Things what I writ

I sometimes write nonsense about things to try and sound clever

Archives
Categories

Finding Servers

Hot, well, warm on the heels of our storage finder, we have recently deployed the new server finder, which replaced a previous incantation of finding functionality which was held together by hello world string and was somewhat creaking at the edges. The new server finder is based on the architecture we developed for the storage finder – they are, in fact, 2 renditions of the same finding platform – and so leverages the features that make it so worth the investment of effort.

As with the storage deployment, the key to making the server finder successful was, well, a number of things, but the main thrust of our efforts was defining the data architecture that make the relationships between product groups, products and product attributes a meaningful one. This can only happen with some Herculean efforts being undertaken by our publishing teams in conjunction with the product marketing teams, who really understand what is important and relevant about the products they market. Really, the finder itself is just a layer of abstraction on top of the data set underneath, and in theory (as we are at pains to try and progress), can be applied to any well-structured data set. What matters, is whether the data that a customer, user, or casual visitor is presented with, and the methods they can use to interrogate that data, enables them to reach an appropriate destination. In other words, they might know where they want to go, they might have a vague idea, or they may have no idea at all, but if we’ve done our job as well as we should be doing it, the directed searches and filters that the finding platform utilizes should provide a the equivalent of a product sat-nav, but avoid the 18-wheelers that get grounded on hump-back bridges in the middle of Hertfordshire on the way to the new Tesco Express.

Probably an analogy too far there, but it is by way of illustrating that the key to the finding platform is the data that it manipulates. I mean, we did a number of detailed usability trails, with various rapid and high-fidelity prototypes, struggled over the tiniest nuances of labels and gradients, fought compromise on page region refreshes and a followed number of other noteworthy user experience best practices, but in the end, if we built our application infrastructure on top of a taxonomy akin to a river bed full of shopping trolleys, we’d only be providing half a solution, which, in fact, is no solution at all.

We’ve still got a number of things to work on that didn’t make it into the first release, such as enabling product comparisons across products and, more difficult, across product in different product families, but take a look for yourself and let us know what you think. Comments are more than welcome, especially ones that are nice.

use it, or not

Hang on, is this still active? Oh dear.

It occurred to me while reading Paul Boag‘s 10 things a web designer would never tell you, that there are a number of things I’ve never told you. They were mainly the things in his list, but accompanied by a cultural caveat that stated I’m English, specifically, from dead-pan capital of the east, Norwich, so anything I might write down here that might apparently be totally, like, lame, I’ve actually written post-modern ironically, which means that any statement of fact that I make that you consider simply ridiculous, is, in fact, a joke. Its just that I deliver it straight. Which is difficult enough when I deliver it face-to-face, but when it’s rendered in a <div>, then it often goes horribly wrong.

Suffice to say that on reading Paul’s list, I’m particularly struck by number 3. Not because I necessarily subscribe to the argument that user testing is an expensive conspiracy prolapsed by some clandestine web design coven in order to prolong delivery (you see how straight I’m delivering that), but because I really dislike useit.com. Of course its probably the most well-respected usability site of all and has mucho gravitas amongst the online design community (he even knows where my eyes are looking. right now!), but, oh, I just can’t look at it. Its not so much that there is just so much valuable, rich, meaningful content there on that home page that it makes me hyperventilate at the thought of actually having to read some of it, or even the fact that when I do get to read it, I’m mildly troubled by it being more statement of fact than informed conjecture. Its worse than that. I don’t like yellow. And that high frequency colour spectrum opposite polarity yellow-blue split-screen thing makes my eyes go all stereogrammatical and gives me a headache. Oh, and there’s no pictures. That alone makes my healing brush finger twitch. Am I wrong?

That is, of course, just something that affects me. I think. Thankfully, the team that delivers the sun.com experience is a very broad church and amongst us walk many much more professional and qualified experts who actually understand, subscribe to, and put into practice the very things that he who shall be called Jakob Nielsen imparts. That’s one of the reasons why sun.com conistently ranks so highly in independent usability evaluations (from folks such as SiteIQ) – because we engage early with stakeholders, customers and our extended communites to get that critical knowledge that keeps us informed. We also continue those relationships with ongoing studies, surveys and evaluations, so that we don’t get complacent. Its easy to evaluate usability on a specific project while you’re in the design and build phase, and then just forget all about it once you move to deploy and maintain. You’ve already tested it right? You don’t need to test it again after it goes live, surely? We’re guilty of that on sun.com, and its one of the things we’re trying to focus on right now. So if we ask you for feedback, or drop you the occasional survey when you’re visiting sun.com, we’d be delighted if you’d spare the time to let us know what we’re doing right or wrong. It doesn’t take long, and hopefully it will help us to help you.

That last bit wasn’t a joke, by the way.

Listening Post: Ladyhawke: My Delirium

Finding Storage

Sounds like it should be a film with Tom Hanks and an emotionally challenged Tupperware box. It is, in fact, the long-awaited solution to one of our common web problems. Whether you call it filtered searching, directed searching, product finding, trans-navigational learning aid cognitive process map hierarchical cross-sell or something, it’s about trying to find the right product for your business. And we’ve just launched it on sun.com.

The new storage finder is built from the ground up with the intention of enabling customers to find the right products for them, based on their unique requirements. We’ve tried this before, you may have noticed, with mixed results. One of the problems we’ve previously encountered is trying to architect a finding solution that’s based on the interaction model alone, rather than really understanding what is important to our customers and how those key criteria drive the user experience. To avoid repeating those mistakes, for the new storage finder, we took a significant step backwards, to understand the product taxonomy and how it maps to business needs and customer expectations. When reviewing the product data, and testing with business groups and customers, it was clear that what seems like an important attribute of a product or product family is not necessarily what matters to the people who are actually wanting to buy it. Seems obvious, but until you get real people to give you real opinions, then you’re just guessing.

After investing such much effort in evaluating the product data and determining what really turns folk on about storage (it does happen), we were in a much better position to look at the interaction model and the representation of the data on sun.com. I mean, we already knew that driving customers down a one-way street with road signs that only the product marketing team can read is a pretty hopeless exercise, but there was still a lot of decision making and testing to be done around the entry points to the customer journey, the complexity of the options (parabolic vs. optional), and the level of detail required to enable a decision to be made. Oh, and whether the Ajax thing would work.

I won’t bore you with the iterations of prototypes, usability testing, data refining, back-end systems, publishing frameworks and specifications that need to collide gracefully in order to get a project like this out of the door, but, suffice to say, a number of dedicated, hard-working folks from across Sun managed to pull this one out of the bag just in time for Christmas, so enjoy. There’s still a shopping day left, by the way…

We’ve a list of enhancements and future work that we’re already planning, but let us know what you think so that we can involve you in the ongoing development of our finding capability on sun.com

Listening Post: Dananananaykroyd: Pink Sabbath

Got Server Content?

You do? Where shall we put that then? No, I mean where shall we put it so people can actually see it?

There’s a ton of great stuff out there about Sun products, and it changes all the time. The trouble is, on sun.com, we sometimes don’t keep up with all the new and updated content out there. This is because we’ve often not really had a good place to surface it on our traditional product landing pages. Think servers, or storage, or software. Those product areas have their own discrete content areas on sun.com, where you might expect a reasonable refresh rate. In particular, the Overview pages in those product areas – the pages you hit at /servers, /storage, /software – should probably be the stickiest pages we can build, with constantly refreshed content. It’s always nice to see new content when you go back to a page. It gives the impression it might actually be current.

Up until very recently, the servers landing page on sun.com wasn’t really a landing page at all. You just landed quite unceremoniously at a server finder, where you were kind of expected to fend for yourself. Fine, of course, if you know what you’re looking for, or if you have some sense of the kind of product attributes that make up the ideal server for your particular business needs. Not so great if you don’t even know what a server is, or does. Or maybe you just want to know how Sun servers can help you, before you actually have to start choosing one. All the kind of stuff we loosely describe as content ‘which tells the product story’. You know, delivering key messages, addressing market sectors, providing system solutions, all that kind of stuff.

A few weeks ago, we put together a servers overview page, so that we could do that story telling, provide sensible paths into product areas, uplevel featured products, show off some great customer success stories, and, yes, tell you what our servers actually are. It’s a delicate balance on these pages between getting the story out there and providing a quick route to the products, but I think we managed it pretty well. I say ‘we’, but, of course, it was the good folks in the product marketing teams that pulled all the content together (kudos Carlos & Lisa), and our publishing team that managed the tricky icky problem of integrating the new content with the existing server finder (heroics from Jing). I just did the bit where I say ‘you’d be better of with a PC00 component there’.

While we were working that project, there was another altogether more dynamic project going on in the design room next door (there’s not a really a design room next door to me, but you know what I mean). A few months ago, the systems group here at Sun, that looks after the server product line, had an idea that they wanted to explore. It was really about addressing the problem I mentioned at the start – there’s great, current content out there, that has marketing dollars behind it, and a plan to develop it, but not a really great place to showcase it. Based on the kind of presentation we use for the product launch events on sun.com, they wanted to see what we could do to support their idea of ‘content channels’. A little bit launch, a little bit back story, a little bit promotion, a whole lot more interesting than a big top banner.

The result is what you now see on the top of the new servers overview page. A rather nice mix of videos, podcasts, product tours, white papers and other supercool server stories (those product tours are very nice. I took the PSU out of a Sun Fire X4140 just now). So now, when you come back to the servers section on sun.com, you can expect regular updates, announcements, product walkthroughs and all that jazz – all hand-picked by your server channel content owners. If you can’t hear it, that’s the sound of a gauntlet dropping to the floor of Menlo Park, by the way…

If its not immediately apparent, this is a product category landing page without right-hand navigation. Well, I’m excited.

Listening Post: Psychedelic Furs: All Of This and Nothing

Sharing Your Opinion

I just had a call with Ben, Mr Usability, regarding some work we need to collaborate on as part of our web feedback program. We’ve been looking into the user experience across our feedback systems on various venues and trying to simplify and standardize a number of the interactions. This effort has been really quite specific in focus for sun.com, based on the nature of how we gather feedback there through our contact forms, but we really do a whole lot more than just ask you to point out broken links and typos.

You may have noticed that we’ve rolled out the ‘floating math’ feedback widget across sun.com. In fact, the widget, in various formats, is rolled out across a wide range of Sun web venues and is gathering mightily useful data from those sites. Well, aside from the comments about how we suck particular parts of primates anatomy, of course, but, in general, specific, constructive and informative.

The whole thing is powered by lovely people at OpinionLab, and I was lucky enough to have Ben walk me through the administration interface to give me a better understanding of the capabilities of their templated comment card system and the deployment of widgets and embedded components. There was a time when we would take a look at a system like this, kind of like it, and then build our own. On Solaris. Using vi. Thankfully, we’re much more ready these days to let folks who really know what they’re doing provide these services (yes, I know we have to pay), and work out how they interconnect and communicate with our own systems. In the case of OpinionLab, is seems this is an exercise that they are more than happy to work with us on to get right, which is good, because now they’ll have to work with me to try and get it right, which is a user experience I can’t possibly comment on.

Listening Post: Bloc Party: Atonement

We Sell Servers

You know that, of course, but how do you buy our servers? For as long as I can remember, and in line with how we structure our organization, we’ve presented our product lines on the web by the product categories by which we refer to them. This means that if you’re looking for our servers on sun.com, we think you might want to look for them by their parent category. Right now, we’d be in a great position to answer customer questions like “What CoolThreads servers have you got?”, or “Show me all your blades”, but, really, is that the kind of question you have in your head when you come to sun.com to look at servers?

Maybe you’d actually prefer to see our servers presented in terms of their attributes, so that you can begin your research by asking “What servers have you got that can run Linux?”, or maybe “I’ve got $5000 and I want a Sun server now. Show me what you’ve got”. In any case, you’d be hard pressed right now to complete a customer journey like that without going through a number of hoops. Backwards, probably.

So, at the moment, we’re looking at what’s important to our customers in terms of the way that they look for our products and how they might expect to see them grouped, or otherwise, so that a subset of products is a meaningful subset of products, that can support directed searching, categorization and a much more targeted presentation model. I mean, do you really need to know everything about why our products are so great when you’ve already come to sun.com to find the products? Is that product category landing page just telling you a bit more than you need to know, when all you really want to do is find the products? Perhaps, in actual fact, you don’t know what you’re looking for and you do need help in understanding just what Sun servers there are and how they are differentiated from the competition. Either way, we want to try and support those interactions as efficiently as possible and, from a user experience perspective, make it a pleasure to be engaging with us.

We have great people in the team conducting user evaluations and interviews and gathering as much data as we can in order to direct our designs, but, you know, you might have something to say about your experiences on sun.com and what you really want to be able to do when you’re researching our products. If you do, let me know, and we’ll feed it directly into the design process. If you don’t want to comment here, you can always email – my name is Tim Caynes and I work at sun.com, so the address isn’t difficult to fathom.

Listening Post: Future Radio Online

Ad Server Finger Drumming

It is quite possibly a consequence of my patience becoming inversely proportional to my age, but recently, waiting for ad servers to respond in order to complete loading a page is really ticking me off. I’m not bothered about about ads which take a while to load while I’m actually reading the page I requested, but what really gets my fingers drumming on the desk and puts my laser mouse in imminent danger of being crashed unceremoniously against the woodwork with accompanying cries of “c’mon! C’MON-AH!”, is ad server code that halts a page load mid-stream until its finished its business. I’m sure the page owners have bought into the most efficient geo-located edge-based web service out there, so why is it increasingly the case that while pages get faster, ad servers seem to get slower? Perhaps it’s a deliberate interaction feature, I mean, nothing grabs your attention more than a broken page, but from a customer experience point of view, I don’t think that’s a journey I would normally care to continue with.

I’m aware that we deploy our own ad server across sun.com, and that’s not always bulletproof, but, as you might imagine, I look at as many sun.com pages as any other commercial/consumer sites, and I never have noticeable ad server lag on sun.com. I’m not exactly co-located with the sun.com servers either, being on the free internet in the UK, so I don’t get any special treatment. Maybe because we own the deployment of our own ad server, we’re in a much better position to monitor performance and make adjustments – I can’t pretend to understand the technology behind it (well, ok, I can) – whereas, as is the case for any web service you buy into, if you get your ads delivered by a 3rd party, you can’t do much about the external reference issues. That’s been true of any page you care to publish since html 1.0 – once you include external references as core components of your page, you’re really asking for trouble, notwithstanding any service level agreements you might have in place (and they’re always great, right?).

Even as I write this, I’m looking at Facebook and waiting for a hair loss ad to appear in the left-hand navigation. It doesn’t actually break the rendering, but it does annoy me all the same – the delays, not because it’s targeted me for hair loss products. Although, that is pretty annoying

Listening Post: Spiral Vertigo: What I’d Really Like To Say

The Return of the Design Comic

They’ve never really been away, but there’s a number of places I’ve been recently where they’d tell the story just perfectly, so I recently dug out all the old slides I had, and got any stuff I was missing from Martin’s site, and I’m looking at running some scenarios past people, with the comic treatment.

There’s no simpler way to get the message across when you’re trying to highlight a particular use case and they’re a great, self-documenting way to describe a unique customer journey. More often than not, because they’re particularly good for delivering bad news, I pull together all the slides with the really scary close-ups of disgruntled customers’ faces, and add suitably appalled call-outs, to make a really heavy-handed point, but, hey, that’s ok, as long as you put a joke in, right? Those ones are generally reserved for ‘problem’ scenarios, where we know there’s something wrong, but clickthrough and omniture data doesn’t always describe the user experience. Its a kind of ‘once more with feeling’ approach to describing a problem. To prove something’s not working isn’t always enough, you have to be able to show what it means to a customer as a result, and the way I’m doing that is with the faces of customers looking, well, pissed off annoyed.

They’re not just for bad news though. Most of the characterizations are at the delighted end of the scale, verging on the ecstatic in some cases (that would be for something like the super download speed on the improved docs.sun.com or something), all the way through to Dr Spock puzzlement (not finding products on a product gateway). Some of my favorite artifacts are the customer scenes, such as the ‘overhead typing’ view, or the ‘yes, I’m still in the office at this time’ view. My very favorite, however, is the ‘cubicle farm’, which, even after working from home for 4 years, makes me twitch a little and look over my shoulder when I see it.

If I come up with anything remotely entertaining, which isn’t entertaining because I’m highlighting some disasterous product portfolio deployment or something, then I’ll share it here. Until then, I’ll just post the usual meaningless kind of nonsense.

Listening Post: Add N to (X): Barry 7’s Contraption

You Know, Like CNET

Before you even get to the point where you ask ‘what is your content?’, there’s an apparent understanding that you need to work out how it surfaces all over your site. Since the very early days of sun.com, one of the biggest goals, as far as maintaining a healthy visitor profile goes, is just how to make things sticky. I’m not talking sticky as in the stuff that makes you go eeuw, but sticky like the invisible elastic brain rubber that compels you, against the gravity of your free will, to revisit those places online that have already visited. It’s the same reason you go back to Fry’s every so often, just to see if there’s any new technology stuff to dribble over, or why you ping last.fm or iTunes to keep up with released, related, and recommended. It might also be the reason you visit Gap every Friday lunchtime – you’re just checking it out to see what’s new.

But how do you know what’s new and where do you expect to find that out? When you’re looking at something the scale of sun.com and trying to determine customer behaviours for a given page type, it’s not alway a simple task to predict. You might be the kind of visitor who would casually visit the sun.com home page and, not unreasonably, expect to see anything newsworthy enough, that you might be compelled to actually invest time in, to be present right there. You might be more specific than that. You might be the CTO for an SMB or some other suitable market research defined acronym pairing, in which case, you’d probably know that we’ve got a place just for you, where you’d expect announcements, deep-dives and news to appear, relevant to your needs. You might even have a large propeller sticking out of your head and be interested only in what’s going on with Sun Virtual Desktop Infrastructure and how that relates to your development requirements for your linear accellerator or something. Either way, when we’ve got news for you, we want you to find it. And we want you to come back again. And again. And again.

So that’s why we’re currently investigating new approaches to surfacing the bestest, most currentest, content around, that’s relevant to you, in a way that’s going to make you want to come back often, but not take all day to consume when you’re engaging with us. One of the ideas we’re floating around (or select another flagpole/envelope/conceptualization buzzword bingo term of your own there) is content channels. You know, like CNET. We could funnel these content streams into various containers on product pages, gateways, category pages, etc., so that what’s most relevant to you is right there, where you want it, on-demand, so to speak. In terms of web design, this a quite a nice proposal, as we can have the content live elsewhere and suck it through a virtual ‘news pipe’, which spits it into, for instance, the servers container. Which would probably be quite sticky. Of course, someone, somewhere, needs to be owning, managing, publishing and maintaining the channels, but on the assumption that that would be possible, then a modular approach to deploying those channels where it makes most sense would be, um, neat.

Listening Post: The Who: I Can See For Miles

Project Overlap

I know you just love it when you find out your project overlaps with about 4 other projects doing kind of the same thing, but from a different place. That just happens in large-scale organizations, however we arrange ourselves and whatever processes we try and stick to. So when you gracefully collide with the business teams, the publishing and engineering teams and at least 1 other team you didn’t actually know existed until this morning, in a conference call that gathers all the stakeholders, it nice to get a good outcome.

We’re currently taking a deep dive, or whatever you call it, into the design framework we need in order to support the content architecture around product lines. In other words, if you happen to be the director in charge of, say, server marketing here at Sun, what is it that sun.com needs to do for you? I mean, we know a bunch of stuff about what people are actually doing when they hit those landing pages (we’re calling then category pages, for the record), but what is it that we’re wanting them to do, and from where did they enter, and to where are they going? Its all very well me just drawing a fancier looking media panel and assuming that we know what’s going to play there, or even if it should be a media panel at all. I can use terms like ‘customer channel’ as if I know what they mean, but in the end, as designers, we’re trying to understand the customer journey, in order for us to determine navigation paths and build a design framework that works for everyone.

Which is where collisions are helpful. As long as you have super efficient people around you to pull those overlapping projects together (designers don’t really do that kind of stuff very well), you might just strike it lucky and start the conversation at the point where you’re all saying “well, that’s kind of what we’re trying to do”. And that’s what happened this week, which made everything fit together way more neatly than it did last week. I finally get to the point where I know what’s required, we’re engaged with the stakeholders, and we’re all talking the same language.

Typically, I’m on vacation all next week, so I’ll forgotten it all by the time I get back (only joking).

Listening Post: The Wombats: Moving To New York

Archives
Categories

Share

Share on twitter
Share on facebook
Share on linkedin