Cixin- 3/5 - Jul
The Moon is a Harsh Mistress by Robert A. Heinlein - 4/5 - Jul
Recursion by Blake Crouch - 3/5 - Jul
Factfulness by Hans Rosling - 3/5 - Jul
Slender Man - 2/5 - Aug
Annihilation by Jeff Vandermeer - 3/5 - Aug
The Wandering Earth by Liu Cixin - 4/5 - Aug
Avatar (series) - 3/5 - Aug
Dark Matter by Blake Crouch - 3/5 - Sep
Into the Woods by John Yorke - 3/5 - Sep
Snow Crash by Neal Stephenson - 4/5 - Oct
Ready Player One by Ernest Cline - 3/5 - Oct
The Peripheral by William Gibson - 4/5 - Nov
Hyperion by Dan Simmons - 4/5 - Dec
Hello World
And so, the blog begins.
At 15:19 CET, on a day with perpetually changing weather — from sunny to windy to rainy in less than a minute — here comes the first post.
So.
Hello World.