Full Width 
Shortcuts 
Sign Up 
Log In 
Articles 
Reading List More from Computer Things
Recently I got a question on formal methods1: how does it help to mathematically model systems when the system requirements are constantly changing? It doesn't make sense to spend a lot of time proving a design works, and then deliver the product and find out it's not at all what the client needs. As the saying goes, the hard part is "building the right thing", not "building the thing right". One possible response: "why write tests"? You shouldn't write tests, especially lots of unit tests ahead of time, if you might just throw them all away when the requirements change. This is a bad response because we all know the difference between writing tests and formal methods: testing is easy and FM is hard. Testing requires low cost for moderate correctness, FM requires high(ish) cost for high correctness. And when requirements are constantly changing, "high(ish) cost" isn't affordable and "high correctness" isn't worthwhile, because a kinda-okay solution that solves a customer's problem is infinitely better than a solid solution that doesn't. But eventually you get something that solves the problem, and what then? Most of us don't work for Google, we can't axe features and products on a whim. If the client is happy with your solution, you are expected to support it. It should work when your customers run into new edge cases, or migrate all their computers to the next OS version, or expand into a market with shoddy internet. It should work when 10x as many customers are using 10x as many features. It should work when you add new features that come into conflict. And just as importantly, it should never stop solving their problem. Canonical example: your feature involves processing requested tasks synchronously. At scale, this doesn't work, so to improve latency you make it asynchronous. Now it's eventually consistent, but your customers were depending on it being always consistent. Now it no longer does what they need, and has stopped solving their problems. Every successful requirement met spawns a new requirement: "keep this working". That requirement is permanent, or close enough to decide our long-term strategy. It takes active investment to keep a feature behaving the same as the world around it changes. (Is this all a pretentious of way of saying "software maintenance is hard?" Maybe!) Phase changes In physics there's a concept of a phase transition. To raise the temperature of a gram of liquid water by 1° C, you have to add 4.184 joules of energy.2 This continues until you raise it to 100°C, then it stops. After you've added two thousand joules to that gram, it suddenly turns into steam. The energy of the system changes continuously but the form, or phase, changes discretely. Software isn't physics but the idea works as a metaphor. A certain architecture handles a certain level of load, and past that you need a new architecture. Or a bunch of similar features are independently hardcoded until the system becomes too messy to understand, you remodel the internals into something unified and extendable. etc etc etc. It's doesn't have to be totally discrete phase transition, but there's definitely a "before" and "after" in the system form. Phase changes tend to lead to more intricacy/complexity in the system, meaning it's likely that a phase change will introduce new bugs into existing behaviors. Take the synchronous vs asynchronous case. A very simple toy model of synchronous updates would be Set(key, val), which updates data[key] to val.3 A model of asynchronous updates would be AsyncSet(key, val, priority) adds a (key, val, priority, server_time()) tuple to a tasks set, and then another process asynchronously pulls a tuple (ordered by highest priority, then earliest time) and calls Set(key, val). Here are some properties the client may need preserved as a requirement: If AsyncSet(key, val, _, _) is called, then eventually db[key] = val (possibly violated if higher-priority tasks keep coming in) If someone calls AsyncSet(key1, val1, low) and then AsyncSet(key2, val2, low), they should see the first update and then the second (linearizability, possibly violated if the requests go to different servers with different clock times) If someone calls AsyncSet(key, val, _) and immediately reads db[key] they should get val (obviously violated, though the client may accept a slightly weaker property) If the new system doesn't satisfy an existing customer requirement, it's prudent to fix the bug before releasing the new system. The customer doesn't notice or care that your system underwent a phase change. They'll just see that one day your product solves their problems, and the next day it suddenly doesn't. This is one of the most common applications of formal methods. Both of those systems, and every one of those properties, is formally specifiable in a specification language. We can then automatically check that the new system satisfies the existing properties, and from there do things like automatically generate test suites. This does take a lot of work, so if your requirements are constantly changing, FM may not be worth the investment. But eventually requirements stop changing, and then you're stuck with them forever. That's where models shine. As always, I'm using formal methods to mean the subdiscipline of formal specification of designs, leaving out the formal verification of code. Mostly because "formal specification" is really awkward to say. ↩ Also called a "calorie". The US "dietary Calorie" is actually a kilocalorie. ↩ This is all directly translatable to a TLA+ specification, I'm just describing it in English to avoid paying the syntax tax ↩
I have a lot in the works for the this month's Logic for Programmers release. Among other things, I'm completely rewriting the chapter on Logic Programming Languages. I originally showcased the paradigm with puzzle solvers, like eight queens or four-coloring. Lots of other demos do this too! It takes creativity and insight for humans to solve them, so a program doing it feels magical. But I'm trying to write a book about practical techniques and I want everything I talk about to be useful. So in v0.9 I'll be replacing these examples with a couple of new programs that might get people thinking that Prolog could help them in their day-to-day work. On the other hand, for a newsletter, showcasing a puzzle solver is pretty cool. And recently I stumbled into this post by my friend Pablo Meier, where he solves a videogame puzzle with Prolog:1 Summary for the text-only readers: We have a test with 10 true/false questions (denoted a/b) and four student attempts. Given the scores of the first three students, we have to figure out the fourth student's score. bbababbabb = 7 baaababaaa = 5 baaabbbaba = 3 bbaaabbaaa = ??? You can see Pablo's solution here, and try it in SWI-prolog here. Pretty cool! But after way too long studying Prolog just to write this dang book chapter, I wanted to see if I could do it more elegantly than him. Code and puzzle spoilers to follow. (Normally here's where I'd link to a gentler introduction I wrote but I think this is my first time writing about Prolog online? Uh here's a Picat intro instead) The Program You can try this all online at SWISH or just jump to my final version here. :- use_module(library(dif)). % Sound inequality :- use_module(library(clpfd)). % Finite domain constraints First some imports. dif lets us write dif(A, B), which is true if A and B are not equal. clpfd lets us write A #= B + 1 to say "A is 1 more than B".2 We'll say both the student submission and the key will be lists, where each value is a or b. In Prolog, lowercase identifiers are atoms (like symbols in other languages) and identifiers that start with a capital are variables. Prolog finds values for variables that match equations (unification). The pattern matching is real real good. % ?- means query ?- L = [a,B,c], [Y|X] = [1,2|L], B + 1 #= 7. B = 6, L = [a, 6, c], X = [2, a, 6, c], Y = 1 Next, we define score/33 recursively. % The student's test score % score(student answers, answer key, score) score([], [], 0). score([A|As], [A|Ks], N) :- N #= M + 1, score(As, Ks, M). score([A|As], [K|Ks], N) :- dif(A, K), score(As, Ks, N). First key is the student's answers, second is the answer key, third is the final score. The base case is the empty test, which has score 0. Otherwise, we take the head values of each list and compare them. If they're the same, we add one to the score, otherwise we keep the same score. Notice we couldn't write if x then y else z, we instead used pattern matching to effectively express (x && y) || (!x && z). Prolog does have a conditional operator, but it prevents backtracking so what's the point??? A quick break about bidirectionality One of the coolest things about Prolog: all purely logical predicates are bidirectional. We can use score to check if our expected score is correct: ?- score([a, b, b], [b, b, b], 2). true But we can also give it answers and a key and ask it for the score: ?- score([a, b, b], [b, b, b], X). X = 2 Or we could give it a key and a score and ask "what test answers would have this score?" ?- score(X, [b, b, b], 2). X = [b, b, _A], dif(_A,b) X = [b, _A, b], dif(_A,b) X = [_A, b, b], dif(_A,b) The different value is written _A because we never told Prolog that the array can only contain a and b. We'll fix this later. Okay back to the program Now that we have a way of computing scores, we want to find a possible answer key that matches all of our observations, ie gives everybody the correct scores. key(Key) :- % Figure it out score([b, b, a, b, a, b, b, a, b, b], Key, 7), score([b, a, a, a, b, a, b, a, a, a], Key, 5), score([b, a, a, a, b, b, b, a, b, a], Key, 3). So far we haven't explicitly said that the Key length matches the student answer lengths. This is implicitly verified by score (both lists need to be empty at the same time) but it's a good idea to explicitly add length(Key, 10) as a clause of key/1. We should also explicitly say that every element of Key is either a or b.4 Now we could write a second predicate saying Key had the right 'type': keytype([]). keytype([K|Ks]) :- member(K, [a, b]), keytype(Ks). But "generating lists that match a constraint" is a thing that comes up often enough that we don't want to write a separate predicate for each constraint! So after some digging, I found a more elegant solution: maplist. Let L=[l1, l2]. Then maplist(p, L) is equivalent to the clause p(l1), p(l2). It also accepts partial predicates: maplist(p(x), L) is equivalent to p(x, l1), p(x, l2). So we could write5 contains(L, X) :- member(X, L). key(Key) :- length(Key, 10), maplist(contains([a,b]), L), % the score stuff Now, let's query for the Key: ?- key(Key) Key = [a, b, a, b, a, a, b, a, a, b] Key = [b, b, a, b, a, a, a, a, a, b] Key = [b, b, a, b, a, a, b, b, a, b] Key = [b, b, b, b, a, a, b, a, a, b] So there are actually four different keys that all explain our data. Does this mean the puzzle is broken and has multiple different answers? Nope The puzzle wasn't to find out what the answer key was, the point was to find the fourth student's score. And if we query for it, we see all four solutions give him the same score: ?- key(Key), score([b, b, a, a, a, b, b, a, a, a], Key, X). X = 6 X = 6 X = 6 X = 6 Huh! I really like it when puzzles look like they're broken, but every "alternate" solution still gives the same puzzle answer. Total program length: 15 lines of code, compared to the original's 80 lines. Suck it, Pablo. (Incidentally, you can get all of the answer at once by writing findall(X, (key(Key), score($answer-array, Key, X)), L).) I still don't like puzzles for teaching The actual examples I'm using in the book are "analyzing a version control commit graph" and "planning a sequence of infrastructure changes", which are somewhat more likely to occur at work than needing to solve a puzzle. You'll see them in the next release! I found it because he wrote Gamer Games for Lite Gamers as a response to my Gamer Games for Non-Gamers. ↩ These are better versions of the core Prolog expressions \+ (A = B) and A is B + 1, because they can defer unification. ↩ Prolog-descendants have a convention of writing the arity of the function after its name, so score/3 means "score has three parameters". I think they do this because you can overload predicates with multiple different arities. Also Joe Armstrong used Prolog for prototyping, so Erlang and Elixir follow the same convention. ↩ It still gets the right answers without this type restriction, but I had no idea it did until I checked for myself. Probably better not to rely on this! ↩ We could make this even more compact by using a lambda function. First import module yall, then write maplist([X]>>member(X, [a,b]), Key). But (1) it's not a shorter program because you replace the extra definition with an extra module import, and (2) yall is SWI-Prolog specific and not an ISO-standard prolog module. Using contains is more portable. ↩
My April Cools is out! Gaming Games for Non-Gamers is a 3,000 word essay on video games worth playing if you've never enjoyed a video game before. Patreon notes here. (April Cools is a project where we write genuine content on non-normal topics. You can see all the other April Cools posted so far here. There's still time to submit your own!) April Cools' Club
Logic for Programmers v0.8 now out! The new release has minor changes: new formatting for notes and a better introduction to predicates. I would have rolled it all into v0.9 next month but I like the monthly cadence. Get it here! Betteridge's Law of Software Engineering Specialness In There is No Automatic Reset in Engineering, Tim Ottinger asks: Do the other people have to live with January 2013 for the rest of their lives? Or is it only engineering that has to deal with every dirty hack since the beginning of the organization? Betteridge's Law of Headlines says that if a journalism headline ends with a question mark, the answer is probably "no". I propose a similar law relating to software engineering specialness:1 If someone asks if some aspect of software development is truly unique to just software development, the answer is probably "no". Take the idea that "in software, hacks are forever." My favorite example of this comes from a different profession. The Dewey Decimal System hierarchically categorizes books by discipline. For example, Covered Bridges of Pennsylvania has Dewey number 624.37. 6-- is the technology discipline, 62- is engineering, 624 is civil engineering, and 624.3 is "special types of bridges". I have no idea what the last 0.07 means, but you get the picture. Now if you look at the 6-- "technology" breakdown, you'll see that there's no "software" subdiscipline. This is because when Dewey preallocated the whole technology block in 1876. New topics were instead to be added to the 00- "general-knowledge" catch-all. Eventually 005 was assigned to "software development", meaning The C Programming Language lives at 005.133. Incidentally, another late addition to the general knowledge block is 001.9: "controversial knowledge". And that's why my hometown library shelved the C++ books right next to The Mothman Prophecies. How's that for technical debt? If anything, fixing hacks in software is significantly easier than in other fields. This came up when I was interviewing classic engineers. Kludges happened all the time, but "refactoring" them out is expensive. Need to house a machine that's just two inches taller than the room? Guess what, you're cutting a hole in the ceiling. (Even if we restrict the question to other departments in a software company, we can find kludges that are horrible to undo. I once worked for a company which landed an early contract by adding a bespoke support agreement for that one customer. That plagued them for years afterward.) That's not to say that there aren't things that are different about software vs other fields!2 But I think that most of the time, when we say "software development is the only profession that deals with XYZ", it's only because we're ignorant of how those other professions work. Short newsletter because I'm way behind on writing my April Cools. If you're interested in April Cools, you should try it out! I make it way harder on myself than it actually needs to be— everybody else who participates finds it pretty chill. Ottinger caveats it with "engineering, software or otherwise", so I think he knows that other branches of engineering, at least, have kludges. ↩ The "software is different" idea that I'm most sympathetic to is that in software, the tools we use and the products we create are made from the same material. That's unusual at least in classic engineering. Then again, plenty of machinists have made their own lathes and mills! ↩
More in programming
I like the job title “Design Engineer”. When required to label myself, I feel partial to that term (I should, I’ve written about it enough). Lately I’ve felt like the term is becoming more mainstream which, don’t get me wrong, is a good thing. I appreciate the diversification of job titles, especially ones that look to stand in the middle between two binaries. But — and I admit this is a me issue — once a title starts becoming mainstream, I want to use it less and less. I was never totally sure why I felt this way. Shouldn’t I be happy a title I prefer is gaining acceptance and understanding? Do I just want to rebel against being labeled? Why do I feel this way? These were the thoughts simmering in the back of my head when I came across an interview with the comedian Brian Regan where he talks about his own penchant for not wanting to be easily defined: I’ve tried over the years to write away from how people are starting to define me. As soon as I start feeling like people are saying “this is what you do” then I would be like “Alright, I don't want to be just that. I want to be more interesting. I want to have more perspectives.” [For example] I used to crouch around on stage all the time and people would go “Oh, he’s the guy who crouches around back and forth.” And I’m like, “I’ll show them, I will stand erect! Now what are you going to say?” And then they would go “You’re the guy who always feels stupid.” So I started [doing other things]. He continues, wondering aloud whether this aversion to not being easily defined has actually hurt his career in terms of commercial growth: I never wanted to be something you could easily define. I think, in some ways, that it’s held me back. I have a nice following, but I’m not huge. There are people who are huge, who are great, and deserve to be huge. I’ve never had that and sometimes I wonder, ”Well maybe it’s because I purposely don’t want to be a particular thing you can advertise or push.” That struck a chord with me. It puts into words my current feelings towards the job title “Design Engineer” — or any job title for that matter. Seven or so years ago, I would’ve enthusiastically said, “I’m a Design Engineer!” To which many folks would’ve said, “What’s that?” But today I hesitate. If I say “I’m a Design Engineer” there are less follow up questions. Now-a-days that title elicits less questions and more (presumed) certainty. I think I enjoy a title that elicits a “What’s that?” response, which allows me to explain myself in more than two or three words, without being put in a box. But once a title becomes mainstream, once people begin to assume they know what it means, I don’t like it anymore (speaking for myself, personally). As Brian says, I like to be difficult to define. I want to have more perspectives. I like a title that befuddles, that doesn’t provide a presumed sense of certainty about who I am and what I do. And I get it, that runs counter to the very purpose of a job title which is why I don’t think it’s good for your career to have the attitude I do, lol. I think my own career evolution has gone something like what Brian describes: Them: “Oh you’re a Designer? So you make mock-ups in Photoshop and somebody else implements them.” Me: “I’ll show them, I’ll implement them myself! Now what are you gonna do?” Them: “Oh, so you’re a Design Engineer? You design and build user interfaces on the front-end.” Me: “I’ll show them, I’ll write a Node server and setup a database that powers my designs and interactions on the front-end. Now what are they gonna do?” Them: “Oh, well, we I’m not sure we have a term for that yet, maybe Full-stack Design Engineer?” Me: “Oh yeah? I’ll frame up a user problem, interface with stakeholders, explore the solution space with static designs and prototypes, implement a high-fidelity solution, and then be involved in testing, measuring, and refining said solution. What are you gonna call that?” [As you can see, I have some personal issues I need to work through…] As Brian says, I want to be more interesting. I want to have more perspectives. I want to be something that’s not so easily definable, something you can’t sum up in two or three words. I’ve felt this tension my whole career making stuff for the web. I think it has led me to work on smaller teams where boundaries are much more permeable and crossing them is encouraged rather than discouraged. All that said, I get it. I get why titles are useful in certain contexts (corporate hierarchies, recruiting, etc.) where you’re trying to take something as complicated and nuanced as an individual human beings and reduce them to labels that can be categorized in a database. I find myself avoiding those contexts where so much emphasis is placed in the usefulness of those labels. “I’ve never wanted to be something you could easily define” stands at odds with the corporate attitude of, “Here’s the job req. for the role (i.e. cog) we’re looking for.” Email · Mastodon · Bluesky
We received over 2,200 applications for our just-closed junior programmer opening, and now we're going through all of them by hand and by human. No AI screening here. It's a lot of work, but we have a great team who take the work seriously, so in a few weeks, we'll be able to invite a group of finalists to the next phase. This highlights the folly of thinking that what it'll take to land a job like this is some specific list of criteria, though. Yes, you have to present a baseline of relevant markers to even get into consideration, like a great cover letter that doesn't smell like AI slop, promising projects or work experience or educational background, etc. But to actually get the job, you have to be the best of the ones who've applied! It sounds self-evident, maybe, but I see questions time and again about it, so it must not be. Almost every job opening is grading applicants on the curve of everyone who has applied. And the best candidate of the lot gets the job. You can't quantify what that looks like in advance. I'm excited to see who makes it to the final stage. I already hear early whispers that we got some exceptional applicants in this round. It would be great to help counter the narrative that this industry no longer needs juniors. That's simply retarded. However good AI gets, we're always going to need people who know the ins and outs of what the machine comes up with. Maybe not as many, maybe not in the same roles, but it's truly utopian thinking that mankind won't need people capable of vetting the work done by AI in five minutes.
Recently I got a question on formal methods1: how does it help to mathematically model systems when the system requirements are constantly changing? It doesn't make sense to spend a lot of time proving a design works, and then deliver the product and find out it's not at all what the client needs. As the saying goes, the hard part is "building the right thing", not "building the thing right". One possible response: "why write tests"? You shouldn't write tests, especially lots of unit tests ahead of time, if you might just throw them all away when the requirements change. This is a bad response because we all know the difference between writing tests and formal methods: testing is easy and FM is hard. Testing requires low cost for moderate correctness, FM requires high(ish) cost for high correctness. And when requirements are constantly changing, "high(ish) cost" isn't affordable and "high correctness" isn't worthwhile, because a kinda-okay solution that solves a customer's problem is infinitely better than a solid solution that doesn't. But eventually you get something that solves the problem, and what then? Most of us don't work for Google, we can't axe features and products on a whim. If the client is happy with your solution, you are expected to support it. It should work when your customers run into new edge cases, or migrate all their computers to the next OS version, or expand into a market with shoddy internet. It should work when 10x as many customers are using 10x as many features. It should work when you add new features that come into conflict. And just as importantly, it should never stop solving their problem. Canonical example: your feature involves processing requested tasks synchronously. At scale, this doesn't work, so to improve latency you make it asynchronous. Now it's eventually consistent, but your customers were depending on it being always consistent. Now it no longer does what they need, and has stopped solving their problems. Every successful requirement met spawns a new requirement: "keep this working". That requirement is permanent, or close enough to decide our long-term strategy. It takes active investment to keep a feature behaving the same as the world around it changes. (Is this all a pretentious of way of saying "software maintenance is hard?" Maybe!) Phase changes In physics there's a concept of a phase transition. To raise the temperature of a gram of liquid water by 1° C, you have to add 4.184 joules of energy.2 This continues until you raise it to 100°C, then it stops. After you've added two thousand joules to that gram, it suddenly turns into steam. The energy of the system changes continuously but the form, or phase, changes discretely. Software isn't physics but the idea works as a metaphor. A certain architecture handles a certain level of load, and past that you need a new architecture. Or a bunch of similar features are independently hardcoded until the system becomes too messy to understand, you remodel the internals into something unified and extendable. etc etc etc. It's doesn't have to be totally discrete phase transition, but there's definitely a "before" and "after" in the system form. Phase changes tend to lead to more intricacy/complexity in the system, meaning it's likely that a phase change will introduce new bugs into existing behaviors. Take the synchronous vs asynchronous case. A very simple toy model of synchronous updates would be Set(key, val), which updates data[key] to val.3 A model of asynchronous updates would be AsyncSet(key, val, priority) adds a (key, val, priority, server_time()) tuple to a tasks set, and then another process asynchronously pulls a tuple (ordered by highest priority, then earliest time) and calls Set(key, val). Here are some properties the client may need preserved as a requirement: If AsyncSet(key, val, _, _) is called, then eventually db[key] = val (possibly violated if higher-priority tasks keep coming in) If someone calls AsyncSet(key1, val1, low) and then AsyncSet(key2, val2, low), they should see the first update and then the second (linearizability, possibly violated if the requests go to different servers with different clock times) If someone calls AsyncSet(key, val, _) and immediately reads db[key] they should get val (obviously violated, though the client may accept a slightly weaker property) If the new system doesn't satisfy an existing customer requirement, it's prudent to fix the bug before releasing the new system. The customer doesn't notice or care that your system underwent a phase change. They'll just see that one day your product solves their problems, and the next day it suddenly doesn't. This is one of the most common applications of formal methods. Both of those systems, and every one of those properties, is formally specifiable in a specification language. We can then automatically check that the new system satisfies the existing properties, and from there do things like automatically generate test suites. This does take a lot of work, so if your requirements are constantly changing, FM may not be worth the investment. But eventually requirements stop changing, and then you're stuck with them forever. That's where models shine. As always, I'm using formal methods to mean the subdiscipline of formal specification of designs, leaving out the formal verification of code. Mostly because "formal specification" is really awkward to say. ↩ Also called a "calorie". The US "dietary Calorie" is actually a kilocalorie. ↩ This is all directly translatable to a TLA+ specification, I'm just describing it in English to avoid paying the syntax tax ↩
While Stripe is a widely admired company for things like its creation of the Sorbet typer project, I personally think that Stripe’s most interesting strategy work is also among its most subtle: its willingness to significantly prioritize API stability. This strategy is almost invisible externally. Internally, discussions around it were frequent and detailed, but mostly confined to dedicated API design conversations. API stability isn’t just a technical design quirk, it’s a foundational decision in an API-driven business, and I believe it is one of the unsung heroes of Stripe’s business success. This is an exploratory, draft chapter for a book on engineering strategy that I’m brainstorming in #eng-strategy-book. As such, some of the links go to other draft chapters, both published drafts and very early, unpublished drafts. Reading this document To apply this strategy, start at the top with Policy. To understand the thinking behind this strategy, read sections in reverse order, starting with Explore. More detail on this structure in Making a readable Engineering Strategy document. Policy & Operation Our policies for managing API changes are: Design for long API lifetime. APIs are not inherently durable. Instead we have to design thoughtfully to ensure they can support change. When designing a new API, build a test application that doesn’t use this API, then migrate to the new API. Consider how integrations might evolve as applications change. Perform these migrations yourself to understand potential friction with your API. Then think about the future changes that we might want to implement on our end. How would those changes impact the API, and how would they impact the application you’ve developed. At this point, take your API to API Review for initial approval as described below. Following that approval, identify a handful of early adopter companies who can place additional pressure on your API design, and test with them before releasing the final, stable API. All new and modified APIs must be approved by API Review. API changes may not be enabled for customers prior to API Review approval. Change requests should be sent to api-review email group. For examples of prior art, review the api-review archive for prior requests and the feedback they received. All requests must include a written proposal. Most requests will be approved asynchronously by a member of API Review. Complex or controversial proposals will require live discussions to ensure API Review members have sufficient context before making a decision. We never deprecate APIs without an unavoidable requirement to do so. Even if it’s technically expensive to maintain support, we incur that support cost. To be explicit, we define API deprecation as any change that would require customers to modify an existing integration. If such a change were to be approved as an exception to this policy, it must first be approved by the API Review, followed by our CEO. One example where we granted an exception was the deprecation of TLS 1.2 support due to PCI compliance obligations. When significant new functionality is required, we add a new API. For example, we created /v1/subscriptions to support those workflows rather than extending /v1/charges to add subscriptions support. With the benefit of hindsight, a good example of this policy in action was the introduction of the Payment Intents APIs to maintain compliance with Europe’s Strong Customer Authentication requirements. Even in that case the charge API continued to work as it did previously, albeit only for non-European Union payments. We manage this policy’s implied technical debt via an API translation layer. We release changed APIs into versions, tracked in our API version changelog. However, we only maintain one implementation internally, which is the implementation of the latest version of the API. On top of that implementation, a series of version transformations are maintained, which allow us to support prior versions without maintaining them directly. While this approach doesn’t eliminate the overhead of supporting multiple API versions, it significantly reduces complexity by enabling us to maintain just a single, modern implementation internally. All API modifications must also update the version transformation layers to allow the new version to coexist peacefully with prior versions. In the future, SDKs may allow us to soften this policy. While a significant number of our customers have direct integrations with our APIs, that number has dropped significantly over time. Instead, most new integrations are performed via one of our official API SDKs. We believe that in the future, it may be possible for us to make more backwards incompatible changes because we can absorb the complexity of migrations into the SDKs we provide. That is certainly not the case yet today. Diagnosis Our diagnosis of the impact on API changes and deprecation on our business is: If you are a small startup composed of mostly engineers, integrating a new payments API seems easy. However, for a small business without dedicated engineers—or a larger enterprise involving numerous stakeholders—handling external API changes can be particularly challenging. Even if this is only marginally true, we’ve modeled the impact of minimizing API changes on long-term revenue growth, and it has a significant impact, unlocking our ability to benefit from other churn reduction work. While we believe API instability directly creates churn, we also believe that API stability directly retains customers by increasing the migration overhead even if they wanted to change providers. Without an API change forcing them to change their integration, we believe that hypergrowth customers are particularly unlikely to change payments API providers absent a concrete motivation like an API change or a payment plan change. We are aware of relatively few companies that provide long-term API stability in general, and particularly few for complex, dynamic areas like payments APIs. We can’t assume that companies that make API changes are ill-informed. Rather it appears that they experience a meaningful technical debt tradeoff between the API provider and API consumers, and aren’t willing to consistently absorb that technical debt internally. Future compliance or security requirements—along the lines of our upgrade from TLS 1.2 to TLS 1.3 for PCI—may necessitate API changes. There may also be new tradeoffs exposed as we enter new markets with their own compliance regimes. However, we have limited ability to predict these changes at this point.
I've been biking in Brooklyn for a few years now! It's hard for me to believe it, but I'm now one of the people other bicyclists ask questions to now. I decided to make a zine that answers the most common of those questions: Bike Brooklyn! is a zine that touches on everything I wish I knew when I started biking in Brooklyn. A lot of this information can be found in other resources, but I wanted to collect it in one place. I hope to update this zine when we get significantly more safe bike infrastructure in Brooklyn and laws change to make streets safer for bicyclists (and everyone) over time, but it's still important to note that each release will reflect a specific snapshot in time of bicycling in Brooklyn. All text and illustrations in the zine are my own. Thank you to Matt Denys, Geoffrey Thomas, Alex Morano, Saskia Haegens, Vishnu Reddy, Ben Turndorf, Thomas Nayem-Huzij, and Ryan Christman for suggestions for content and help with proofreading. This zine is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, so you can copy and distribute this zine for noncommercial purposes in unadapted form as long as you give credit to me. Check out the Bike Brooklyn! zine on the web or download pdfs to read digitally or print here!
