Articles

This week, a significant portion of the Web fell over when on Tuesday, sites powered by Fastly were impacted by a massive outage that affected around 85% of the network.

The near-total collapse — which was quickly identified and remedied — took out sites including GitHub, Stack Overflow, PayPal, Shopify, Stripe, Reddit, Amazon, and CNN. Furthermore, it was all but impossible to express rage on Twitter because the server that handles the social network’s emojis was also affected.

This outage was broad and severe, and we’re truly sorry for the impact to our customers and everyone who relies on them.

Nick Rockwell, Senior VP of Engineering and Infrastructure, Fastly Inc.

The incident occurred at around 10:00 UST (06:00 EST) and prompted mass “Error 503” messages. It was identified by Fastly in less than a minute and patched within an hour.

Initial analysis indicates that the whole episode was triggered by a single customer updating their settings (in a perfectly valid way) — you know those nightmares you have about clicking the wrong button and deleting the whole Web? Yeah, imagine being that person. The precise combination of settings triggered a bug in an update that had been missed in Fastly’s QA and had been sitting in production code since May 12th.

If you’ve ever visited a serious server center, you’ll know the kind of security they employ in defense of potential criminal attacks. The only center I’ve visited in person was inside a nuclear-proof bunker, involved multiple security checks, and I wasn’t even allowed into the really secure part. But it turns out, all the terrorists need to do to crash the global economy is open a CDN account and update their settings.

Fastly actually reacted far faster than previous CDN mass-outages by its competitors — one possible reason its share price soared this week. But it is still trapped in a cycle of competition in which fast and cheap are easily compared, and good is somewhat abstract…until it’s not.

Most of us feel like seasoned hands at the Web when the truth is we’re very early adopters. It will be a century or more before the Web is truly integrated into society. Still, we are building the foundations now, and future generations need those foundations to be robust. We need less focus on clawing back a few pennies, less focus on delivering sites 3 nanoseconds before a user opens their browser, and a greater focus on resilience.

Like everyone, I love eye-peelingly fast sites, and I’m more than happy to get a good deal, but personally, I don’t feel either of those things is worth waking up to an Error 503 on a site I’m responsible for.

Image via Unsplash.

Source

The post Poll: Fast CDN, Cheap CDN, Good CDN, Pick Any One… first appeared on Webdesigner Depot.


Source de l’article sur Webdesignerdepot

This week at I/O, Google unveiled the latest version of its Material Design design system, Material You.

Initially presented as an upgrade for Android 12, Material You is the biggest revision to Material Design since its launch in 2014 and will be rolling out across all Google products in the coming year.

Material You is an adaptable system that takes the building blocks of Material Design — the spacing and component approach — and skins it to allow a more personal design language, albeit a distinctly Google personal design language.

Google isn’t shy about its intention to define what constitutes good UI design, even if its efforts have so far fallen short of its ambition. Material You potentially leads us back to that undesirable state where every new site looks like a Google clone. The more cynical might suggest that Google actually does perceive all websites as Google products — a view that’s not entirely without merit given the source of most web traffic — so a homogenous approach is warranted.

Material You will certainly make its way into web design. Expect a rash of Material You ‘updates’ to popular themes and site builders that will consist mainly of pastel color palettes.

Hopefully, two major benefits of Material You will not be overlooked: firstly, Material You introduces far more emotion than Material Design allowed; secondly, Material You is flexible enough to ensure accessible design is harder to ignore.

Fundamentally, Material You is still Material Design. The basic approach remains, but it’s less rigidly enforced. Think of it less as a rulebook and more as a disapproving parent who, despite their better judgment, is willing to let you make your own mistakes.

Material Design has looked dated for a few years now, and it’s possible that Material You is just Google hoping to nail down a trend that’s escaping them. But it’s equally possible that Material You is a step closer to what Material Design was meant to be: an invisible design system that feels natural to all seven billion individuals on the planet.

Source

The post Poll: Will You Adopt Google’s Material You? first appeared on Webdesigner Depot.


Source de l’article sur Webdesignerdepot

This week, in a move like something from a particularly eventful episode of The Office, popular project management app company Basecamp banned political and societal discussion in the company’s internal communications.

In a post that has been revised for “clarification,” the company’s co-founder Jason Fried listed six rules for employees: No societal or political discussions at work; No more ‘paternalistic’ benefits; No more committees; No more lingering on past decisions; No more 360 reviews; No forgetting what we do here.

A follow-up post from Heinemeier Hansson notes that Basecamp will still permit discussion of issues deemed central to its business like anti-trust and privacy; certain civil liberties are to be championed, while others, like racism and climate change, are not.

On the surface, it seems reasonable, Fried and co-founder David Heinemeier Hansson would like you to believe that it is. After all, people are paid to work, not soapbox, right?

So why, if they’re the ones being protected, are Basecamp’s employees angry about the move?

It turns out, multiple sources from inside Basecamp are reporting that the ‘political’ and ‘societal’ issues referred to in Fried’s public memo were, in fact, frank and open conversations about Basecamp itself.

As reported by The Verge, way back in 2009, a list of ‘funny’ customer names began circulating at the company — hardly respectful, potentially racist, and certainly inappropriate. The misalignment between co-founders and staff occurred when staff members attempted to hold discrete conversations about this and numerous other diversity and inclusivity failings at the company. Fried’s move appears to be a direct attempt to halt criticism of the status quo at Basecamp.

Basecamp itself is a highly political organization: The co-founders have written several books advocating certain societal change; they even provided a campaign headquarters and substantial donation for a candidate for Chicago mayor. Both co-founders are highly active on social media, using their business positions to elevate their personal views.

The truth is that the solo entrepreneur is an almost mythical beast. Successful startups require contributions from a range of skills and experience beyond any one individual. Jason Fried may be the frontman, strutting up and down the stage in spandex pants, with David Heinemeier Hansson playing lead guitar with his teeth, but behind them, there’s a drummer keeping time, and behind them all, there’s a crew of roadies without whom none of the equipment will arrive, let alone sound good.

Basecamp’s founders argue that the company has a mission, and that mission is to create apps that streamline the workplace. But how can you develop a product that is inclusive if staff cannot discuss what inclusive means? The answer is, you can’t.

Discussing racial bias in advertising or the impact of company wastage, climate change, or gender pay gaps in HR meetings are all political and societal and lead to a healthier, more united company.

As designers, we often say that you cannot not communicate; every decision is a design decision; there is no such thing as “adesign.” Likewise, choosing to be apolitical is itself a political choice. The only way it is feasible to run a company like this is to treat employees like robots (in the word’s original sense).

If employees feel the need to discuss exclusionary policies in the workplace, do the company founders, who benefit from those policies (or they would not be in place), have a moral or legal right to restrict those discussions?

Although it is the first point in Fried’s list that has drawn most ire, it is the fourth item on the list that is most telling: “No more lingering or dwelling on past decisions.” Like a parent answering, “Because I said so,” Fried’s attitude to his staff is laid bare in one statement.

It turns out two wealthy white men would rather their employees not try to change the world or even their workplace.

When Coinbase announced a similar move last year, it lost 5% of its staff. If Basecamp suffered the same loss, it would amount to three people. Hardly a disaster. The question for the founders — who, judging by the number of follow-ups and clarifications they’ve published, are aware the ice they’re on is perilously thin — is whether this kind of controversy creates irreparable reputational damage.

 

Featured image via Pexels.

Source

The post Poll: Is Basecamp Right To Shutdown Politics At Work? first appeared on Webdesigner Depot.


Source de l’article sur Webdesignerdepot

We’re no longer arguing about whether climate change exists because climate change is now extensively documented. Instead, we’re arguing over whether climate change is a natural cycle or human-accelerated.

At this point, the reasonable best-case (yes, best-case) scenario is that global temperatures will rise by 2˚C — if we don’t meet our global emissions targets, then it will be higher.

A 2˚C rise in temperature means a rise in sea levels of approximately 20m in the next 100 years. Major cities, including New York and London, will be decimated. Florida will become an island off the coast of Georgia. Venice is in serious trouble, as is the whole of the Netherlands. Many low-lying island nations will cease to exist. If we do nothing, there may not be enough farmable land left to sustain the population.

According to sustainablewebdesign.org, the whole Internet currently churns out 3.8% of global carbon emissions. Next Tuesday is Earth Day 2021, making this a great time to think about your own sites’ carbon footprint.

The key to reducing a website’s carbon footprint is to use as little electricity as possible — including battery power because batteries need to be charged. Some simple ways to achieve this are using dark mode, limiting the amount of resource-hungry JavaScript you use, and limiting your site’s payload. Small incremental gains like these can radically reduce the damage your sites are doing, and conveniently they also happen to be good for UX.

But let’s look at the numbers. An environmentally consciously designed site that gets 100,000 page views per year will output around 0.055 tons of CO2 per annum. By comparison, the average human produces 4 tons of CO2 (in Western countries, it’s closer to 16 tons) over the same period. The world as a whole produces 43.1 billion tons of CO2 per year. If you redesign your site to reduce power usage by 10%, you’ll have solved 0.0000000000001% of the problem.

With such a small impact on such a huge problem, is it really worth changing how we design websites to limit climate change?

Featured image via USGS.

Source

The post Poll: Is Environmentally Conscious Web Design Really Worth It? first appeared on Webdesigner Depot.


Source de l’article sur Webdesignerdepot

We all get excited about new projects; we’re daydreaming about possibilities from the first contact with a potential client. Most professionals have an established onboarding process, with contracts to sign and business assets to acquire; if you’re a coder, you probably set up a fresh new repository; if you’re a designer, you create a new project folder. All of us start imagining how the case study will look in our portfolio.

But few, if any, plan for the end of a project. Offboarding clients simply isn’t a thing. We build their site, and then one day, we don’t.

It may be that the client moves on; hopefully, you’ve done a good enough job that they can’t resist bringing you on board for their next startup. All too often, projects languish in some half-life, with occasional security patches that net you a whole $5 in service charges; is that why you got into web design? Probably not. There is the desirable option of upselling; if your client’s business grows due to your work, then more work should grow it some more.

If you’re great at startups, you’re probably not great at maintaining sites in the long term. If you’re great at maintaining sites, you’re probably not great at growing them.

For every cycle of a project’s life, there are different kinds of professionals who suit it best. And conversely, different cycles of a project suit you and your skillset better than others.

We all know that a bad client — demanding, rude, late at paying — should be fired. But what about a good client — a client who pays quickly, is friendly, professional, accommodating? Would you fire a good client if you’d outgrown the work?

Source

The post Poll: Should You Fire Good Clients? first appeared on Webdesigner Depot.


Source de l’article sur Webdesignerdepot

We tend not to think about it, but the Internet has a physical dimension. It’s a complex network of wires, cables, servers, and technical odds and ends — if you really want to, you can track it down; doing so is particularly easy on small islands because there tends to be a single cable tethering the region to the wider world.

Those physical cables run all the way to your building, and although an ISP manages them, they are normally rented from public bodies as part of your national infrastructure.

Beyond the physical, international bodies govern protocols like ARP, IEEE, HTTP, NTP, FTP, and others, which control how data is transmitted through the network and keep everything playing nice.

Then, at the other end of the equation, there’s your device. It may be a phone, a tablet, a notebook, a desktop. It’s probably several of these. And because it’s your device, everything on it feels like yours. We tend to think of it as our method of accessing the Internet instead of being part of the Internet — in reality, it’s both.

On your device, the software you use to access the Internet is your browser. For 65% of people, that’s Chrome. Even if you’re reading this on Edge, it’s created with the Blink engine, an extension of Chromium, which is the basis for Chrome. In fact, almost every browser is built using a variation of Chromium, except those on Apple devices that require Apple’s own WebKit to be used instead.

Chromium is ostensibly open-source. WebKit is not, but both are geared towards their primary contributors’ business goals; neither Chromium nor WebKit will make a change that negatively impacts Alphabet or Apple.

Your browser is just a copy of a pre-compiled set of source files sat in a Git repo somewhere. You may have installed a few plugins in your browser. You may have bookmarked a few pages. You’ve probably moved it to your dock or your home screen. Those features are just nice add-ons for the GUI; what really matters is what decisions are made about how to render web technologies.

Imagine a world in which every single car used the same mid-range Ford engine. Add in a stereo, and paint it any color you like, you can even pick your own tires, but under the hood, it has to be that mid-range Ford engine. And the only justification is that it’s too much work to create an alternative.

The 2020s are going to be a time of enormous change. You can smell the panic in traditional banking sectors every time Cryptocurrency is mentioned. Real estate billionaires are desperately trying to get us back into offices we don’t want to return to. And yes, I’m sorry, but the climate crisis is looming, and it will force our hand. The values of a whole generation have been rapidly reassessed. Innovation and the potential for innovation are rife, except, ironically, on the Internet, where we’re still chugging away with the mid-range Ford engine under the hood.

The web has reached the point at which the browser engines we choose define real-world infrastructure. There’s a fork in the road: either browser engines are part of an infrastructure that should be rationalized into a single browser protocol, or alternative browser engines need to be nurtured, encouraged, and accessible by choice.

Featured image via Pexels.

Source

The post Poll: Is It Time to Merge Browser Engines Into a Browser Protocol? first appeared on Webdesigner Depot.


Source de l’article sur Webdesignerdepot

Since the skeuomorphism of the early 00s, the design trend of choice has been minimalism. In fact, minimalism has been de rigueur for substantially longer than that.

You can make a fair case that minimalism was the defining theme of the 20th century. Beginning with the widespread adoption of grotesque typefaces in mass advertising the century before, continuing through the less-is-more philosophies of the middle part of the century, and culminating in the luxury of 80s and 90s consumerism.

Minimalism has been central to the design practice of almost every designer that we recognize as a designer. It underpins the definition of the discipline itself.

With the weight of such illustrious history, it’s no wonder that the fledgling web — and in the scale of history, the web is still a very new phenomenon — adopted minimalism.

And then there’s the fact that a minimalist approach works on the web. Multiple viewports, multiple connection speeds, multiple user journeys, all of these things are so much easier to handle if you reduce the number of visual components that have to adapt to each context.

And yet, despite this, an increasing number of designers are abandoning minimalism in favor of a more flamboyant approach where form is function. It’s clearly happening. What’s not clear is whether this is a short-lived, stylistic fad or something altogether more fundamental. In other words, are designers about to abandon grids, or are they just slapping some gradients on an otherwise minimal design?

Featured image via Pexels.

Source

The post Poll: Is Minimalism a Dead Trend Walking? first appeared on Webdesigner Depot.


Source de l’article sur Webdesignerdepot