Scott Jehl

Scott works as a web designer and developer with the bright folks at Filament Group, where he creates sites and applications for a range of clients like the Boston Globe and LEGO Inc., and commonly contributes to the open source community. He enjoys writing about web design. He co-authored the book, Designing with Progressive Enhancement (2010, New Riders), and is currently writing Responsible Responsive Design, due out in early 2014. Scott speaks regularly at conferences such as An Event Apart, Breaking Dev, and Mobilism.

Scott lives in Seagrove Beach, Florida and tweets over at @scottjehl

Published Thoughts

Under-promise and over-deliver is a piece of advice I’ve heard repeated many times since I started working in this field. As a web designer, following this advice has long brought me comfort and insurance; if I ran into problems, it made it okay to fall short of my best-case without missing a delivery, and when the work went faster than I expected, it made for a pleasant surprise for everyone involved. And well, that’s always been nice for me. But recently, I realized I’d only ever thought of this advice as a means of protecting my own interests…

That changed when I began searching for a contractor to do some remodeling on my house. The process would start similarly each time. I’d call a contractor and tell them that I needed help on some projects and they’d begin asking the sort of questions that would inform an estimate. Invariably, they would tell me a date in the near future when I should expect to receive a quote. Invariably again, that day would arrive and I’d hear nothing. Usually days would pass after that—often a week or more. I’d send them a meek nudge via email or text, “just wondering if you might get around to that estimate sometime soon,” and again I’d hear, “absolutely - should have you something by [day].”

In hindsight, the key word in their communication was “should,” and I heard it enough times that I came to realize that they meant it to convey a best-case scenario, if all went according to plan. In other words, they were over-promising, and since they were as busy as any of us they were leaving themselves little chance of meeting the expectations that they set. As a client, it was frustrating because it left me unsure of whether to call another contractor, bother them again, or just keep waiting it out. In the cases where I did receive a quote, it was often so late that I was left feeling uncertain I could trust them to actually do the work they had quoted, or at least in a reasonable amount of time. Cold feet would lead me to call yet another company and start the process over again.

Of course, this is not just about the construction industry. This communication breakdown happens in our field all the time. It feels good to please our clients, and I suspect the urge to be a “yes”-person often leads us to overload ourselves with projects, deadlines, and best-case promises that sometimes fall short. In that regard, under-promising can certainly aid us service providers in maintaining our own sanity, but playing the role of a client reminded me that in the early stages of a project, we service providers hold a great deal of power over our clients who don’t have the luxury of such strategies. Clients have a problem to solve and they’re asking us for help. It’s important that we never let our urge to say the most pleasing thing cause us to take advantage of people’s time and trust.

Thankfully for me, things have improved lately on the remodeling front. In yet another meeting with a new contractor last week, they told me they should be able to get me a quote by the end of the week. There was that word again. This time though, two days before the deadline they really did send me a quote! It was a little more expensive than others I’d received, but the trust they conveyed made me feel like they were worth it, and I decided to hire them for the job. They start tomorrow.

In our industry, we talk a lot about empathy. But it’s most often about our users and less frequently about our clients. Frustrations aside, I’m glad that this small role reversal placed me in the shoes of the client, and reminded me that this oft-repeated phrase is not just insurance for our service, but a service in itself.

For the past couple of years I’ve been roasting coffee at home. Did you know that is a thing that people do? Well I didn’t… at least until my friend Karl introduced me to the idea. As it turns out it can be really easy to do, it’s cost-effective, and most importantly it’s a good way to make a consistently great (or at worst, decent yet interesting) cup of coffee. From what I have read, there are plenty of ways to roast green coffee beans. My setup isn’t particularly fancy. I have a FreshRoast SR 300, which is basically a glass cup with holes in the bottom that sits atop a hair dryer. This particular model has only one variable in my control: time. Hot air passes through the cylinder and rotates the beans around as they roast. After starting a roast, I flip the hot air to cool after around 6 and a half minutes, as that duration typically gets me past the first “crack” and leaves me with a nice light to medium hue. Which reminds me: since I started roasting at home I’ve really taken to lighter roasts, which are pretty much the opposite of the super-dark pine-tar french roasts I used to demand. Much to my surprise, light roasts carry just as much pixel-pushing superpower (caffeine) as a pungent dark roast, but they allow a particular bean’s own flavor characteristics be brighter and more noticeable, which makes the whole home-roasting and bean selection process a bit more enjoyable (I should mention that Sweet Maria’s is a fantastic place to shop for beans if you're in the market). These days, I typically make one roast a week. In less than 10 minutes my coffee is ready to grind up and brew, though it’s noticeably better if I wait a few days after roasting before consuming. Oh, and I probably should mention that toasty aroma that fills my porch during roasting… ugh. It is just so, so nice. Worth getting into it for that alone, I say.

Anyway, I’m still pretty new to home roasting and certainly can't claim any expertise. But I can say that since I started I’ve had no need or desire to buy roasted coffee beans from a store, and as a hobby home roasting brings me enough enjoyment each week that I felt compelled to share with you folks, who I gather drink a fair amount of coffee yourselves. I’d encourage you to give it a try.

I’ve worked on many projects for clients who came with pre-established style and interactivity guidelines for how their brand should be presented online. In building multi-device sites for these clients, these guidelines have posed interesting, sometimes unrealistic expectations to live up to (at least, if attempting to follow them consistently across several devices). Many browsers simply aren’t capable of rendering design details at a level of fidelity that may be considered “on-brand,” or say, of utilizing reliable fixed-position toolbars, responsive touch gestures, and luxurious momentum-based scrolling animations. Attempts to force unnatural presentation and behavior into a given device’s experience often lead to ugly workarounds that cost users in time and bandwidth, and that can certainly have a negative impact on how a company is perceived as well. The variety of devices accessing our sites can and should lead to discussions about the many acceptable ways in which a service can be presented to users, and hopefully lead to a more fluid definition of how a brand should present itself in general.

I’m not really sure what guidelines that consider these constraints would look like, though I suspect they could involve some degree of component-level grading. One thing I am pretty sure of though, is that having a fast, accessible, user-friendly site can reflect incredibly well on a company, and I’d love to see more guidelines and expectations that prioritize these aspects of a service as branding requirements in addition to the usual visual details. Perhaps a user interface that pairs appropriately with a device’s own features and constraints reflects better on a company than one that painstakingly conforms to that company’s aesthetic ideals. I tend to think that’s the case, at least. To that end, I’ve made an effort to avoid using terms like “degraded” when comparing the ways a user interface may manifest itself on a particular device and instead focus on whether the interface feels appropriate and at-home on that device. After all, from a customer service perspective, “meeting our users wherever they are” may be more “on-brand” than anything a style guide can hope to suggest.

It has been a complicated couple of weeks for our industry. With no intention of trivializing matters, I’ll try and start this week on a light note.

Here is some footage of a sea turtle making its way to the Gulf of Mexico, moments after hatching on the beach last summer.

Have a nice week. Maybe spend some time outside, yeah?

Where I work, we spend a lot of time talking about how to make websites that work “everywhere” and work even better in favorable browsing conditions. The term often used to describe this approach is “Progressive Enhancement,” or “PE”—and lately, I’ve noticed it may have an unfair stigma attached.

I sometimes hear concerns from developers that building with PE will hold them back from innovating and “moving the web forward,” because it requires spending a lot of time making things work in old or obscure browsers that aren’t presumed to be common within their audiences.

I find the opposite to be true. For us, building with Progressive Enhancement moves almost all of our development time and costs to newer browsers, not older ones. Perhaps that’s because our approach to PE has evolved beyond a mere separation of concerns to a more controlled test-driven application of technologies—which is a rather abstruse way of saying that we just serve something basic and useful to everyone and then heavily qualify everything we do from there. This means that ancient or under-featured browsers like IE 6 or BlackBerry 5 still have access to the core content and functionality, but newer code doesn’t reach them and potentially break their experience. There are so many new, popular browsers for us to care about nowadays that we rarely have time to test in older/unqualified browsers until the late stages of a project, but when we do, things frequently work as expected—which is to say very simply, and that’s often appropriate anyhow.

The hard part of building with Progressive Enhancement isn’t supporting older browsers, it’s supporting newer browsers. But that’s hard no matter the approach. I suspect that building without PE would actually increase time spent fiddling with older browsers, that is, as long as basic access remains a priority. But in fairness, I can only speak to the approach that works for us. Progressive Enhancement frees us to focus on the costs of building features for modern browsers, without worrying much about leaving anyone out. With a strongly qualified codebase, older browser support comes nearly for free.

They say a picture is worth a thousand words. As web users, I think we should be so lucky. 1000 words of English text weighs a mere 5 kilobytes, and less than half that when compressed for delivery across the web. I’ve got favicons bigger than that. The average “picture” on the web is nearly 11 times that size, at 26 kb, and in JPG format, that doesn’t deliver a whole lot of quality. Which is why we encounter images much, much heavier than that every time we use the web – each one adding time to our wait, chipping away at our limited data plans, and for some, chipping away at our ability to access the web regularly.

This month the average webpage weighs just shy of 1700 kb, and at its current rate it’ll hit 2 megabytes within the year. Images alone hold a 61% share of that weight. It’s a hard problem. Perhaps we should just modernize our proverb… say, “a picture is worth about 11,000 words, if it’s a pretty small image, and if the words in question would have been compressed with gzip.” Or maybe something hip for the kids: “an avatar is worth a thousand words.”

Fortunately, we won’t have to consider such things, because the past few years in our field have witnessed a movement to address this very problem directly, and from many angles. At the core of the issue is that not all browsers, devices, networks, and data plans are equal. We need better tools to tailor our imagery to our users’ needs, and finally starting this month, we’re getting to see the tireless work of the folks in the Responsive Images Community Group (and the web community at large) start to come to fruition. Real, standardized tools are beginning to land in browsers such as HTTP Client-Hints (which may help us create assets that retain responsive characteristics when reposted and shared), and client-side (HTML) tools like the srcset attribute and its wonderful parent element, picture (another “picture” that cost well over a thousand words, if you read the standards mailing lists). These tools will be available to us very soon, and will be easy to polyfill for existing browsers as well. I simply can not wait to see the effect that they will have across the web.

That’s all I’ve got this month. I hope this post was worth 407 words (2306 bytes, 1201 after gzip).

At some point in 2013, I started surfing. Well okay, surfing doesn’t really describe whatever that is that I’m doing out there in the water… “not-drowning,” “panic-swearing,” “pearl-diving” are all a bit more pertinent. I’m not good, is what I’m getting at, and the prospect of that changing soon isn’t very good either. For one, the northwest stretch of Florida coastline I now call home is not exactly known as an exemplar of consistent wave conditions. What’s more, surfing is perhaps unsurprisingly the lowest priority thing I do, long after I finish making websites, spending time with my wife and baby daughter, doing household chores, writing about websites, etc., etc.

But damned if I don’t love being out there when I get the chance. Stripped of the campy ways in which surfing is typically marketed to us, it’s hard to think of a purer, more independent, mentally and physically challenging, environmentally sustainable, and often starkly beautiful activity to invest time into. And heck, learning something genuinely new, a bit scary, and thankfully un-tethered to a screen in my 30s feels profoundly refreshing.

Winter has set in here, which in north-Florida terms doesn’t mean the season is over, but just that I’ll be donning a wetsuit to ward off some chillier conditions whenever I get the chance to venture out. One day a week or so ago, it was 39°F outside with a water temperature of 61°. The waves were up for the third day in a row, and when they’re up around here, you try to take advantage. I got off work, snatched my wetsuit off the clothesline outside, and cringed as I slid one foot into the cold, spongey, still-soggy-from-2-days-ago neoprene. I briefly reconsidered the idea of paddling out at all, but a thought occurred to me that maybe putting on an already-wet wetsuit could be framed as a sort of goal, rather than the minor deterrent that it was. As the thinking went, if I’m putting on a wet wetsuit, I must have done a few things right that week; I balanced my priorities well enough to find time for this low-priority thing I enjoy, a thing that makes me feel alive (and nearly not, at times — see pearl diving, above), and repeatedly at that.

Look, if you came here for a “disruptive,” mind-blowing epiphany, I’m sorry — this one may not have a lot of legs.

But here’s the thing. Many of us builders of the web have little trouble finding inspiration to do more work; it’s the getting-away-to-do-seemingly-unimportant-things part that many of us struggle with. And yet, those things really are pretty important… for sanity, for fitness, for perspective, or at least to recharge for another round of the obviously-important stuff. Perhaps your unimportant-on-paper activity is experimenting in the kitchen, restoring a motorcycle, or heck, building the websites you can’t build during work hours. If I’ve observed anything at work (and particularly from maintaining our own portfolio website), if you don’t consider your lowest priority to-dos important, they have a way of never happening.

And so, while I’ve never been much for New Year’s resolutions, for 2014, here’s a simple one that’s been working for me so far: I’ll aim to keep my wetsuit wet.