Stuart Langridge
Stuart is a web hacker, author, and speaker living in the UK. He writes books about JavaScript and convinces people to use Ubuntu, and also works for Canonical creating the Ubuntu One online service and travels around the world talking about it. Stuart is well known for being the presenter of the late LugRadio podcast, along with co-host Jono Bacon, a show which contributed to the cause of free software. Stuart is widely considered to be an expert in all things DOM and JavaScript.
Code and writings (and the occasional rant) are to be found at kryogenix.org and @sil on Twitter; Stuart is to be found outside in the rain looking for the smoking area.
More thoughts by Stuart Langridge:
-
Don't be creative. Be a creator. No one ever looks back and wishes that they'd given the world less stuff.
-
There's a lot to think about when you're building something on the web. Is it accessible? How do I handle translations of the text? Is the design OK on a 320px-wide screen? On a 2320px-wide screen? Does it work in IE8? In Android 4.0? In Opera Mini? Have I minimized the number of HTTP requests my page requires? Is my JavaScript minified? Are my images responsive? Is Google Analytics hooked up properly? AdSense? Am I handling Unicode text properly? Avoiding CSRF? XSS? Have I encoded my videos correctly? Crushed my pngs? Made a print stylesheet?
We've come a long way since:
<HEADER> <TITLE>The World Wide Web project</TITLE> <NEXTID N="55"> </HEADER> <BODY> <H1>World Wide Web</H1>The WorldWideWeb (W3) is a wide-area<A NAME=0 HREF="WhatIs.html"> hypermedia</A> information retrieval initiative aiming to give universal access to a large universe of documents.Look at http://html5boilerplate.com/—a base level page which helps you to cover some (nowhere near all) of the above list of things to care about (and the rest of the things you need to care about too, which are the other 90% of the list). A year in development, 900 sets of changes and evolutions from the initial version, seven separate files. That's not over-engineering; that's what you need to know to build things these days.
The important point is: one of the skills in our game is knowing what you don't need to do right now but still leaving the door open for you to do it later. If you become the next Facebook then you will have to care about all these things; initially you may not. You don't have to build them all on day one: that is over-engineering. But you, designer, developer, translator, evangelist, web person, do have to understand what they all mean. And you do have to be able to layer them on later without having to tear everything up and start again. Feel guilty that you're not addressing all this stuff in the first release if necessary, but you should feel a lot guiltier if you didn't think of some of it.
-
Publish and be damned, said the Duke of Wellington; these days, in between starting wars in France and being sick of everyone repeating the jokes about his name from Blackadder, he’d probably say that we should publish or be damned. If you’re anything like me, you’ve got folders full of little experiments that you never got around to finishing or that didn’t pan out. Put ’em up somewhere. These things are useful.
Twitter, autobiographies, collections of letters from authors, all these have shown us that the minutiae can be as fascinating as carefully curated and sieved and measured writings, and who knows what you’ll inspire the next person to do from the germ of one of your ideas?
-
It's OK to not want to build websites for everybody and every browser. Making something which is super-dynamic in Chrome 18 and also works excellently in w3m is jolly hard work, and a lot of the time you might well be justified in thinking it's not worth it. If your site stats, or your belief, or your prediction of the market's direction, or your favourite pundit tell you that the best use of your time is to only support browsers with
querySelector, or only support browsers with JavaScript, or only support WebKit, or only support iOS Safari, then that's a reasonable decision to make; don't let anyone else tell you what your relationship with your users and customers and clients is, because you know better than them.Just don't confuse what you're doing with supporting "the web". State your assumptions up front. Own your decisions, and be prepared to back them up, for your project. If you're building something which doesn't work in IE6, that requires JavaScript, that requires mobile WebKit, that requires Opera Mobile, then you are letting some people down. That's OK; you've decided to do that. But your view's no more valid than theirs, for a project you didn't build. Make your decisions, and state what the axioms you worked from were, and then everyone else can judge whether what you care about is what they care about. Just don't push your view as being what everyone else should do, and we'll all be fine.
-
Ever notice how the things you slave over and work crushingly hard on get less attention, sometimes, than the amusing things you threw together in a couple of evenings?
I can't decide whether this is a good thing or not.
Here are the dates of Stuart Langridge's future thoughts:
- Saturday, 16 June
- Monday, 30 July
- Saturday, 18 August
- Tuesday, 18 September
- Tuesday, 16 October
- Sunday, 18 November
- Tuesday, 18 December