Brad McAllister

Something that landed unnoticed: Playwright support in the Angular CLI 🚀 We've been collaborating with the Playwright team to land this community schematic by Jean-François Greffier and Cédric Exbrayat in the CLI Pick between Puppeteer, Playwright, Cypress, WebdriverIO, or NightWatch! #angular #typescript #programming #webdev #webdevelopment #javascript #webperformance #webtesting

905

35 Comments

One criticism of "strict TDD" is that you're not allowed to add tests after the code's complete. I don't recognise this version of "strict" TDD. If the tests I'm left with from TDD-ing code aren't giving me that warm, fuzzy feeling, I'll explore more, maybe write more tests (or refactor to data-driven or property-based tests if it really has to work, or grab a colleague and do guided inspections if it really, really has to work, or even a proof of correctness if it really, really, really has to work. Whatever gives me sufficient warm fuzzies, basically). As far as I'm aware, after ~25 years of doing and teaching TDD, that's allowed. Not just allowed, in fact, but recommended. 5/5 on Tripadvisor. Would go again. And likewise, if I think, going forward, I don't need some of these regression tests, I'll remove them or consolidate etc. The tests that drive our design often correspond to good regression tests, but certainly not always. More work usually needs to be done. Once the design's in place from TDD-ing the code, we're in the realms of something called "testing", where tests we add are expected to pass first time (though they might not, and that's a good reason to do it). Oh, and this is probably a good time to mention mutation testing. I'll routinely mutation test new or changed code and if that throws up gaps in my regression tests, I'll fill them.

290

54 Comments

The 5 Levels of .Net Full Stack Unicorns 🦄 – FinTech Edition (Yes, I'm Hiring) 🔍 Level 1: The Focused Unicorn They’ve got .NET down pat, but if you ask them to flex into Angular or Azure? Let’s just say they’re a bit single-minded in their magic. 🔍 Level 2: The Flexible Unicorn They can work wonders on the front-end with Angular, but getting them to handle the back-end or take the reins in Azure might require a few extra spells. 🔍 Level 3: The Cloud-Savvy Unicorn They’ve got Azure expertise for days, but we’re after someone who knows the ropes of both the cloud and the craziness of a FinTech startup. That combo? Truly rare. 🔍 Level 4: The Startup-Seasoned Unicorn They’ve been there, done that in the FinTech startup world, so they can handle the heat. The real trick is finding one who doesn’t need hand-holding and is ready to own this project. 🔍 Level 5: The All-Star Unicorn This is the full package—on-site, FinTech startup experience, a master of both .NET and Angular, with Azure cloud expertise. Plus, they’ve got that “go-getter” spirit that takes projects to the finish line without breaking a sweat. Total unicorn status here. If you know where to find one of these all-stars in the #Dallas area, let’s chat. I've got an opportunity. #Fullstackdevelopers #FinTech #UnicornHunt

7

3 Comments

🚨 JavaScript Contractor Pay Rates 🚨 Over the last 2 weeks I've spoken to 80+ JavaScript Engineers, who are all contractors, and are all highly skilled and experienced with React. They have other skills of course, but their core skill is React/Typescript. Below is a scatter graph of what they are getting paid right now in a contract role. I did this a couple of years ago and I remember there was a huge variety of rates regardless of experience, what I can see from below is that rates have stabilised massively, and people are now getting paid pretty consistently dependant on their level. It's good to see the market has stabilised and that people are getting paid what they are worth. If you feel you are getting paid below market rates, reach out and we can see what other opportunities I have for you. Or if you are looking to ramp up contingent staff quickly, and need to know what you should be paying, reach out. #JavaScript #TypeScript #React #Contractors #Sydney

26

5 Comments

Was playing around with the #Zed IDE from Zed Industries this evening. There's no Windows installer for it yet but you can compile it pretty easily. Yesterday I updated my #Rust toolchain, and with Visual Studio Community made sure the MSVC C++ build tools and Windows 11 SDK were installed. I let `cargo run` do its thing while I did other stuff, and when it eventually finished, it just launched without any build errors. Love that. Still early days though, it's not yet the most intuitive to use. I do like the docs as they are now, and also the familiar keybindings from VS Code. I wonder how easy a complete newcomer would find it though. The idea of built in #collaboration features for the IDE is great. I first saw this kind of feature many years ago in JetBrains #IntelliJ IDEA, where you could do things like viewing a colleague's session and diffing files against their local copy. I thought that was amazing! We've also seen it in recent years in Visual Studio Code and VS proper with #LiveShare. One thing these kinds of tools have issues with usually though is challenges within corporations - networking policies, proxies and firewalls, data privacy policies, etc. Having an ability to use an on-premise/local server for this would be nice too I think. As far as I understand, you can run a local collaboration version, but it's just a dev tool at the moment more than a product. The ability to follow collaborators is useful for #remotePairing or #mobProgramming. Channels with a side-by-side chat window, and note links to Markdown notes is nice. I haven't really used all the features fully, but I think it might support voice? Since COVID days I've wanted a way to combine something like Discord with Live Share or MS Teams, where you can enter a "room" with your colleagues/team, similar to being in an open office while working on-site, but now working remotely. Need a break, go over into the "coffee corner" virtual room and see what people in there are chatting or talking about - whether its collaborating on something, or just socializing. Need to have a quick team breakaway discussion, use another room/channel. Having AI as another "collaborator" in the IDE is of course nice. Interesting to see the support of multiple language model integrations too. I've recently been wanting to get back into the command line world as a primary environment, and like that it has built in vim emulation (yes... I'm finally wanting to figure out how to be a vim power user... maybe). Glad to see there's a #csharp extension, as well as an #asciidoc extension, think I might be testing those pretty soon :) Looking forward to not just playing around a bit, but trying it out on a proper project, to see how it performs. zed.dev https://lnkd.in/e5Shjx8P

2

It's not all non-symbolic AI... I presented today to Accord Project Technology Working Group on recent enhancements to the Template Engine and how TypeScript expressions within templates are compiled and evaluated. Accord Project uses TypeScript over Concerto types (with Concerto also compiled to TypeScript) to provide type-safe logic/calculations for templates; ensuring that a wide array of programming errors are caught at template compilation time, not at runtime. #typesafety #soundness #typescript #templating #legaltech

17

2 Comments

There’s a fascinating ‘leaked’ PDF[1] doing the rounds at the moment. It, allegedly, was penned by Mr Beast, and in it, he outlines key factors for success at his production company. I’d recommend reading it as he speaks with great candour, and it’s packed with broadly applicable insights, but my personal favourite is this: === USE CONSULTANTS Consultants are literally cheat codes. Need to make the world’s largest slice of cake? Start off by calling the person who made the previous world’s largest slice of cake lol. He’s already done countless tests and can save you weeks worth of work. === Whenever I’ve mentored people who wanted to start out in consultancy, I’ve always said the same thing: a consultant is just someone who’s done something at least once more than their client. Consultants aren’t there to be smarter than their clients—they’re there because they’ve BEEN dumber. Consultants have already seen and made all of the right mistakes in all of the right permutations, and they’re there to stop their clients from doing the same. Your client CAN DO what you do—you’re not special. It’s just going to take them longer. You’re not hired because you know what to do—it’s because you also know what not to do. You’re not special; you’re a shortcut. 1. https://lnkd.in/eD6SYchZ

111

6 Comments

this is a really important point. Of course it's allowed! the only people I have met with that interpretation of tdd are people who have been forced into it by people who think it's the simple act of writing tests first that matters rather than the mindset that makes you realise doing otherwise is likley play. I think it's really important. When code works according to the spec we should be allowed to ship it. that doesn't mean we are finished - just hit the bare minimum standard. Then we should be allowed to invest more into it and ship it again, and again. This whole madness and obsession with tickets gets in the way of the feedback loop of determining value, shipping and exploration with the consumer of the code. The tests are the *most* important part of a codebase. I can rewrite code but I can only rewrite tests whilst the domain knowledge is clear in my mind. Tests are the source of truth for the behaviour of the system. After that point I depend on the tests to replay it to me. So refactoring and development of tests is a noble persuit that I think many teams and engineers undervalue.

1

Yesterday I was reminded that no matter how far I've come as a dev I will never be immune from forehead-slapping embarrassment in debugging... I hit an error trying to make some slightly unusual customizations using the Web API in a PCF to relate two records, neither of which were the record I was "on". the error message was your usual unhelpful "it broke, maybe check the documentation?" I've come to expect from a Microsoft product, so I dove in, head-first. I tried at least half a dozen different ways to tackle the problem, and kept getting increasingly esoteric and useless error messages. finally, after over 8 hours of debugging, I got the same original error and threw up my hands in resignation. then I clicked the generic documentation link in the error message and it took me directly to a "common errors" page, the first one on the list being exactly my problem: "this WebApi operation is case sensitive, dummy." One capitalized letter (and a reversion back to 3 commits ago) and everything was working as it should. Thank you, gods of code, for keeping me ever humbled.

33

6 Comments

I thought it might be fun to start a bit of a LinkedIn Challenge - to meet 100 Developers in 100 Days I've been a C#/.NET software developer for over 2 decades. I've actually been coding with the .NET Framework since one of its beta releases Post a comment of what your primary coding language is and how long you've been mastering your craft #softwaredevelopers #coders #100devsin100days #linkedinchallenge

2

Developers.... Can someone explain why SRE acronym has been hijacked as the moniker of the Software Engineer? According to https://lnkd.in/e2Wj2kCy - SRE stands for Site Reliability Engineer. Whilst this and a Software Engineer role may overlap, the first is not the second and the second may not be dealing with Sites at all. If you have that role or are a software engineer that uses SRE, can you explain what it means and why you have it.

5

2 Comments

Sometimes when we think about technical debt, the immediate reaction is to want to stamp it out. Fix things, upgrade everything, refactor. Now. But sometimes you have to pause and ask: will an application gain value when we refactor it? Try flipping that question around: does a badly written app have enough value that it's worth refactoring? Maybe that's the question you need to ask.

16

🔥🔥🔥amazing software engineering posts from the past week🔥🔥🔥 Ryan Murphy rejection means only that: rejection. Keep going. https://lnkd.in/ecfy5tBV Tiger Abrodi consistently posting TypeScript knowledge bombs https://lnkd.in/ebeapvQ9 Taha Hussain are you fueling the micromanager in your life?? https://lnkd.in/eFtrhiHN Eric Roby please keep your processes simple! your success depends on it! https://lnkd.in/epk7Q_qJ Owain Lewis principles for a delegation: I needed this 2 years ago. https://lnkd.in/eNvt564u Pramoda Vyasarao master the basics https://lnkd.in/eMKTMesj Matt Watson new, "exciting" and complex tech excites only one person: the person who uses it to build. long term success needs tech that is boring, known and comes with a talent pool. https://lnkd.in/e8aK9aQt Austin Belcak above average results require above average effort. Go the extra mile - the road is never crowded. https://lnkd.in/edWQGDPH Francis Nayan take. action. today! https://lnkd.in/ej-HbcpJ Callie Buruchara 7 principles for giving feedback. it's easy to see that Callie is leading from experience! https://lnkd.in/e9XXnDyq

29

Developers: Why are you using an old laptop? I did the maths on my work laptop. It’s a reasonable spec, that I’ve had since 2022. M1 Pro, 16GB RAM / 512 SSD and is 2 years old. Buying new equivalent, M4 Pro 24GB RAM / 512GB SSD would cost around £2k. Currently you can get the same model on backmarket for just under £1k. That would mean the cost of replacing is £1000 over 2 years which works out £10 per week or £2 per day if you account for working days. My pre tax daily cost to the business is around £300 per day. Split that over 7.5 hours is £40 per hour. Or 75p per minute. I run my test suite multiple times a day and it can take maybe 5 minutes a run for the full suite. All I need to do is for my laptop to save me 3 minutes a day in run time, for it to cover the £2 per day it costs to upgrade. There's no environment impact if you sell your laptop on or feed it down the chain to someone who doesn't need the power. So the question is, apart from being attached to your laptop like some panini sticker album, why would you not upgrade? #prodeveloper

73

196 Comments

Code readability: prefixes. If you've got a method which checks if a status is true or false, instead of calling it 'status', why not try 'isStatus' or 'isPending'? If you've got a method which checks if someone has replied, don't call it 'replied', call it 'hasReplied'. In my opinion, it's better to hint at whats being returned. Lots of micro-improvements like this make your code easier to read.

9

4 Comments

I dislike the excessive dogmatism in the TDD world. Yes, you absolutely can write good and bug-free code without using TDD. Having said that, a lot of criticism of TDD I see is completely unfounded. Most people who criticise TDD don't even understand what it is. The two most popular misconceptions are these: 1. You write all of your tests up front 2. You write a test for every implementation detail In TDD, you don't write all your tests upfront. Yes, you would write all the test cases for the behavior upfront, but those are not the same as the actual executable tests. Those are just very brief descriptions of the expected behavior. In TDD, you write one test and then you write the implementation. In fact, in the classic canonical TDD as defined by Robert C. Martin (aka Uncle Bob), you start with the test and then you keep switching between the test and the implementation until both are done. You use the test to "drive" your production code. This is why it's known as "test driven development". And no, you don't write a test for every implementation detail. You write a test at the public API level of the module. The implementation details are tested implicitly. Sometimes you have to go down a level and write tests against some implementation details. This is especially useful if you are developing a complex multi-part algorithm. Kent Beck refers to this concept as "shifting gears". However, this is an exception rather than the rule. A "unit" in "unit test" is not a unit of implementation. It's either a unit of behavior or a module. The name causes a lot of confusion and this is why many developers are ditching it. Some refer to these types of tests as "developer tests", as they are the tests that developers write. Some call them "component tests", as they are designed to test individual components. In my project at Microsoft, we simply refer to them as "level 1" tests, or "L1" tests.

66

47 Comments

TypeScript 5.5 – It's a Banger! It's been slightly more than two weeks since the release of TypeScript 5.5, and if you haven't explored the latest updates yet (just like me 😊), I highly recommend checking out this insightful video by Matt Pocock [https://lnkd.in/gdVeWBni] The new features include Set methods, regex syntax checking, inferred type predicates, and many more exciting additions. My favorite update is the new tsconfig template variable configDir, which enhances support for monorepos. Setting up those three levels of tsconfig used to be quite tricky, and this update makes it significantly easier. Check out the full release notes here (https://lnkd.in/gNKy-5F7). What's your favorite update? #TypeScript #TypeScript5_5 #JavaScript #WebDevelopment #Programming #Coding #SoftwareDevelopment #TechUpdates #DeveloperCommunity #Angular

23

Different types of 'I don't know' I've met lots of software engineers who excelled at school. And in doing so learned a terrible lesson: you should always know the answer. It's OK not to know the answer. In fact, as your career progresses you are going to find more and more questions where the answer is unclear - or even unknowable. Here are three types of "I don't know" which are worth exploring: "I need help". It isn't bad to ask for help. Learning is important. What I do recommend, however, before asking for help is to write down all the things you have tried. Because that way, you can tell the person you are asking what you've tried, and more importantly you'll often discover new things to try while justifying why you need help. There is a special form of "I need help" which is rather subtle. it comes across as "I don't want to do it" or "I think it is a bad idea". If you find yourself feeling either of these, check with yourself to determine if it's an I need help in disguise. 'I'll find out' this is the answer to any question where you don't know the answer. Don't blather, just say you'll find out. For extra credit, say when you'll find out by (and of course meet that commitment). "I'll find out" also leads to experimenting rather than following dogma. "How can we find out" is a manager's multiplying question. 'This is interesting' - the most valuable I don't know comes when you realise that no-one knows... or that you think everyone else is wrong. This is where you discover the truly new, and where you learn to understand other people. Every successful business has a "This is interesting" behind it. It's also the key to self awareness (which may sound like fluffy bullshit, but is actually the biggest lever for personal and career growth) So maybe you are right not to say "I don't know", but not knowing is how you improve, and should be celebrated. (incidentally, if a senior ever chides you for not knowing, you can be sure they are lacking some of the skills a person in their position should have... but reread this article, and consider if any of these framings might have made you a star rather than a target.)

9

3 Comments