James Prickitt

We don’t have much usability testing in developer tools. The reason is we don’t have an answer to the question “Usable for whom?” Programming has been made into an arcane art with high barriers to entry. So there are relatively few people in the world with the know-how to create a tool that optimizes ~other developer’s~ workflow. Those same folks with the expertise to create a dev tool will not be able to see the delta between what’s obvious to them and what’s obvious to the other ~25 million developers in the world, with their distinct specialites and experience levels. I think about the rise of personal computing, and the economic opportunity that came with converting as many new computer users as possible. UX came from the scramble to meet this opportunity, with usability as a core goal. Decades later, we know that working to meet users where they are, and mapping your product to their context, pays off big time in consumer software. Developers are also users. But DX is far less established than UX, and most developer tools companies don’t get clear enough on who their ideal user is. Without that, it's not possible to discern what "accessible" would mean. A lot of money is being left on the table as a result.

6

New post up on the blog. In this one, I reminisce about PHP and warn against abusing lambdas. https://lnkd.in/gDGuQFsE

5

I have some interesting experimental prompt frameworks coming. I will provide templates for these publicly. These prompts will focus on reasoning and thinking with Claude 3.5 Sonnet.

3

1 Comment

Bad #marketingops: - Undocumented processes  - Too many tools - Constant fire drills  - CMS and systems littered with hacks and work arounds - Limited resources supporting GTM motions Bada$$ #marketingops: - Clear definitions and process documentation - Lean, purposeful #martech stack  - Prioritized, quarterbacked project management - Right mix of internal/external expertise #opsenabledgrowth #demandgen #revops

20

5 Comments

Oy. Yesterday I saw someone present to CTOs a formula for measuring individual #DeveloperProductivity* . Then Laura Tacho kicked off another talk admitting that devs get a 'nervous twinge' because this topic has been 'weaponized' against you. So how do you measure productivity without breaking trust? In this #MustRead, Quotient's Lizzie Matusov tells me it's all about Trust and Transparency. And making it simple to collect Team and Org-level Metrics, to make sure Everyone has the same view, and to offer Suggested Next Steps. *To be clear, while I believe we should privately check in with our own productivity (check in later this month for my column!), but individual productivity measurements are far too likely to be abused. I attended that talk thinking I'd cover it, but I refuse to do any work that facilitates folks losing their jobs.

103

32 Comments

New addition to my #WebComponents collection: <bluesky-replies> A very simple component to render post replies on any web page. Could be used for blog comments, also works with my <bluesky-post> component

6

I think this is a really important topic that we don't see much in Engineering recruiting. A great way to lead to #burnout, especially when you're already giving your all in a full-time dev job, is to take on side projects that aren't work-related. For some, programming is their whole life, and if that is who you are looking for in an employee, then this might be a great metric to use. Otherwise, it most definitely is not. #TechCareers #BurnoutPrevention #WorkLifeBalance #Programming #SoftwareDevelopment #TechCommunity #CareerAdvice #TechTalk #GitHub #EngineeringLeadership

6

2 Comments

How’d you know if someone is a devout TDD-er? How’d you know if someone hates PHP? How’d you know if someone is Vegan? They tell you. I love coding. I don’t care about the language I use or technology. I care about solving problems. Software dogmatism is rife. Do you believe your programming language is better than the rest because it’s the only one you know? Do you assume a certain platform / framework is better because it’s popular? Do you bash people that use a different tech stack to what you do? Dont be a “confidently wrong” developer. Dont be dogmatic, intolerant and impractical. Be curious. Be humble. Be open-minded. Be a problem solver.

70

77 Comments

New #WebComponents article: This is a write-up and some extra context for my <heading-anchors> Web Component. Thanks to Curtis Wilcox for the accessibility feedback and Zach Leatherman for recommending others to use it with 11ty 💜

15

PSA - The !important (below) blocks the #JavaScript scroll eventListener. Yeah, who knew??! Note: Not my code. It came from someone else's #WordPress theme / work. html, body { max-width: 100% overflow-x: hidden !important; }

1

📝New guide: "How to detect website text content changes" using #SkybearNET. Specifically, detecting changes in PagerDuty's vCard latest version date.🎉 A real-world scenario I have been using for a few years now as part of my on-call duties😅 Also notifies you by email when changes are detected! Don't miss on price changes, important text changes, or anything else you can check on the response headers or body. Ultimate flexibility! https://lnkd.in/dsyqKNVM

2

I want to talk about “branding debt,” which is kinda like the reverse of technical debt. Technical debt is when software engineers cut corners in coding to get something out on time. While the software will work, more will need to be added for the product to be effective long term. Branding debt, by contrast, is when marketers add stuff into the brand that don’t need to be there. Tag lines, products, positions, distinctive assets, etc. It turns the whole brand into a tangled mess I’m working right now on the exam for Mark Ritson and Marketing Week’s Mini MBA, where you must create a marketing plan for a fictional company. They have a pile of previous work that is mostly junk. Wading through this junk to find ways to be consistent is a challenge. For the exam, I can quickly decide to drop stuff that isn’t working, but that obviously isn’t how real life works. There are egos and internal politics to care about. This is one of the many reasons to keep your marketing disciplined. It’s easy for bad marketers to add more to their brand. By being focused and consistent your brand will have a longer life.

20

7 Comments

Caching can make or break the experience. I’ve seen sites soar with the right strategy and struggle when caching is overlooked. It’s not just about speed; it’s about keeping readers happy and your site running smoothly. https://lnkd.in/eUgqYRzG

2

New CSS that can actually be used in 2024 https://buff.ly/4fya8Kn The amount of CSS novelty in the last two to four years has been staggering. Multiple innovations have been released and are now supported in all modern browsers, and some of them fundamentally change how to make websites. This page is a cheat sheet for new CSS things I want to start using in 2024. Note that, to support older browsers, it’s sometimes a necessity to use polyfills. Source: New CSS that can actually be used in 2024 (https://buff.ly/3NS7R16)

86

6 Comments

Is This the Future of JavaScript Tooling? 🚀 Faster, Leaner, and All-in-One for Linting, Testing, and Bundling I’ve just checked out voidZero and their new toolset, and it looks like a serious upgrade for JavaScript developers. They’re rolling out tools like oxc and RollDown – both Rust-based and built to tackle some major pain points. oxc is more than just a linter. It’s handling linting, parsing, testing, and bundling with up to 50-100x faster speeds than traditional tools like ESLint. This speed boost alone is huge for anyone working with large codebases. And then there’s RollDown, voidZero’s take on bundling. Designed as a high-performance alternative to Rollup, it’s all about precision and speed without the usual heavy lifting. RollDown aims to make the bundling process faster and more efficient, especially in production environments, where performance really matters. With tools like these, voidZero’s goal seems clear: simplify workflows, reduce toolchain clutter, and keep things lightweight and developer-friendly. If you’re ready to streamline and speed up your setup, this might be worth a closer look. Anyone else exploring this? Would love to hear your experiences and thoughts 🙏🏻

1