rowid,title,contents,year,author,author_slug,published,url,topic 219,Speed Up Your Site with Delayed Content,"Speed remains one of the most important factors influencing the success of any website, and the first rule of performance (according to Yahoo!) is reducing the number of HTTP requests. Over the last few years we’ve seen techniques like sprites and combo CSS/JavaScript files used to reduce the number of HTTP requests. But there’s one area where large numbers of HTTP requests are still a fact of life: the small avatars attached to the comments on articles like this one. Avatars Many sites like 24 ways use a fantastic service called Gravatar to provide user images. As a user, you can sign up to Gravatar, give them your e-mail address, and upload an image to represent you. Sites can then include your image by generating a one way hash of your e-mail address and using that to build an image URL. For example, the markup for the comments on this page looks something like this:

Drew McLellan

This is a great article!

The Gravatar URL contains two parts. 100 is the size in pixels of the image we want. 13734b0cb20708f79e730809c29c3c48 is an MD5 digest of Drew’s e-mail address. Using MD5 means we can request an image for a user without sharing their e-mail address with anyone who views the source of the page. So what’s wrong with avatars? The problem is that a popular article can easily get hundreds of comments, and every one of them means another image has to be individually requested from Gravatar’s servers. Each request is small and the Gravatar servers are fast but, when you add them up, it can easily add seconds to the rendering time of a page. Worse, they can delay the loading of more important assets like the CSS required to render the main content of the page. These images aren’t critical to the page, and don’t need to be loaded up front. Let’s see if we can delay loading them until everything else is done. That way we can give the impression that our site has loaded quickly even if some requests are still happening in the background. Delaying image loading The first problem we find is that there’s no way to prevent Internet Explorer, Chrome or Safari from loading an image without removing it from the HTML itself. Tricks like removing the images on the fly with JavaScript don’t work, as the browser has usually started requesting the images before we get a chance to stop it. Removing the images from the HTML means that people without JavaScript enabled in their browser won’t see avatars. As Drew mentioned at the start of the month, this can affect a large number of people, and we can’t completely ignore them. But most sites already have a textual name attached to each comment and the avatars are just a visual enhancement. In most cases it’s OK if some of our users don’t see them, especially if it speeds up the experience for the other 98%. Removing the images from the source of our page also means we’ll need to put them back at some point, so we need to keep a record of which images need to be requested. All Gravatar images have the same URL format; the only thing that changes is the e-mail hash. Storing this is a great use of HTML5 data attributes. HTML5 data what? Data attributes are a new feature in HTML5. The latest version of the spec says: A custom data attribute is an attribute in no namespace whose name starts with the string “data-”, has at least one character after the hyphen, is XML-compatible, and contains no characters in the range U+0041 to U+005A (LATIN CAPITAL LETTER A to LATIN CAPITAL LETTER Z). […] Custom data attributes are intended to store custom data private to the page or application, for which there are no more appropriate attributes or elements. These attributes are not intended for use by software that is independent of the site that uses the attributes. In other words, they’re attributes of an HTML element that start with “data-” which you can use to share data with scripts running on your site. They’re great for adding small bits of metadata that don’t fit into an existing markup pattern the way microformats do. Let’s see this in action Take a look at the markup for comments again:

Drew McLellan

This is a great article!

Let’s replace the element with a data-gravatar-hash attribute on the element:

Drew McLellan

This is a great article!

Once we’ve done this, we’ll need a small bit of JavaScript to find all these attributes, and replace them with images after the page has loaded. Here’s an example using jQuery: $(window).load(function() { $('a[data-gravatar-hash]').prepend(function(index){ var hash = $(this).attr('data-gravatar-hash') return '' }) }) This code waits until everything on the page is loaded, then uses jQuery.prepend to insert an image into every link containing a data-gravatar-hash attribute. It’s short and relatively simple, but in tests it reduced the rendering time of a sample page from over three seconds to well under one. Finishing touches We still need to consider the appearance of the page before the avatars have loaded. When our script adds extra content to the page it will cause a browser reflow, which is visually annoying. We can avoid this by using CSS to reserve some space for each image before it’s inserted into the HTML: #comments div { padding-left: 110px; min-height: 100px; position: relative; } #comments div h4 img { display: block; position: absolute; top: 0; left: 0; } In a real world example, we’ll also find that the HTML for a comment is more varied as many users don’t provide a web page link. We can make small changes to our JavaScript and CSS to handle this case. Put this all together and you get this example. Taking this idea further There’s no reason to limit this technique to sites using Gravatar; we can use similar code to delay loading any images that don’t need to be present immediately. For example, this year’s redesigned Flickr photo page uses a “data-defer-src” attribute to describe any image that doesn’t need to be loaded straight away, including avatars and map tiles. You also don’t have to limit yourself to loading the extra resources once the page loads. You can get further bandwidth savings by waiting until the user takes an action before downloading extra assets. Amazon has taken this tactic to the extreme on its product pages – extra content is loaded as you scroll down the page. So next time you’re building a page, take a few minutes to think about which elements are peripheral and could be delayed to allow more important content to appear as quickly as possible.",2010,Paul Hammond,paulhammond,2010-12-18T00:00:00+00:00,https://24ways.org/2010/speed-up-your-site-with-delayed-content/,ux 203,Jobs-to-Be-Done in Your UX Toolbox,"Part 1: What is JTBD? The concept of a “job” in “Jobs-To-Be-Done” is neatly encapsulated by a oft-quoted line from Theodore Levitt: “People want a quarter-inch hole, not a quarter inch drill”. Even so, Don Norman pointed out that perhaps Levitt “stopped too soon” at what the real customer goal might be. In the “The Design of Everyday Things”, he wrote: “Levitt’s example of the drill implying that the goal is really a hole is only partially correct, however. When people go to a store to buy a drill, that is not their real goal. But why would anyone want a quarter-inch hole? Clearly that is an intermediate goal. Perhaps they wanted to hang shelves on the wall. Levitt stopped too soon. Once you realize that they don’t really want the drill, you realize that perhaps they don’t really want the hole, either: they want to install their bookshelves. Why not develop methods that don’t require holes? Or perhaps books that don’t require bookshelves.” In other words, a “job” in JTBD lingo is a way to express a user need or provide a customer-centric problem frame that’s independent of a solution. As Tony Ulwick says: “A job is stable, it doesn’t change over time.” An example of a job is “tiding you over from breakfast to lunch.” You could hire a donut, a flapjack or a banana for that mid-morning snack—whatever does the job. If you can arrive at a clearly identified primary job (and likely some secondary ones too), you can be more creative in how you come up with an effective solution while keeping the customer problem in focus. The team at Intercom wrote a book on their application of JTBD. In it, Des Traynor cleverly characterised how JTBD provides a different way to think about solutions that compete for the same job: “Economy travel and business travel are both capable candidates applying for [the job: Get me face-to-face with my colleague in San Francisco], though they’re looking for significantly different salaries. Video conferencing isn’t as capable, but is willing to work for a far smaller salary. I’ve a hiring choice to make.” So far so good: it’s relatively simple to understand what a job is, once you understand how it’s different from a “task”. Business consultant and Harvard professor Clay Christensen talks about the concept of “hiring” a product to do a “job”, and firing it when something better comes along. If you’re a company that focuses solutions on the customer job, you’re more likely to succeed. You’ll find these concepts often referred to as “Jobs-to-be-Done theory”. But the application of Jobs-to-Be-Done theory is a little more complicated; it comprises several related approaches. I particularly like Jim Kalbach’s description of how JTBD is a “lens through which to understand value creation”. But it is also more. In my view, it’s a family of frameworks and methods—and perhaps even a philosophy. Different facets in a family of frameworks JTBD has its roots in market research and business strategy, and so it comes to the research table from a slightly different place compared to traditional UX or design research—we have our roots in human-computer interaction and ergonomics. I’ve found it helpful to keep in mind is that the application of JTBD theory is an evolving beast, so it’s common to find contradictions across different resources. My own use of it has varied from project to project. In speaking to others who have adopted it in different measures, it seems that we have all applied it in somewhat multifarious ways. As we like to often say in interviews: there are no wrong answers. Outcome Driven Innovation Tony Ulwick’s version of the JTBD history began with Outcome Driven Innovation (ODI), and this approach is best outlined in his seminal article published in the Harvard Business Review in 2002. To understand his more current JTBD approach in his new book “Jobs to Be Done: Theory to Practice”, I actually found it beneficial to read his approach in the original 2002 article for a clearer reference point. In the earlier article, Ulwick presented a rigorous approach that combines interviews, surveys and an “opportunity” algorithm—a sequence of steps to determine the business opportunity. ODI centres around working with “desired outcome statements” that you unearth through interviews, followed by a means to quantify the gap between importance and satisfaction in a survey to different types of customers. Since 2008, Ulwick has written about using job maps to make sense of what the customer may be trying to achieve. In a recent article, he describes the aim of the activity is “to discover what the customer is trying to get done at different points in executing a job and what must happen at each juncture in order for the job to be carried out successfully.” A job map is not strictly a journey map, however tempting it is to see it that way. From a UX perspective, is one of many models we can use—and as our research team at Clearleft have found, how we use model can depend on the nature of the jobs we’ve uncovered in interviews and the characteristics of the problem we’re attempting to solve. Figure 1. Universal job map Ulwick’s current methodology is outlined in his new book, where he describes a complete end-to-end process: from customer and competitor research to framing market and product strategy. The Jobs-To-Be-Done Interview Back in 2013, I attended a workshop by Chris Spiek and Bob Moesta from the ReWired Group on JTBD at the behest of a then-MailChimp colleague, and I came away excited about their approach to product research. It felt different from anything I’d done before and for the first time in years, I felt that I was genuinely adding something new to my research toolbox. A key idea is that if you focus on the stories of those who switched to you, and those who switch away from you, you can uncover the core jobs through looking at these opposite ends of engagement. This framework centres around the JTBD interview method, which harnesses the power of a narrative framework to elicit the real reasons why someone “hired” something to do a job—be it something physical like a new coffee maker, or a digital service, such as a to-do list app. As you interview, you are trying to unearth the context around the key moments on the JTBD timeline (Figure 2). A common approach is to begin from the point the customer might have purchased something, back to the point where the thought of buying this thing first occurred to them. Figure 2. JTBD Timeline Figure 3. The Four Forces The Forces Diagram (Figure 3) is a post-interview analysis tool where you can map out what causes customers to switch to something new and what holds them back. The JTBD interview is effective at identifying core and secondary jobs, as well as some context around the user need. Because this method is designed to extract the story from the interviewee, it’s a powerful way to facilitate recall. Having done many such interviews, I’ve noticed one interesting side effect: participants often remember more details later on after the conversation has formally ended. It is worth scheduling a follow-up phone call or keep the channels open. Strengths aside, it’s good to keep in mind that the JTBD interview is still primarily an interview technique, so you are relying on the context from the interviewee’s self-reported perspective. For example, a stronger research methodology combines JTBD interviews with contextual research and quantitative methods. Job Stories Alan Klement is credited for coming up with the term “job story” to describe the framing of jobs for product design by the team at Intercom: “When … I want to … so I can ….” Figure 4. Anatomy of a Job Story Unlike a user story that traditionally frames a requirement around personas, job stories frame the user need based on the situation and context. Paul Adams, the VP of Product at Intercom, wrote: “We frame every design problem in a Job, focusing on the triggering event or situation, the motivation and goal, and the intended outcome. […] We can map this Job to the mission and prioritise it appropriately. It ensures that we are constantly thinking about all four layers of design. We can see what components in our system are part of this Job and the necessary relationships and interactions required to facilitate it. We can design from the top down, moving through outcome, system, interactions, before getting to visual design.” Systems of Progress Apart from advocating using job stories, Klement believes that a core tenet of applying JTBD revolves around our desire for “self-betterment”—and that focusing on everyone’s desire for self-betterment is core to a successful strategy. In his book, Klement takes JTBD further to being a tool for change through applying systems thinking. There, he introduces the systems of progress and how it can help focus product strategy approach to be more innovative. Coincidentally, I applied similar thinking on mapping systemic change when we were looking to improve users’ trust with a local government forum earlier this year. It’s not just about capturing and satisfying the immediate job-to-be-done, it’s about framing the job so that you can a clear vision forward on how you can help your users improve their lives in the ways they want to. This is really the point where JTBD becomes a philosophy of practice. Part 2: Mixing It Up There has been some misunderstanding about how adopting JTBD means ditching personas or some of our existing design tools or research techniques. This couldn’t have been more wrong. Figure 5: Jim Kalbach’s JTBD model Jim Kalbach has used Outcome-Driven Innovation for around 10 years. In a 2016 article, he presents a synthesised model of how to think about that has key elements from ODI, Christensen’s theories and the structure of the job story. More interestingly, Kalbach has also combined the use of mental models with JTBD. Claire Menke of UDemy has written a comprehensive article about using personas, JTBD and customer journey maps together in order to communicate more complete story from the users’ perspective. Claire highlights an especially interesting point in her article as she described her challenges: “After much trial and error, I arrived at a foundational research framework to suit every team’s needs — allowing everyone to share the same holistic understanding, but extract the type of information most applicable to their work.” In other words, the organisational context you are in likely can dictate what works best—after all the goal is to arrive at the best user experience for your audiences. Intercom can afford to go full-on on applying JTBD theory as a dominant approach because they are a start-up, but a large company or organisation with multiple business units may require a mix of tools, outputs and outcomes. JTBD is an immensely powerful approach on many fronts—you’ll find many different references that lists the ways you can apply JTBD. However, in the context of this discussion, it might also be useful to we examine where it lies in our models of how we think about our UX and product processes. JTBD in the UX ecosystem Figure 6. The Elements of User Experience (source) There are many ways we have tried to explain the UX discipline but I think Jesse James Garrett’s Elements of User Experience is a good place to begin. I sometimes also use little diagram to help me describe the different levels you might work at when you work through the complexity of designing and developing a product. A holistic UX strategy needs to address all the different levels for a comprehensive experience: your individual product UI, product features, product propositions and brand need to have a cohesive definition. Figure 7. Which level of product focus? We could, of course, also think about where it fits best within the double diamond. Again, bearing in mind that JTBD has its roots in business strategy and market research, it is excellent at clarifying user needs, defining high-level specifications and content requirements. It is excellent for validating brand perception and value proposition —all the way down to your feature set. In other words, it can be extremely powerful all the way through to halfway of the second diamond. You could quite readily combine the different JTBD approaches because they have differences as much as overlaps. However, JTBD generally starts getting a little difficult to apply once we get to the details of UI design. The clue lies in JTBD’s raison d’être: a job statement is solution independent. Hence, once we get to designing solutions, we potentially fall into a existential black hole. That said, Jim Kalbach has a quick case study on applying JTBD to content design tucked inside the main article on a synthesised JTBD model. Alan Klement has a great example of how you could use UI to resolve job stories. You’ll notice that the available language of “jobs” drops off at around that point. Job statements and outcome statements provide excellent “mini north-stars” as customer-oriented focal points, but purely satisfying these statements would not necessarily guarantee that you have created a seamless and painless user experience. Playing well with others You will find that JTBD plays well with Lean, and other strategy tools like the Value Proposition Canvas. With every new project, there is potential to harness the power of JTBD alongside our established toolbox. When we need to understand complex contexts where cultural or socioeconomic considerations have to be taken into account, we are better placed with combining JTBD with more anthropological approaches. And while we might be able to evaluate if our product, website or app satisfies the customer jobs through interviews or surveys, without good old-fashioned usability testing we are unlikely to be able to truly validate why the job isn’t being represented as it should. In this case, individual jobs solved on the UI can be set up as hypotheses to be proven right or wrong. The application of Jobs-to-be-Done is still evolving. I’ve found it to be very powerful and I struggle to remember what my UX professional life was like before I encountered it—it has completely changed my approach to research and design. The fact JTBD is still evolving as a practice means we need to be watchful of dogma—there’s no right way to get a UX job done after all, it nearly always depends. At the end of the day, isn’t it about having the right tool for the right job?",2017,Steph Troeth,stephtroeth,2017-12-04T00:00:00+00:00,https://24ways.org/2017/jobs-to-be-done-in-your-ux-toolbox/,ux 197,Designing for Mobile Performance,"Last year, some colleagues at Google ran a research study titled “The Need for Mobile Speed” to find out what the impact of performance and perception of speed had on the way people use the web on their mobile devices. That’s not a trivial distinction; when considering performance, how fast something feels is often more important than how fast it actually is. When dealing with sometimes underpowered mobile devices and slow mobile networks, designing experiences that feel fast is exceptionally important. One of the most startling numbers we found in the study was that 53% of mobile site visits are abandoned if pages take longer than 3 seconds to load. We wanted to find out more, so following on from this study, we conducted research to define what the crucial elements of speed are. We took into consideration the user experience (UX), overall perception of speed, and how differing contexts the user finds themselves in can alter how fast a user thinks something loaded. To understand speed and load times first we must understand that user mobile web behaviour is broken down into three buckets; Intention Location State of mind Let’s look at each of those in turn. Intention Users browse sites on a mobile device for many different reasons. To be able to effectively design a performant user experience for them, it’s important to understand what those reasons might be. When asked to describe their reason for visiting a site, approximately 30% of people asked by the study claimed that they were simply browsing without a particular purpose in mind. Looking deeper, we found that this number increased slightly (34%) for retail sites. 30% said they were just there to find out some information for a future task or action, such as booking a flight. Interestingly, the research shows that users are actually window shopping using their mobile browser. Only 29% actually said they had a specific goal or intent in mind, and this number increases significantly for financial services like banking sites (57%). This goes against a traditionally held view of users wanting to perform simple actions efficiently on their mobile device. Sure, some users are absolutely doing that, but many are just browsing around without a goal in mind, just like they would on a desktop browser. This gives great insight into the user’s intentions. It tells us that users are actively using sites on their mobile, but a large majority do not intend to do anything instantly. There’s no goal they’re under pressure to achieve. If a site’s performance is lousy or janky, this will only reaffirm to the user to that they can hold off on completing a task, so they might just give up. However, if a site is quick to load, sophisticated in expressing its value proposition quickly, and enables the user to perform their actions seamlessly, then turning that “browsing user” into a “buying user” becomes all that much easier. When the user has no goal, there’s more opportunity to convert, and you stand a greater chance of doing that if the performance is good enough so they stick around. Location Obviously, mobile devices by their nature can be used in many different locations. This is an interesting consideration, because it’s not something we traditionally need to take into account designing experiences for static platforms like desktop computers. The in the study, we found that 82% of users browse the web on their mobile phone while in their home. In contrast, only 7% do the same while at work. This might come across as a bit of a shock, but when you look at mobile usage – in particular app usage – most of the apps being used are either a social network or some sort of entertainment or media app. Due to the unreliability of network connections, users will often alternate between these two types of apps. The consequence being that if a site doesn’t work offline, or otherwise compensate for bad network connectivity in some way by providing opportunities to allow users to browse their site, then it becomes a self-fulfilling prophecy as to why users mostly view the mobile web from the comfort of their homes where there is a strong WiFi connection. They’re using mobile devices, but they’re not actually mobile themselves. Another thing to bear in mind is how users alternate between devices, a study by comScore found that 80% of transactions take place on desktop while 69% of the browsing takes place on mobile. Any given user might access from more than one location - they might visit one day from a bus queue on their phone, and then next day from a laptop at home. State of mind One more thing we need to take into consideration is the user’s state of mind. Whilst browsing at home, users tend to be more relaxed, and in the research 76% stated they were generally calmer at home. The user’s state of mind can have quite a big impact on how they perceive things. The calmer they are, the quicker a site might appear to load. If the user is anxious and impatiently drumming their fingers on the table, things seem to take longer, and even a small wait can feel like an eternity. This is quite key. Over 40% of sites take longer than 4 seconds to load for users who are are out and about and using a mobile data connection. Coupled with our perception, and amplified by a potentially less-than-calm state of mind, this can seem like an age. What does this all mean? I think we can all agree that users prefer strong, steady connections and comfort when completing transactions. It seems like common sense when we say it out loud. Recreating these feelings and sensations of comfort and predictability under all circumstances therefore becomes paramount. Equally, when asked in the study, users all claimed that speed was the most important factor impacting their mobile web usage. Over 40%, in fact, said it was the most important UX feature of a site, and nobody asked considered it to be of no importance at all. The meaning of speed When it comes to performance, speed is measured in two ways – real speed; as measured on a clock, and perceived speed; how responsive an interaction feels. We can, of course, improve how quickly a site loads by simply making files smaller. Even then, the study showed that 32% of users felt a site can feel slow even when it loads in less than 4 seconds. This gets even worse when you look at it by age group, with 50% if young people (18-24 year olds) thinking a site was slower than it actually was. When you add to the mix that users think a site loaded faster when they are sitting compared to when they are standing up, then you are in a world of trouble if your site doesn’t have any clear indicators to let the user know the loading state of you website or app. So what can we do about this to improve our designs? How to fix / hack user perception There are some golden rules of speed, the first thing is hacking response time. If a page takes more than 3 seconds to load, you will certainly start to lose your users. However, if that slowness is part of a UX flow such as processing information, the user understands it might take a little time. Under those circumstances, a load time of under 5 seconds is acceptable, but even then, you should take caution. Anything above 8 seconds and you are in very real danger of losing your audience completely. Load time User impression 200 ms Feels instant 1 s Feels it is performing smoothly 5 s Part of user flow 8 s Lose attention Remove the tap delay Mobile browsers often use a 300-350ms delay between the triggering of the touchend and click events. This delay was added so the browser could determine if there was going to be a double-tap triggered or not, since double-tap is a common gesture used to zoom into text. This delay can have the side effect of making interactions feel laggy, and therefore giving the user the impression that the site is slow, even if it’s their own browser causing the problem. Fortunately there’s a way to remove the delay. Add following in the of your page, and the delay no longer takes effect: You can also use touch-action: manipulation in newer browsers to eliminate click delay. For old browsers, FastClick by FT Labs uses touch events to trigger clicks faster and remove the double-tap gesture. Make use of Skeleton Screens A skeleton layout is a wireframe version of your app that displays while content is being loaded. This demonstrates to the user that content is about to be loaded, giving the impression that something is happening more quickly than it really is. Consider also using a preloader UI as well, with a text label informing the user that the app is loading. One example would be to pulsate the wireframe content giving the app the feeling that it is alive and loading. This reassures the user that something is happening and helps prevent resubmissions or refreshes of your app. Razvan Caliman created a Codepen example of how to create this effect in purely CSS. One thing to consider though, if data doesn’t load then you might need to create a fallback 404 or error page to let the user know what happened. Example by Owen-Campbell Moore Responsive Touch Feedback Carefully designing the process by which items load is one aspect of increasing the perceived speed of your page, but reassuring the user that an action they have taken is in process is another. At Google we use something called a Ripple, which is is animating dot that expands or ripples in order to confirm to the user that their input has been triggered. This happens immediately, expanding outward from the point of touch. This reaffirms to the user that their input has been received and is being acted on, even before the site has had a chance to process or respond to the action. From the user’s point of view, they’ve tapped and the page has responded immediately, so it feels really quick and satisfying. You can mimic this same behavior using our Material Design Components Web GitHub repo. Progress bars These UI elements have existed for a very long time, but research conducted by Chris Harrison and published in New Scientist found that the style of a progress bar can alter the perception of speed drastically. As a matter of fact, progress bars with ripples that animate towards the left appear like they are loading faster by at least 11% percent. So when including them in your site, take into consideration that ripples and progress bars that pulsate faster as they get to the end will make your sites seem quicker. Faster Progress Bars: Manipulating Perceived Duration with Visual Augmentations Navigation The speed with which a user can locate navigational items or call to actions adds to their perceived performance of a site. If the user’s next action is quick to spot on the screen, they don’t spend time hunting around the interface with their eyes and fingers. So no matter how quickly your code runs, hiding items behind a nav bar will make a site feel slower than it actually is. Facebook found that switching to using bottom navigation saw an increase in engagement, satisfaction, revenue, speed, and importantly, perception of speed. If the user sees the navigation items they’re looking for quickly, the interaction feels fast. What’s more, end-to-end task completion is quicker too, as the interface not only feels quicker, but actually measures quicker as well. Similar reactions were found with Spotify and Redbooth. Luke Wroblewski gave a talk last year in Ireland titled “Obvious Always Wins” which he demonstrated through the work he did with Google+. Luke’s message is that by making the core features of your app obvious to your user, you will see engagement go up. This again seems obvious, right? However, it is important to note that adding bottom navigation doesn’t just mean a black bar at the bottom of your screen like some kind of performance magic bullet. The goal is to make the items clear to the user so they know what they need to be doing, and how you achieve that could be different from one interface to the next. Google keeps experimenting with different navigation styles, but finally settled with the below when they conducted user research and testing. Conclusion By utilizing a collection of UI patterns and speed optimisation techniques, you can improve not only the actual speed of a site, but the perception of how quickly a user thinks your site is loading. It is critical to remember that users will not always be using your site in a calm and relaxed manner and that even their age can impact how they will use or not use your site. By improving your site’s stability, you increase the likelihood of positive user engagement and task completion. You can learn more about techniques to hack user perception and improve user speed by taking a look at an E-Book we published with Awwwards.com called Speed Matters: Design for Mobile Performance.",2017,Mustafa Kurtuldu,mustafakurtuldu,2017-12-18T00:00:00+00:00,https://24ways.org/2017/designing-for-mobile-performance/,ux 166,Performance On A Shoe String,"Back in the summer, I happened to notice the official Wimbledon All England Tennis Club site had jumped to the top of Alexa’s Movers & Shakers list — a list that tracks sites that have had the biggest upturn or downturn in traffic. The lawn tennis championships were underway, and so traffic had leapt from almost nothing to crazy-busy in a no time at all. Many sites have similar peaks in traffic, especially when they’re based around scheduled events. No one cares about the site for most of the year, and then all of a sudden – wham! – things start getting warm in the data centre. Whilst the thought of chestnuts roasting on an open server has a certain appeal, it’s less attractive if you care about your site being available to visitors. Take a look at this Alexa traffic graph showing traffic patterns for superbowl.com at the beginning of each year, and wimbledon.org in the month of July. Traffic graph from Alexa.com Whilst not on the same scale or with such dramatic peaks, we have a similar pattern of traffic here at 24ways.org. Over the last three years we’ve seen a dramatic pick up in traffic over the month of December (as would be expected) and then a much lower, although steady load throughout the year. What we do have, however, is the luxury of knowing when the peaks will be. For a normal site, be that a blog, small scale web app, or even a small corporate site, you often just cannot predict when you might get slashdotted, end up on the front page of Digg or linked to from a similarly high-profile site. You just don’t know when the peaks will be. If you’re a big commercial enterprise like the Super Bowl, scaling up for that traffic is simply a cost of doing business. But for most of us, we can’t afford to have massive capacity sat there unused for 90% of the year. What you have to do instead is work out how to deal with as much traffic as possible with the modest resources you have. In this article I’m going to talk about some of the things we’ve learned about keeping 24 ways running throughout December, whilst not spending a fortune on hosting we don’t need for 11 months of each year. We’ve not always got it right, but we’ve learned a lot along the way. The Problem To know how to deal with high traffic, you need to have a basic idea of what happens when a request comes into a web server. 24 ways is hosted on a single small virtual dedicated server with a great little hosting company in the UK. We run Apache with PHP and MySQL all on that one server. When a request comes in a new Apache process is started to deal with the request (or assigned if there’s one available not doing anything). Each process takes a bunch of memory, so there’s a finite number of processes that you can run, and therefore a finite number of pages you can serve at once before your server runs out of memory. With our budget based on whatever is left over after beer, we need to get best performance we can out of the resources available. As the goal is to serve as many pages as quickly as possible, there are several approaches we can take: Reducing the amount of memory needed by each Apache process Reducing the amount of time each process is needed Reducing the number of requests made to the server Yahoo! have published some information on what they call Exceptional Performance, which is well worth reading, and compliments many of my examples here. The Yahoo! guidelines very much look at things from a user perspective, which is always important. Server tweaking If you’re in the position of being able to change your server configuration (our set-up gives us root access to what is effectively a virtual machine) there are some basic steps you can take to maximise the available memory and reduce the memory footprint. Without getting too boring and technical (whole books have been written on this) there are a couple of things to watch out for. Firstly, check what processes you have running that you might not need. Every megabyte of memory that you free up might equate to several thousand extra requests being served each day, so take a look at top and see what’s using up your resources. Quite often a machine configured as a web server will have some kind of mail server running by default. If your site doesn’t use mail (ours doesn’t) make sure it’s shut down and not using resources. Secondly, have a look at your Apache configuration and particularly what modules are loaded. The method for doing this varies between versions of Apache, but again, every module loaded increases the amount of memory that each Apache process requires and therefore limits the number of simultaneous requests you can deal with. The final thing to check is that Apache isn’t configured to start more servers than you have memory for. This is usually done by setting the MaxClients directive. When that limit is reached, your site is going to stop responding to further requests. However, if all else goes well that threshold won’t be reached, and if it does it will at least stop the weight of the traffic taking the entire server down to a point where you can’t even log in to sort it out. Those are the main tidbits I’ve found useful for this site, although it’s worth repeating that entire books have been written on this subject alone. Caching Although the site is generated with PHP and MySQL, the majority of pages served don’t come from the database. The process of compiling a page on-the-fly involves quite a few trips to the database for content, templates, configuration settings and so on, and so can be slow and require a lot of CPU. Unless a new article or comment is published, the site doesn’t actually change between requests and so it makes sense to generate each page once, save it to a file and then just serve all following requests from that file. We use QuickCache (or rather a plugin based on it) for this. The plugin integrates with our publishing system (Textpattern) to make sure the cache is cleared when something on the site changes. A similar plugin called WP-Cache is available for WordPress, but of course this could be done any number of ways, and with any back-end technology. The important principal here is to reduce the time it takes to serve a page by compiling the page once and serving that cached result to subsequent visitors. Keep away from your database if you can. Outsource your feeds We get around 36,000 requests for our feed each day. That really only works out at about 7,000 subscribers averaging five-and-a-bit requests a day, but it’s still 36,000 requests we could easily do without. Each request uses resources and particularly during December, all those requests can add up. The simple solution here was to switch our feed over to using FeedBurner. We publish the address of the FeedBurner version of our feed here, so those 36,000 requests a day hit FeedBurner’s servers rather than ours. In addition, we get pretty graphs showing how the subscriber-base is building. Off-load big files Larger files like images or downloads pose a problem not in bandwidth, but in the time it takes them to transfer. A typical page request is very quick, a few seconds at the most, resulting in the connection being freed up promptly. Anything that keeps a connection open for a long time is going to start killing performance very quickly. This year, we started serving most of the images for articles from a subdomain – media.24ways.org. Rather than pointing to our own server, this subdomain points to an Amazon S3 account where the files are held. It’s easy to pigeon-hole S3 as merely an online backup solution, and whilst not a fully fledged CDN, S3 works very nicely for serving larger media files. The roughly 20GB of files served this month have cost around $5 in Amazon S3 charges. That’s so affordable it may not be worth even taking the files back off S3 once December has passed. I found this article on Scalable Media Hosting with Amazon S3 to be really useful in getting started. I upload the files via a Firefox plugin (mentioned in the article) and then edit the ACL to allow public access to the files. The way S3 enables you to point DNS directly at it means that you’re not tied to always using the service, and that it can be transparent to your users. If your site uses photographs, consider uploading them to a service like Flickr and serving them directly from there. Many photo sharing sites are happy for you to link to images in this way, but do check the acceptable use policies in case you need to provide a credit or link back. Off-load small files You’ll have noticed the pattern by now – get rid of as much traffic as possible. When an article has a lot of comments and each of those comments has an avatar along with it, a great many requests are needed to fetch each of those images. In 2006 we started using Gravatar for avatars, but their servers were slow and were holding up page loads. To get around this we started caching the images on our server, but along with that came the burden of furnishing all the image requests. Earlier this year Gravatar changed hands and is now run by the same team behind WordPress.com. Those guys clearly know what they’re doing when it comes to high performance, so this year we went back to serving avatars directly from them. If your site uses avatars, it really makes sense to use a service like Gravatar where your users probably already have an account, and where the image requests are going to be dealt with for you. Know what you’re paying for The server account we use for 24 ways was opened in November 2005. When we first hit the front page of Digg in December of that year, we upgraded the server with a bit more memory, but other than that we were still running on that 2005 spec for two years. Of course, the world of technology has moved on in those years, prices have dropped and specs have improved. For the same amount we were paying for that 2005 spec server, we could have an account with twice the CPU, memory and disk space. So in November of this year I took out a new account and transferred the site from the old server to the new. In that single step we were prepared for dealing with twice the amount of traffic, and because of a special offer at the time I didn’t even have to pay the setup cost on the new server. So it really pays to know what you’re paying for and keep an eye out of ways you can make improvements without needing to spend more money. Further steps There’s nearly always more that can be done. For example, there are some media files (particularly for older articles) that are not on S3. We also serve our CSS directly and it’s not minified or compressed. But by tackling the big problems first we’ve managed to reduce load on the server and at the same time make sure that the load being placed on the server can be dealt with in the most frugal way. Over the last 24 days we’ve served up articles to more than 350,000 visitors without breaking a sweat. On a busy day, that’s been nearly 20,000 visitors in just 24 hours. While in the grand scheme of things that’s not a huge amount of traffic, it can be a lot if you’re not prepared for it. However, with a little planning for the peaks you can help ensure that when the traffic arrives you’re ready to capitalise on it. Of course, people only visit 24 ways for the wealth of knowledge and experience that’s tied up in the articles here. Therefore I’d like to take the opportunity to thank all our authors this year who have given their time as a gift to the community, and to wish you all a very happy Christmas.",2007,Drew McLellan,drewmclellan,2007-12-24T00:00:00+00:00,https://24ways.org/2007/performance-on-a-shoe-string/,ux 160,Tracking Christmas Cheer with Google Charts,"A note from the editors: Since this article was written Google has retired the Charts API. Let’s get something out in the open: I love statistics. As an informatician I can’t get enough graphs, charts, and numbers. So you can imagine when Google released their Charts API I thought Christmas had come early. I immediately began to draw up graphs for the holiday season using the new API; and using my new found chart-making skills I’ll show you what you can and can’t do with Google Charts. Mummy, it’s my first chart The Google Charts API allows you to send data to Google; in return they give you back a nicely-rendered graph. All the hard work is done on Google’s servers — you need only reference an image in your HTML. You pass along the data — the numbers for the charts, axis labels, and so on — in the query string of the image’s URL. If you want to add charts to your blog or web site, there’s probably no quicker way to get started. Here’s a simple example: if we add the following line to an HTML page: Then we’ll see the line graph in Figure 1 appear in our page. That graph is hosted on Google’s own server1: http://chart.apis.google.com/. Figure 1: A simple example of a line graph created with Google Charts. If you look at the URL used in the example you’ll notice we’re passing some parameters along in the query string (the bit after the question mark). The query string looks like this: cht=lc&chs=200x125&chd=s:ZreelPuevfgznf2008 It’s contains everything Google Charts needs to draw the graph. There are three parameters in the query string: cht; this specifies the type of chart Google Charts will generate (in this case, lc is a line chart). chs, the value of which is 200x125; this defines the chart’s size (200 pixels wide by 125 pixels high). chd, the value of which is s:ZreelPuevfgznf2008; this is the actual chart data, which we’ll discuss in more detail later. These three parameters are the minimum you need to send to Google Charts in order to create a chart. There are lots more parameters you can send too (giving you more choice over how a chart is displayed), but you have to include at least these three before a chart can be created. Using these three parameters you can create pie charts, scatter plots, Venn diagrams, bar charts (and more) up to 1,000 pixels wide or 1,000 pixels high (but no more than 300,000 pixels in total). Christmas pie After I discovered the option to create a pie chart I instantly thought of graphing all the types of food and beverages that I’ll consume at this year’s Christmas feast. I can represent each item as a percentage of all the food on a pie chart (just thinking about that makes me hungry). By changing the value of the cht parameter in the image’s query string I can change the chart type from a line chart to a pie chart. Google Charts offers two different types of pie chart: a fancy three-dimensional version and a two-dimensional overhead version. I want to stick with the latter, so I need to change cht=lc to cht=p (the p telling Google Charts to create a pie chart; if you want the three-dimensional version, use cht=p3). As a pie chart is circular I also need to adjust the size of the chart to make it square. Finally, it would be nice to add a title to the graph. I can do this by adding the optional parameter, chtt, to the end of the image URL. I end up with the chart you see in Figure 2. Figure 2: Pie chart with a title. To add this chart to your own page, you include the following (notice that you can’t include spaces in URLs, so you need to encode them as plus-signs.): Ok, that’s great, but there are still two things I want to do before I can call this pie chart complete. First I want to label each slice of the pie. And second I want to include the proper data (at the moment the slices are meaningless). If 2007 is anything like 2006, the break down will be roughly as follows: Egg nog 10% Christmas Ham 20% Milk (not including egg nog) 8% Cookies 25% Roasted Chestnuts 5% Chocolate 3% Various Other Beverages 15% Various Other Foods 9% Snacks 5% I have nine categories of food and drink to be tracked, so I need nine slice labels. To add these to the chart, I use the chl parameter. All nine labels are sent in one value; I use the vertical-pipe character, |, to separate them. So I need to append the following to the query string: chl=Egg+nog|Christmas+Ham|Milk+(not+including+egg+nog)|Cookies|Roast+Chestnuts|Chocolate|Various+Other+Beverages|Various+Other+Foods|Snacks Next I need to add the corresponding percentage values to the chart labels. Encoding the chart data is the trickiest part of the Google Charts API — but by no means complicated. There are three different ways to encode your data on a chart. As I’m only dealing with small numbers, I’m going to use what Google calls simple encoding. Simple encoding offers a sixty-two value spectrum in which to represent data. Remember the mandatory option, chd, from the first example? The value for this is split into two parts: the type of encoding and the graph data itself. These two parts are separated with a colon. To use simple encoding, the first character of the chd option must be a lower case s. Follow this with a colon and everything after it is considered data for the graph. In simple encoding, you have sixty-two values to represent your data. These values are lowercase and uppercase letters from the Latin alphabet (fifty-two characters in total) and the digits 0 to 9. Each letter of the alphabet represents a single number: A equals 0, B equals 1, and so on up to Z, which equals 25; a equals 26, b equals 27, and so on up to z, which equals 51. The ten digits represent the numbers 52 to 61: 0 equals 52, 1 equals 53, and 9 equals 61. In the previous two examples we used the string ZreelPuevfgznf2008 as our chart data; the Z is equal to 25, the r is equal to 42, the e is equal to 30, and so on. I want to encode the percentage values 10, 20, 8, 25, 5, 3, 15, 9 and 5, so in simple encoding I would use the string KUIZFDPJF. If you think figuring this out for each chart may make your head explode, don’t worry: help is out there. Do you remember I said I needed to change the image dimensions to be square, to accommodate the pie chart? Well now I’m including labels I need even more room. And as I’m in a Christmassy mood I’m going to add some festive colours too. The optional chco parameter is used to change the chart color. You set this using the same hexadecimal (“hex”) notation found in CSS. So let’s make our pie chart green by adding chco=00AF33 (don’t start it with a hash character as in CSS) to the image URL. If we only specify one hex colour for the pie chart Google Charts will use shades of that colour for each of the slices. To choose your own colours, pass a comma separated list of colours. The “Milk” and “Cookies” slices were consumed together, so we can make those two slices more of a redish colour. I’ll use shades of green for the other slices. My chco parameter now looks like this: chco=00AF33,4BB74C,EE2C2C,CC3232,33FF33,66FF66,9AFF9A,C1FFC1,CCFFCC. After all this, I’m left with the following URL: http://chart.apis.google.com/chart?chco=00AF33,4BB74C,EE2C2C,CC3232,33FF33,66FF66,9AFF9A,C1FFC1,CCFFCC&chl=Egg+nog|Christmas+Ham|Milk+(not+including+egg+nog)|Cookies|Roast+Chestnuts|Chocolate|Various+Other+Beverages|Various+Other+Foods|Snacks&chtt=Food+and+Drink+Consumed+Christmas+2007&cht=p&chs=600x300&chd=s:KUIZFDPJF What does that give us? I’m glad you asked. I have the rather beautiful 600-pixel wide pie chart you see in Figure 3. Figure 3: A Christmassy pie chart with labels. But I don’t like pie charts The pie chart was invented by the Scottish polymath William Playfair in 1801. But not everyone is as excited by pie charts as wee Billy, so if you’re an anti-pie-chartist, what can you do? You can easily reuse the same data but display it as a bar graph in a snap. The first thing we need to do is change the value of the cht parameter from p to bhg. This creates a horizontal bar graph (you can request a vertical bar graph using bvg). The data and labels all remain the same, but we need to decide where the labels will appear. I’ll talk more about how to do all this in the next section. In Figure 4 you’ll see the newly-converted bar graph. The URL for the graph is: http://chart.apis.google.com/chart?cht=bhg&chs=600x300&chd=s:KUIZFDPJF&chxt=x,y&chtt=Food+and+Drink+Consumed+Christmas+2007&chxl=1:|Egg+nog|Christmas+Ham|Milk+(not+including+egg+nog)|Cookies|Roast+Chestnuts|Chocolate|Various+Other+Beverages|Various+Other+Foods|Snacks&chco=00AF33 Figure 4: The pie chart from Figure 3 represented as a bar chart. Two lines, one graph Pie charts and bar charts are interesting, but what if I want to compare last year’s Christmas cheer with this year’s? That sounds like I’ll need two lines on one graph. The code is much the same as the previous examples; the most obvious difference is I need to set up the chart as a line graph. Creating some dummy values for the required parameters, I end up with: The chs=800x300 sets the dimensions of the new chart, while cht=lxy describes the type of chart we are using (in this case a line chart with x and y co-ordinates). For the chart data I’m going to demostrate a different encoding, text encoding. To use this I start the value of the chd parameter with “t:” instead of “s:”, and follow it with a list of x coordinates, a vertical pipe, |, and a list of y coordinates. Given the URL above, Google Charts will render the chart shown in Figure 5. Figure 5: A simple line graph with x and y co-ordinates. To make this graph a little more pleasing to the eye, I can add much the same as I did to the pie chart. I’ll add a chart title. Maybe something like “Projected Christmas Cheer for 2007”. Just as before I would add a chtt parameter to the image URL: &chtt=Projected+Christmas+Cheer+for+2007 Next, let’s add some labels on the y axis to represent a scale from 0 to 100. On the x axis let’s label for the most important days of December. To do this I need to use the chart axis type parameter, chxt. This allows us to specify the axes and associate some labels with them. As I’m only interested in the y-axis (to the left of the chart) and the x-axis (below the chart), we add chxt=x,y to our image URL. Now I need my label data. This is slightly more tricky because I want the data evenly spaced without labelling every item. The parameter for labels is chxl, the chart axis label. You match a label to an axis by using a number. So 0:Label1 is the zero index of chxt — in this case the x-axis. 1:Label2 is the first index of chxt — the y-axis. The order of these parameters or labels doesn’t matter as long as you associate them to their chxt correctly. The next thing to know about chxl is that you can add an empty label. Labels are separated by vertical pipe; if you don’t put any text in a label, you just leave the two vertical pipes empty (“||”) and Google Charts will allocate space but no label. For our vertical y axis, we want to label only 50% and 100% on the graph and plot them in their respective places. Since the y-axis is the second item, 1: (remember to start counting at zero), we add ten spaces to our image URL, chxl=1:||||||50|||||100 This will output the 50 halfway and the 100 at the top; all the other spaces will be empty. We can do the same thing to get specific dates along the x-axis as well. Let’s add the 1st of December, St. Nick’s Day (the 6th), Christmas Day, Boxing Day (a holiday common in the UK and the Commonwealth, on the 26th), and the final day of the month, the 31st. Since this is the x-axis I’ll use 0: as a reference in the chxt parameter tell Google Charts which axis to label. In full, the chxl parameter now looks like: chxl=1:||||||50|||||100|0:|Dec+1st|||||6th||||10th|||||15th|||||20th|||||25th|26th|||||Dec+31st That’s pretty. Before we begin to graph our data, I’ll do one last thing: add some grid lines to the chart so to better connect the data to the labels. The parameter for this is chg, short for chart grid lines. The parameter takes four comma-separated arguments. The first is the x-axis spacing for the grid. I have thirty-one days, so I need thirty vertical lines. The chart is 100% wide, so 3.33 (100 divided by 30) is the required spacing. As for the y-axis: the axis goes up to 100% but we probably only need to have a horizontal line every 10%, so the required spacing is 10 (100 divided by 10). That is the second argument. The last two arguments control the dash-style of the grid-lines. The first number is the length of the line dash and the second is the space between the dashes. So 6,3 would mean a six-unit dash with a three-unit space. I like a ratio of 1,3 but you can change this as you wish. Now that I have the four arguments, the chg parameter looks like: chg=3.333,10,1,3 If I add that to the chart URL I end up with: http://chart.apis.google.com/chart?chs=800x300&cht=lxy&chd=t:0,100|0,100&chtt=Projected+Christmas+Cheer+for+2007&chxt=x,y&chxl=0:|Dec+1st|||||6th|||||||||||||||||||25th|26th|||||Dec+31st|1:||||||50|||||100&chg=3.3333,10,1,3 Which results in the chart shown in Figure 6. Figure 6: Chart ready to receive the Christmas cheer values. Real data Now the chart is ready I can add historical data from 2006 and current data from 2007. Having a look at last year’s cheer levels we find some highs and lows through-out the month: Dec 1st Advent starts; life is good 30% Dec 6th St. Nick’s Day, awake to find good things in my shoes 45% Dec 8th Went Christmas carolling, nearly froze 20% Dec 10th Christmas party at work, very nice dinner 50% Dec 18th Panic Christmas shopping, hate rude people 15% Dec 23rd Off Work, home eating holiday food 80% Dec 25th Opened presents, good year, but got socks again from Grandma 60% Dec 26th Boxing Day; we’re off and no one knows why 70% Dec 28th Third day of left overs 40% Dec 29th Procured some fireworks for new years 55% Dec 31st New Year’s Eve 80% Since I’m plotting data for 2006 and 2007 on the same graph I’ll need two different colours — one for each year’s line — and a key to denote what each colour represents. The key is controlled by the chdl (chart data legend) parameter. Again, each part of the parameter is separated by a vertical pipe, so for two labels I’ll use chdl=2006|2007. I also want to colour-code them, so I’ll need to add the chco as I did for the pie chart. I want a red line and a green line, so I’ll use chco=458B00,CD2626 and add this to the image URL. Let’s begin to plot the 2006 data on the Chart, replacing our dummy data of chd=t:0,100|0,100 with the correct information. The chd works by first listing all the x coordinates (each separated by a comma), then a vertical pipe, and then all the y coordinates (also comma-separated). The chart is 100% wide, so I need to convert the days into a percentage of the month. The 1st of December is 0 and the 31st is 100. Everything else is somewhere in between. Our formula is: (d – 1) × 100 ÷ (31 – 1) Where d is the day of the month. The formula states that each day will be printed every 3.333 units; so the 6th of December will be printed at 16.665 units. I can repeat the process for the other dates listed to get the following x coordinates: 0,16.7,23.3,33.3,60,76.7,83.3,86.7,93.3,96.7. The y axis coordinates are easy because our scale is 100%, just like our rating, so we can simply copy them across as 30,45,20,50,15,80,60,70,40,55,80. This gives us a final chd value of: chd=t:0,16.7,23.3,33.3,60,76.7,83.3,86.7,93.3,96.7,100|30,45,20,50,15,80,60,70,40,55,80 Onto 2007: I can put the data for the month so far to see how we are trending. Dec 1st Christmas shopping finished already 50% Dec 4th Computer hard disk drive crashed (not Christmas related accident, but put me in a bad mood) 10% Dec 6th Missed St. Nick’s Day completely due to travelling 30% Dec 9th Dinner with friends before they travel 55% Dec 11th 24ways article goes live 60% Using the same system we did for 2006, I can take the five data points and plot them on the chart. The new x axis values will be 0,10,16.7,26.7 and the new y axis 50,10,30,65. We incorporate those into the image URL by appending these values onto the chd parameter we already have, which then becomes: chd=t:0,16.7,23.3,33.3,60,76.7,83.3,86.7,93.3,96.7,100|30,45,20,50,15,80,60,70,40,55,80|0,10,16.7,26.7,33.3|50,10,30,55,60 Passing this to Google Charts results in Figure 7. http://chart.apis.google.com/chart?chs=800x300&cht=lxy&chd=t:0,100|0,100&chtt=Projected+Christmas+Cheer+for+2007&chxt=x,y&chxl=0:|Dec+1st|||||6th|||||||||||||||||||25th|26th|||||Dec+31st|1:||||||50|||||100&chg=3.3333,10,1,3&chd=t:0,16.7,23.3,33.3,60,76.7,83.3,86.7,93.3,96.7,100|30,45,20,50,15,80,60,70,40,55,80|0,10,16.7,26.7,33.3|50,10,30,55,60&chco=458B00,CD2626&chdl=2006|2007 Figure 7: Projected Christmas cheer for 2006 and 2007. Did someone mention Edward Tufte? Google Charts are a robust set of chart types that you can create easily and freely using their API. As you can see, you can graph just about anything you want using the line graph, bar charts, scatter plots, venn diagrams and pie charts. One type of chart conspicuously missing from the API is sparklines. Sparklines were proposed by Edward Tufte as “small, high resolution graphics embedded in a context of words, numbers, images”. They can be extremely useful, but can you create them in Google Charts? The answer is: “Yes, but it’s an undocumented feature”. (The usual disclaimer about undocumented features applies.) If we take our original line graph example, and change the value of the cht parameter from lc (line chart) to lfi (financial line chart) the axis-lines are removed. This allows you to make a chart — a sparkline — small enough to fit into a sentence. Google uses the lfi type all throughout the their financial site, but it’s not yet part of the official API. MerryChristmas http://chart.apis.google.com/chart?cht=lfi&chs=100x15&chd=s:MerryChristmas 24ways http://chart.apis.google.com/chart?cht=lfi&chs=100x15&chd=s:24ways&chco=999999 HappyHolidays http://chart.apis.google.com/chart?cht=lfi&chs=100x15&chd=s:HappyHolidays&chco=ff0000 HappyNewYear http://chart.apis.google.com/chart?cht=lfi&chs=100x15&chd=s:HappyNewYear&chco=0000ff Summary The new Google Charts API is a powerful method for creating charts and graphs of all types. If you apply a little bit of technical skill you can create pie charts, bar graphs, and even sparklines as and when you need them. Now you’ve finished ready the article I hope you waste no more time: go forth and chart! Further reading Google Charts API More on Google Charts How to handle negative numbers 12 Days of Christmas Pie Chart 1 In order to remain within the 50,000 requests a day limit the Google Charts API imposes, chart images on this page have been cached and are being served from our own servers. But the URLs work – try them!",2007,Brian Suda,briansuda,2007-12-11T00:00:00+00:00,https://24ways.org/2007/tracking-christmas-cheer-with-google-charts/,ux 159,How Media Studies Can Massage Your Message,"A young web designer once told his teacher ‘just get to the meat already.’ He was frustrated with what he called the ‘history lessons and name-dropping’ aspects of his formal college course. He just wanted to learn how to build Web sites, not examine the reasons why. Technique and theory are both integrated and necessary portions of a strong education. The student’s perspective has direct value, but also holds a distinct sorrow: Knowing the how without the why creates a serious problem. Without these surrounding insights we cannot tap into the influence of the history and evolved knowledge that came before. We cannot properly analyze, criticize, evaluate and innovate beyond the scope of technique. History holds the key to transcending former mistakes. Philosophy encourages us to look at different points of view. Studying media and social history empowers us as Web workers by bringing together myriad aspects of humanity in direct relation to the environment of society and technology. Having an understanding of media, communities, communication arts as well as logic, language and computer savvy are all core skills of the best of web designers in today’s medium. Controlling the Message ‘The computer can’t tell you the emotional story. It can give you the exact mathematical design, but what’s missing is the eyebrows.’ – Frank Zappa Media is meant to express an idea. The great media theorist Marshall McLuhan suggests that not only is media interesting because it’s about the expression of ideas, but that the media itself actually shapes the way a given idea is perceived. This is what McLuhan meant when he uttered those famous words: ‘The medium is the message.’ If instead of actually serving a steak to a vegetarian friend, what might a painting of the steak mean instead? Or a sculpture of a cow? Depending upon the form of media in question, the message is altered. Figure 1 Must we know the history of cows to appreciate the steak on our plate? Perhaps not, but if we begin to examine how that meat came to be on the plate, the social, cultural and ideological associations of that cow, we begin to understand the complexity of both medium and message. A piece of steak on my plate makes me happy. A vegetarian friend from India might disagree and even find that that serving her a steak was very insensitive. Takeaway: Getting the message right involves understanding that message in order to direct it to your audience accordingly. A Separate Piece If we revisit the student who only wants technique, while he might become extremely adept at the rendering of an idea, without an understanding of the medium, how is he to have greater control over how that idea is perceived? Ultimately, his creativity is limited and his perspective narrowed, and the teacher has done her student a disservice without challenging him, particularly in a scholastic environment, to think in liberal, creative and ultimately innovative terms. For many years, web pundits including myself have promoted the idea of separation as a core concept within creating effective front-end media for the Web. By this, we’ve meant literal separation of the technologies and documents: Markup for content; CSS for presentation; DOM Scripting for behavior. While the message of separation was an important part of understanding and teaching best practices for manageable, scalable sites, that separation is really just a separation of pieces, not of entire disciplines. For in fact, the medium of the Web is an integrated one. That means each part of the desired message must be supported by the media silos within a given site. The visual designer must study the color, space, shape and placement of visual objects (including type) into a meaningful expression. The underlying markup is ideally written as semantically as possible, promote the meaning of the content it describes. Any interaction and functionality must make the process of the medium support, not take away from, the meaning of the site or Web application. Examination: The Glass Bead Game Figure 2 Figure 2 shows two screenshots of CoreWave’s historic ‘Glass Bead Game.’ Fashioned after Herman Hesse’s novel of the same name, the game was an exploration of how ideas are connected to other ideas via multiple forms of media. It was created for the Web in 1996 using server-side randomization with .htmlx files in order to allow players to see how random associations are in fact not random at all. Takeaway: We can use the medium itself to explore creative ideas, to link us from one idea to the next, and to help us better express those ideas to our audiences. Computers and Human Interaction Since our medium involves computers and human interaction, it does us well to look to foundations of computers and reason. Not long ago I was chatting with Jared Spool on IM about this and that, and he asked me ‘So how do you feel about that?’ This caused me no end of laughter and I instantly quipped back ‘It’s okay by me ELIZA.’ We both enjoyed the joke, but then I tried to share it with another dare I say younger colleague, and the reference was lost. Raise your hand if you got the reference! Some of you will, but many people who come to the Web medium do not get the benefit of such historical references because we are not formally educated in them. Joseph Weizenbaum created the ELIZA program, which emulates a Rogerian Therapist, in 1966. It was an early study of computers and natural human language. I was a little over 2 years old, how about you? Conversation with Eliza There are fortunately a number of ELIZA emulators on the Web. I found one at http://www.chayden.net/eliza/Eliza.html that actually contains the source code (in Java) that makes up the ELIZA script. Figure 3 shows a screen shot of the interaction. ELIZA first welcomes me, says ‘Hello, How do you do. Please state your problem’ and we continue in a short loop of conversation, the computer using cues from my answers to create new questions and leading fragments of conversation. Figure 3 Albeit a very limited demonstration of how humans could interact with a computer in 1966, it’s amusing to play with now and compare it to something as richly interactive as the Microsoft Surface (Figure 4). Here, we see clear Lucite blocks that display projected video. Each side of the block has a different view of the video, so not only does one have to match up the images as they are moving, but do so in the proper directionality. Figure 4 Takeway: the better we know our environment, the more we can alter it to emulate, expand and even supersede our message. Leveraging Holiday Cheer Since most of us at least have a few days off for the holidays now that Christmas is upon us, now’s a perfect time to reflect on ones’ environment and examine the messages within it. Convince your spouse to find you a few audio books for stocking stuffers. Find interactive games to play with your kids and observe them, and yourself, during the interaction. Pour a nice egg-nog and sit down with a copy of Marshall McLuhan’s ‘The Medium is the Massage.’ Leverage that holiday cheer and here’s to a prosperous, interactive new year.",2007,Molly Holzschlag,mollyholzschlag,2007-12-22T00:00:00+00:00,https://24ways.org/2007/how-media-studies-can-massage-your-message/,ux 150,"A Gift Idea For Your Users: Respect, Yo","If, indeed, it is the thought that counts, maybe we should pledge to make more thoughtful design decisions. In addition to wowing people who use the Web sites we build with novel features, nuanced aesthetics and the new new thing, maybe we should also thread some subtle things throughout our work that let folks know: hey, I’m feeling ya. We’re simpatico. I hear you loud and clear. It’s not just holiday spirit that moves me to talk this way. As good as people are, we need more than the horizon of karma to overcome that invisible demon, inertia. Makers of the Web, respectful design practices aren’t just the right thing, they are good for business. Even if your site is the one and only place to get experience x, y or zed, you don’t rub someone’s face in it. You keep it free flowing, you honor the possible back and forth of a healthy transaction, you are Johnny Appleseed with the humane design cues. You make it clear that you are in it for the long haul. A peek back: Think back to what search (and strategy) was like before Google launched a super clean page with “I’m Feeling Lucky” button. Aggregation was the order of the day (just go back and review all the ‘strategic alliances’ that were announced daily.) Yet the GOOG comes along with this zen layout (nope, we’re not going to try to make you look at one of our media properties) and a bold, brash, teleport-me-straight-to-the-first-search-result button. It could have been titled “We’re Feeling Cocky”. These were radical design decisions that reset how people thought about search services. Oh, you mean I can just find what I want and get on with it? It’s maybe even more impressive today, after the GOOG has figured out how to monetize attention better than anyone. “I’m Feeling Lucky” is still there. No doubt, it costs the company millions. But by leaving a little money on the table, they keep the basic bargain they started to strike in 1997. We’re going to get you where you want to go as quickly as possible. Where are the places we might make the same kind of impact in our work? Here are a few ideas: Respect People’s Time As more services become more integrated with our lives, this will only become more important. How can you make it clear that you respect the time a user has granted you? User-Oriented Defaults Default design may be the psionic tool in your belt. Unseen, yet pow-er-ful. Look at your defaults. Who are they set up to benefit? Are you depending on users not checking off boxes so you can feel ok about sending them email they really don’t want? Are you depending on users not checking off boxes so you tilt privacy values in ways most beneficial for your SERPs? Are you making it a little too easy for 3rd party applications to run buckwild through your system? There’s being right and then there’s being awesome. Design to the spirit of the agreement and not the letter. See this? Make sure that’s really the experience you think people want. Whenever I see a service that defaults to not opting me in their newsletter just because I bought a t shirt from them, you can be sure that I trust them that much more. And they are likely to see me again. Reduce, Reuse It’s likely that people using your service will have data and profile credentials elsewhere. You should really think hard about how you can let them repurpose some of that work within your system. Can you let them reduce the number of logins/passwords they have to manage by supporting OpenID? Can you let them reuse profile information from another service by slurping in (or even subscribing) to hCards? Can you give them a leg up by reusing a friends list they make available to you? (Note: please avoid the anti-pattern of inviting your user to upload all her credential data from 3rd party sites into your system.) This is a much larger issue, and if you’d like to get involved, have a look at the wiki, and dive in. Make it simple to leave Oh, this drives me bonkers. Again, the more simple you make it to increase or decrease involvement in your site, or to just opt-out altogether, the better. This example from Basecamp is instructive: At a glance, I can see what the implications are of choosing a different type of account. I can also move between account levels with one click. Finally, I can cancel the service easily. No hoop jumping. Also, it should be simple for users to take data with them or delete it. Let Them Have Fun Don’t overlook opportunities for pleasure. Even the most mundane tasks can be made more enjoyable. Check out one of my favorite pieces of interaction design from this past year: Holy knob fiddling, Batman! What a great way to get people to play with preference settings: an equalizer metaphor. Those of a certain age will recall how fun it was to make patterns with your uncle’s stereo EQ. I think this is a delightful way to encourage people to optimize their own experience of the news feed feature. Given the killer nature of this feature, it was important for Facebook to make it easy to fine tune. I’d also point you to Flickr’s Talk Like A Pirate Day Easter egg as another example of design that delights. What a huge amount of work for a one-off, totally optional way to experience the site. And what fun. And how true to its brand persona. Brill. Anti-hype Don’t talk so much. Rather, ship and sample. Release code, tell the right users. See what happens. Make changes. Extend the circle a bit by showing some other folks. Repeat. The more you hype coming features, the more you talk about what isn’t yet, the more you build unrealistic expectations. Your genius can’t possibly match our collective dreaming. Disappointment is inevitable. Yet, if you craft the right melody and make it simple for people to hum your tune, it will spread. Give it time. Listen. Speak the Language of the Tribe It’s respectful to speak in a human way. Not that you have to get all zOMGWTFBBQ!!1 in your messaging. People respond when you speak to them in a way that sounds natural. Natural will, of course, vary according to context. Again, listen in and people will signal the speech that works in that group for those tasks. Reveal those cues in your interface work and you’ll have powerful proof that actual people are working on your Web site. This example of Pownce‘s gender selector is the kind of thing I’m talking about: Now, this doesn’t mean you should mimic the lingo on some cool kidz site. Your service doesn’t need to have a massage when it’s down. Think about what works for you and your tribe. Excellent advice here from Feedburner’s Dick Costolo on finding a voice for your service. Also, Mule Design’s Erika Hall has an excellent talk on improving your word fu. Pass the mic, yo Here is a crazy idea: you could ask your users what they want. Maybe you could even use this input to figure out what they really want. Tools abound. Comments, wikis, forums, surveys. Embed the sexy new Get Satisfaction widget and have a dynamic FAQ running. The point is that you make it clear to people that they have a means of shaping the service with you. And you must showcase in some way that you are listening, evaluating and taking action against some of that user input. You get my drift. There are any number of ways we can show respect to those who gift us with their time, data, feedback, attention, evangelism, money. Big things are in the offing. I can feel the love already.",2007,Brian Oberkirch,brianoberkirch,2007-12-23T00:00:00+00:00,https://24ways.org/2007/a-gift-idea-for-your-users-respect-yo/,ux 144,"The Mobile Web, Simplified","A note from the editors: although eye-opening in 2006, this article is no longer relevant to today’s mobile web. Considering a foray into mobile web development? Following are four things you need to know before making the leap. 1. 4 billion mobile subscribers expected by 2010 Fancy that. Coupled with the UN prediction of 6.8 billion humans by 2010, 4 billion mobile subscribers (source) is an astounding 59% of the planet. Just how many of those subscribers will have data plans and web-enabled phones is still in question, but inevitably this all means one thing for you and me: A ton of potential eyes to view our web content on a mobile device. 2. Context is king Your content is of little value to users if it ignores the context in which it is viewed. Consider how you access data on your mobile device. You might be holding a bottle of water or gripping a handle on the subway/tube. You’re probably seeking specific data such as directions or show times, rather than the plethora of data at your disposal via a desktop PC. The mobile web, a phrase often used to indicate “accessing the web on a mobile device”, is very much a context-, content-, and component-specific environment. Expressed in terms of your potential target audience, access to web content on a mobile device is largely influenced by surrounding circumstances and conditions, information relevant to being mobile, and the feature set of the device being used. Ask yourself, What is relevant to my users and the tasks, problems, and needs they may encounter while being mobile? Answer that question and you’ll be off to a great start. 3. WAP 2.0 is an XHTML environment In a nutshell, here are a few fundamental tenets of mobile internet technology: Wireless Application Protocol (WAP) is the protocol for enabling mobile access to internet content. Wireless Markup Language (WML) was the language of choice for WAP 1.0. Nearly all devices sold today are WAP 2.0 devices. With the introduction of WAP 2.0, XHTML Mobile Profile (XHTML-MP) became the preferred markup language. XHTML-MP will be familiar to anyone experienced with XHTML Transitional or Strict. Summary? The mobile web is rapidly becoming an XHTML environment, and thus you and I can apply our existing “desktop web” skills to understand how to develop content for it. With WML on the decline, the learning curve is much smaller today than it was several years ago. I’m generalizing things gratuitously, but the point remains: Get off yo’ lazy butt and begin to take mobile seriously. I’ll even pass you a few tips for getting started. First, the DOCTYPE for XHTML-MP is as follows: As for MIME type, Open Mobile Alliance (OMA) specifies using the MIME type application/vnd.wap.xhtml+xml, but ultimately you need to ensure the server delivering your mobile content is configured properly for the MIME type you choose to use, as there are other options (see Setting up WAP Servers). Once you’ve made it to the body, the XHTML-MP markup is not unlike what you’re already used to. A few resources worth skimming: Developers Home XHTML-MP Tutorial – An impressively replete resource for all things XHTML-MP XHTML-MP Tags List – A complete list of XHTML-MP elements and accompanying attributes And last but certainly not least, CSS. There exists WAP CSS, which is essentially a subset of CSS2 with WAP-specific extensions. For all intents and purposes, much of the CSS you’re already comfortable using will be transferrable to mobile. As for including CSS in your pages, your options are the same as for desktop sites: external, embedded, and inline. Some experts will argue embedded or inline over external in favor of reducing the number of HTTP connections per page request, yet many popular mobilized sites and apps employ external linking without issue. Stocking stuffers: Flickr Mobile, Fandango Mobile, and Popurls Mobile. A few sites with whom you can do the View Source song and dance for further study. 4. “Cell phone” is so DynaTAC If you’re a U.S. resident, listen up: You must rid your vocabulary of the term “cell phone”. We’re one of the few economies on the planet to refer to a mobile phone accordingly. If you care to find yourself in any of the worthwhile mobile development circles, begin using terms more widely accepted: “mobile” or “mobile phone” or “handset” or “handy”. If you’re not sure which, go for “mobile”. Such as, “Yo dog, check out my new mobile.” More importantly, however, is overcoming the mentality that access to the mobile web can be done only with a phone. Instead, “device” encourages us to think phone, handheld computer, watch, Nintendo DS, car, you name it. Simple enough?",2006,Cameron Moll,cameronmoll,2006-12-19T00:00:00+00:00,https://24ways.org/2006/the-mobile-web-simplified/,ux 142,Revealing Relationships Can Be Good Form,"A few days ago, a colleague of mine – someone I have known for several years, who has been doing web design for several years and harks back from the early days of ZDNet – was running through a prototype I had put together for some user testing. As with a lot of prototypes, there was an element of ‘smoke and mirrors’ to make things look like they were working. One part of the form included a yes/no radio button, and selecting the Yes option would, in the real and final version of the form, reveal some extra content. Rather than put too much JavaScript in the prototype, I took a preverbial shortcut and created a link which I wrapped around the text next to the radio button – clicking on that link would cause the form to mimic a change event on the radio button. But it wasn’t working for him. Why was that? Because whereas I created the form using a