Design for Web 2.0
Jason Fried, Jeff Veen, and I did a workshop yesterday on Design for Web 2.0. In preparing for the informal chat we had among ourselves and with the audience, we prepared a list of questions to consider. There’s about 15 of them, presented here unedited without context or answers:
- Right now, Web design feels like talking to the del.icio.us API and blending Flickr RSS with Upcoming iCal subscriptions. What happens when design(ers) has little to do with what’s on the page?
- Blogs democratized publishing, now “tags” could be considered to democratize information architecture. What’s behind this? Are powerful tools in the hands of millions really better than well-trained experts?
- How do we justify the high upfront costs of doing user research? Is there a magic bullet formula that will tell us if it’s worth it?
- I love quick wins. Find something you can fix in two weeks. Measure how it works now. Fix it. See how the numbers change. Repeat until you run out of stuff. Why is this so shocking to corporate Web sites?
- It feels to me that IT departments still operate under the assumption that technology is a precious resource that should be guarded carefully and trickled out. This is like a pair of handcuffs to most Web teams. Why do so many enterprises treat their Web sites like shrink-wrapped software and not publications?
- Can usability drive innovation? What’s the balance between giving the user what they need but also giving them what they do not yet know that they need?
- How do you go about designing for groups? When the “user” is a collection of people rather than a single person?
- Do brochureware sites still have a place on today’s Web?
- What does user experience mean in the context of cross-media services? How do you keep it consistent when you’re using T-Mobile’s interface to email your photos to Flickr or updating your blog with your TiVo using your Blackberry as an input device?
- home pages -> sites -> “posts” -> ????
- Q for Veen re: your content management is a process, not a software package mantra. Is there a lesson here for software in general?
- How would you design a web-based application differently today than 3 years ago? What do we have in our design war chests today that is capable of making the experience feel 3-years more mature?
- Do you think design “talk” is too focused on technological achievement (“Look mom, no tables!”) these days, or is it a step in the right direction?
- Should one design fit all? Should designers worry about their web designs working on alternate devices by default, or should each device have its own unique design?
- What is your feeling on Personas? Do they really help drive the visual design process or are they just process for process sake? What does it really mean to know your audience might be represented by a 30-something single female who likes to watch Friends, prefers paying her bills by mail, gets coffee every morning at Starbucks, and has a 56K connection at home?
- Who should we follow into the Web 2.0? What are some of the best examples of interaction design today?
I’ve opened the comments if you’d like to discuss any of these questions amongst yourselves.
Reader comments
MattOct 06, 2004 at 2:50PM
Lots of great, tough questions that no one seems to be asking or answering. Let me just grab one that I've been thinking about:
How do we justify the high upfront costs of doing user research? Is there a magic bullet formula that will tell us if it's worth it?
For commercial enterprises, it's easy to go with ROI here. If we test a dozen folks and find a number of problems which we fix and make 2x the money we used to, hopefully that more than pays for the research costs.
For a non-profit, it's a bit tougher, but it's still possible to find some metrics that you can show improvement on, post-research phase.
Personally, after doing user research and testing for the first time on one of my projects, the information gathered was insanely valuable, even if a metric or ROI couldn't be proven. No matter how perfect you think your WidgetCo site is, testing among regular folks will show you at least half a dozen major bugs you overlooked. They're often low-hanging, easy wins too.
A HunnyOct 06, 2004 at 2:56PM
I think personas are being oversold, but like archetypes they have some use in suggesting approximate boundaries. The 30-something you mention probably has certain parameters of taste (grunge unlikely to sell), tempo (may not be as fast-paced as gamers, nor as slow as seniors), content (a few lightweight salaams to the idea of ambition, glamor/prestige/sexiness, relief from stress...) The preference of paying bills by mail and a 56K modem are very eloquent locators in the user landscape.
Personae need to have porous boundaries. I'm 60&female, and run screaming from the BoomerHip of the AARP and other memes. Persona developers should have refined focus groups, not make up a persona and stereotype her/him.
The persona process may be most useful to get geeks out of generalizing from themselves. Until you study the Myers Briggs and lots of other personality style stuff, you won't believe how many "Doesn't everyone?" perspectives there are. All different
Skott KlebeOct 06, 2004 at 3:17PM
...Why is this so shocking to corporate Web sites?
Because of the number of stakeholders involved. Theoretically, the site belongs to Marketing, but Sales wants to make sure that it speaks to customer X, IT is licking its chops to switch from tables to CSS, and the CEO hates orange and certain color combinations he can't describe but knows them when he sees them.
If anyone's unhappy when it goes live, only IT and/or the designer will be blamed, so the presumptively guilty parties document and gain signatures in advance to make sure that there are plenty of additional targets.
If you have an IT department do the work, the work will be a precious resource - IT departments are all cost in most organizations.
Richard MacManusOct 06, 2004 at 3:47PM
Re: What happens when design(ers) has little to do with what's on the page?
This is a fascinating question and it reminds me of a recent Tim Berners-Lee interview, where he talked about how the Semantic Web is all about re-using information. Yes I know TBL always talks about SemWeb, but there were some gem quotes in this one. eg:
"The Semantic Web is just the application of weblike design to data; it will be many more decades before we will be able to say we have really implemented the Web idea in the full, if ever we can."
(emphasis mine)
As I wrote a week or so ago about that: Nowadays it's not just about designing a beautiful website, it's about designing for re-use of information. In a way, that's what people are already doing with RSS - designing with data.
Actually the really interesting "data design" is happening with Atom right now - Matt Webb and friends.
This is how I interpreted your question anyway, hope it's along the lines you will discuss at Web 2.0. And I look forward to your write-up of the results of that workshop, for those of us stuck on the other side of the world :-(
Emily PetrickOct 06, 2004 at 3:55PM
Who should we follow into the Web 2.0? What are some of the best examples of interaction design today?
Flickr. This is a no-brainer. They are smart, have a sense of humor, innovate NEARLY EVERY DAY ( I always see new subtle changes that they don't even publicize), and have a product that just works better than the rest.
They are not afraid!
This addresses a number of the discussion points actually.
Edward VielmettiOct 06, 2004 at 4:15PM
Given a design cycle of 2 weeks, what can we predict about the state of the net in 3 years? (75 turns of the design / implement crank)
BJOct 06, 2004 at 6:31PM
Who should we follow into the Web 2.0? What are some of the best examples of interaction design today?
Definitely Flickr and Basecamp. Both smart and fast moving companies that, like Emily says, are innovating at breakneck speed. They are also leverage things like RSS in useful ways and not just in generic ways.
jeremiasOct 07, 2004 at 7:36AM
- Do brochureware sites still have a place on today's Web?
All great questions, but I was just thinking of this the other day. Are we at a point where your "average" designer *must* bring some sort of interactivity to a site? Brochureware sites are static and have to be updated by the designer or someone who understands the tech.
I would say we're almost there. A designer these days should understand that customers want to be able to easily modify their own sites without having to go back to the creator. Thats been true for some large companies for a while, but now even your mom and pop small businesses will be expecting this and designers need to be able to respond.
NickOct 07, 2004 at 9:13AM
All good questions — but I really like how you've settled (purposefully?) on the idea of "website as process" as addressed by questions like "Q for Veen re: your content management is a process, not a software package mantra. Is there a lesson here for software in general?", "Find something you can fix in two weeks... Repeat until you run out of stuff. ", and "What's the balance between giving the user what they need but also giving them what they do not yet know that they need?".
The best sites are those designed as verbs instead of nouns. That's where I think we're going in the next 3-years (and have been going with blogs / content management systems / etc), towards a web of processes.
Joshua PorterOct 07, 2004 at 9:49AM
Can usability drive innovation? I doubt it, because that's not really the role of usability. Usability, being only a measurement of how easy something is to use, can only help us establish benchmarks from which we can decide whether or not it something is "good enough".
If something isn't "good enough", we set designers on the problem. They'll either incrementally make it better, or innovate and really change the paradigm. Usability may have a hand in all this (and may be seen as a driver), but it's the designers who are doing the innovating.
Steven SmithOct 07, 2004 at 10:07AM
Are powerful tools in the hands of millions really better than well-trained experts?
I'd be interested to hear about the discussion (if any) surrounding the democratization of design and production. I have a relatively small and modest skill set and a sort of ramshackle "learn-by-doing" method in my approach to building my website. With that caveat in mind, I'm not really trying to leverage myself as a designer. While the vanguard of standards and pristine design do interest me, I think it's all outside of my wheelhouse for the time being.
I'm conflicted about where to come down on the question of "more=better". One peek at the 'Web' section of craigslist shows you that 1 in 2 people fancy themselves a pro web designer (in so much as they're asking to be paid for their services). The logical end of this reality, in my opinion, is a marginalization of professional and qualified web designers. I'm all for a stronger set of tools that'll raise all boats in terms of the quality of design. But I hope that it won't mark the end of the sort of design that makes me a little jealous and the use of white space and standards that give me naughty thoughts.
philOct 07, 2004 at 5:00PM
that list of questions--Answered--sounds like a great book.
Do brochureware sites still have a place on today's Web?
I would like to say no, but so many clients don't understand the potential, and they are unwilling to really make anything but a brochure site.
paulOct 08, 2004 at 12:18PM
Do you think design "talk" is too focused on technological achievement ("Look mom, no tables!") these days, or is it a step in the right direction?
No. Design should always be the use of available tools to achieve a desired effect or look, not an ever-changing cluster fudge of hacks and leaky theories. It befuddles me that so-called "designers" blogs contain so many pops and whistles instead of any kind of art theory.
It occurs to me that graphic design, basically the idea of creating an easily reproduced, mass-oriented media has always been mired by technical hang-ups due to ever-changing processes. This is no different in the case of the web, in fact, it's accelerated. It seems every day since, oh, about 1995 there's another way to publish a web page, although standards should change all this once integrated, but I guarantee you when that day comes there will be a whole new way to design for the web, and Zeldman will be whoring it's potential.
History is chock full of people that tell you that "The new way is the one and only right way to do everything for now and forever, and I can show you for a price." How many of these people, throughout the course of history, have been right? Not many - atoms are not the smallest particle, elements can be further broken down, evolution is not necessarily correct, the earth, in fact, is not flat. Now, many people have been correct, and advancements have been made, I'm just saying I think we need to be careful not to fall into the same "miracle cure" mentallity.
The best way to design is to learn what is available to you and create the best thing you know how. That, and be willing to try new things and adapt. All that this all this standards talk is, when you think about it, is a pissing contest for geeks. Less tech, more design!
paulOct 08, 2004 at 12:19PM
sorry about all the italics! forgot the tag.
Bud LandryOct 08, 2004 at 5:14PM
A bit off topic, but given the testimonials... what am I missing about Flickr being so damn juicy? It looks like a flash front end latched onto the protypical here is my personal photo websites, from iPhoto>dotmac from similar software bundled with every digital camera.
Perhaps it blooms into something fabulous with a broadband connection, or via blogging from a cell phone camera, but I just don't get it.
Bill SeitzOct 08, 2004 at 6:30PM
re: brochureware
Sure it has its place. How often does your offering/message change?
You can add value with other stuff, but it doesn't take long for a small biz to get to the wasting-money point of the curve.
Probably the next phase is a static website with a blog attached for doing simple time-sensitive messages, etc.
Phil WolffOct 10, 2004 at 6:41AM
Great questions, and comments.
Much programming work of the past 30 years was custom, then packaged with much tweaking, then turn-keyed, then rented, then open sourced, and is becoming free. Almost as though following Moore's Law, I can now install for free a $2000 web server, a $5k directory service, $25k app server, $75k content management system, $15k bulletin board, etc. That's money not going to programmers. And it's due to people standardizing what they do with software, and how they do it. Fewer choices and better defaults, the products of experience.
Is this happening to aesthetic, interaction, experience, or information design? If not now, soon?
Phil WolffOct 10, 2004 at 6:46AM
Cognitive scientists have been working hard for the last ten years. There's a treasure trove of new models of how we think, perceive, communicate, use tools, etc. How much are designers informed by current psychology, neuroscience, etc.? How much are common/(best?) practices in design validated by psychology?
In other words, when will web/app/software design stop being an art and start being a science? Or did I miss something?
Phil WolffOct 10, 2004 at 6:52AM
Just read Moral Politics and am reading Lakoff's latest ("Don't think about the elephant") on the differences between conservatives and liberals. The differences are rooted in how we rank our values.
Do you perform a values analysis as part of your persona definitions? How would you incorporate the values of a given community into the design of a community site or a shopping cart or a school registration program? Do you have a structured, validated approach to this?
Phil WolffOct 10, 2004 at 6:58AM
Is design now a planet-ready practice? How much has knowledge of the world's cultures - and their design considerations - seeped in to Anglospheric design? Is design for the world wide web still really design for the Western web? Or can you be confident that your giftware store will work fine in Russia? or the UK?
Phil WolffOct 10, 2004 at 7:00AM
Have lessons from the world of game design percollated into web and other application design? How long until we can't tell the difference?
CharlesOct 11, 2004 at 2:43AM
Do you think design "talk" is too focused on technological achievement ("Look mom, no tables!") these days, or is it a step in the right direction?
It's a step in the right direction. On the contrary, too much time, talk, and theory is still spent on design - how a site "looks" or its "elements" as opposed to what it can do. Of course the two are related and the design is still important, but it's way less important in the real world than most design gurus think it is.
Someone mentioned Craigslist - great example. No ground is being broken there design-wise, but it's a hugely successful site in terms of how it functions and community interaction. While its true you could argue not a lot is going on in Craigslist in terms of functionality, either, you have the options for user-posting, email subscriptions, RSS subscriptions by topic, and MyYahoo inclusion. That's still more than many "nicely designed" sites have.
markOct 12, 2004 at 2:50AM
Can usability drive innovation?
usability can and should drive innovation - although i take usability quite broadly. whenever there is a technology that still behaves like a technology, i.e. people have to learn it, configure it, and deal with it on its terms, there's plenty of room for innovation.
the web is getting there. but there's still a ways to go.
This thread is closed to new comments. Thanks to everyone who responded.