Something is wrong.
OTHER media 2002 - Take it away: "Take away food should be a real winner online."
Reminds me of my experience with Verizon online billing lately: it took me over an hour and two tries and I couldn't set up online payment. And I consider myself pretty web savvy. It took me a few minutes to pay over the automated phone system. Something is wrong when a phone interface is more efficient than a web interface, isn't it?
Origins of a language
TheMetaDataWetDreamTool was a tool I wanted to design a while ago, one of my crazy projects (most of which never will come to pass). Somehow I ended up designing a language first. But now on to the tools! I'm gonna spend some time programming during the next few weeks, and I hope to get a working tool for playing around with XFML out there in a month or so... Watch this space :)
Thinking alike...
"Blogs indexed by a structured thesaurus makes it much easier to find other blogs that talk about similar topics without having to rely on the bloggers themselves to create the association via direct links. This could be a supplemental tool to the referrals that currently drive traffic between blogs." Ah!
Matt Jones calles XFML "Potentially
Matt Jones calles XFML "Potentially wondeful" - and I agree (as I would :). Let's make it happen.
What's an archaeologist doing at a design firm?
What's an archaeologist doing at a design firm? (via Peterme)
"By now, I'm used to the question, "How did your background lead you to this job?" People aren't quite sure how several degrees in Archaeology and Anthropology connect to a career in high tech design. The best way to explain the connection is to talk about the other question I get as soon as people find out I'm an archaeologist: "Have you found any cool stuff?" The answer is yes, but not exactly the kinds of cool stuff most people imagine. "
Amazon marketing info
I got an invitation to answer a questionaire from Amazon, and it turns out they use zTelligence. It must be the reporting tools, because the questionaire software was pretty bad: stupid error messages (and fairly dumb questions as well, although that's not the fault of the software).
XFML presentation
I wrote this presentation (with Radio's funky outliner tool): An introduction to XFML (presentation as OPML)
NOW I like Radio
Test slideshow generated from an outline: I tried Radio a year ago, and didn't really like it. I installed it again recently and started playing with it, and now I am really starting to like it. It really is next generation stuff.
OPML is getting more and more useful
The RadioPoint Tool lets you turn an OPML outline into a powerpoint-like silde presentation. Funky!
bringing the RSS "keep it simple" approach to Topic Maps
Edd Dumbill writes at XMLhack about XFML: "bringing the RSS "keep it simple" approach to Topic Maps."
Victor provides a bunch of
Victor provides a bunch of good links on the SIGIA list for when you need to align metadata between a bunch of different systems:
Basic introduction to what and why
Examples
Mapping Between Taxonomies
Semantic Problems of Thesaurus Mapping
Issues in Crosswalking Content Metadata Standards
I would definitely suggest looking into TopicMaps when you need to merge a bunch of systems metadatawise! Really, do it. They are very cool.
InContext Enterprises: if you are
InContext Enterprises: if you are an out of work HTML/design guru, you may want to offer these guys a free makeover of their homepage. They wrote the classic "Contextual design" (At Amazon.com), so it would be a nice high profile job, and you can't really go much worse than their current homepage.
Turns out XFML is also
Turns out XFML is also a file format:
"OmniForm's XFML electronic files are based off the OmniForm proprietary XML format, and used exclusively with the OmniForm ActiveX Control in OmniForm Developer's Edition."
XFML update
Victor hopes for software to play around with XFML soon - so do I. The language was designed so it would be easy to write software for. I'm also writing some myself and hope to get that finished in a month or so...
Internet Banks
I was looking at some Internet banks, and my god, they still live in the dark ages! How can they expect me to trust them with websites like that?
XFML.org: shared, faceted metadata format for the web.
XFML.org: shared, faceted metadata format for the web.
A good day: I finally published eXchangable Faceted Metadata Language version 0.1 today.
Amazon likes John (screenshot) from
Amazon likes John (screenshot) from Webword. For me, having my own tab on Amazon really makes me like them. I wonder how they test these things? (Not for usability, but whether it makes people feel good / buy more)
jaredRESEARCH: portfolio with some nice
jaredRESEARCH: portfolio with some nice case studies, focussing on business goals and explanations of methodology.
Book blog: User centered book
Book blog: User centered book design. It's kinda exiting to see UCD techniques work so well for a different medium.
User Interface Engineering -- "Market
User Interface Engineering -- "Market Maturity".
Rereading this old article made me realize weblog software is already hovering between Stage 2: Checklist Battles and Stage 3: Productivity Wars. Wow. Things move fast!
GUUUI - The Interaction Designer's
GUUUI - The Interaction Designer's Coffee Break (via HeyToWell): a pretty cool site focussed on usability (notice the picture next to the coffee cup). I like the fact they have a humor section.
Research is everyone's job
New Architect: Inspirational Guidance (via Victor): a review of the book The Art of Innovation.
" [The talked to users ...] But not just any users. Instead of hunting for "typical users", the pre-screened, carefully segmented, demographically consistent humdrums who tend to fill up focus groups, IDEO sought out the "crazy user."
For example, an office worker named Sally had developed her own idiosyncratic approach to teleconferencing that involved carting a bunch of individual speakerphones into a conference room, setting the phones around a table, dialing in each caller separately, and conducting the conference call in the open air of the room."
You see now that's interesting.
"The authors also talk about the importance of what they call "cross-dressers," or team members who switch disciplines or specialties. Engineers turned designers, for instance, and vice versa. IDEO blurs disciplinary boundaries wherever possible. That's especially true when it comes to research, a cornerstone of IDEO's design process. Many design firms still treat research as a stand-alone discipline practiced by Researchers (with a capital R). At IDEO, research is everyone's job." (See Better questions, not more answers.)
Interaction seminar
interactionmasters (Flash 5 content, via HeytoWell): some good stuff at this site: ideas from a seminar in Finland.
The presentation is really nice: it's good to see photos of the seminar floating by, together with quotes showing the main ideas. The site has some typical Flash UI fproblems (popup windows! Confusing navigation!) but I really like the way it lets you quickly get an impression of the seminar, very nice.
This little UI innovation is
This little UI innovation is so darn useful I really want it in Windows, not just in one application. Can you guess what it does? (I could - a good sign.)
The humble browser
Time for a humble browser (Advogato, via the ever reliable Webword)
"And while I really like the idea of standards, I also shudder to think of where we'll be in fifty years, if Opera's position becomes dominant. Innovation will dry up because everything is 'standards-based' and the standards are ALWAYS a couple of years behind innovations. Having worked in a giant bureaucracy, and also for a small company, I prefer the latter because the standards are so much freer. Easier to navigate. [...] I do not want a standards-based Internet which tolerates no new extensions, and for THIS reason, I am an Open Source advocate who is really excited about IE's innovations, frustrated by Opera's simple inability to allow a drag-n-drop interface within a browser. "
Evolution Robotics offers Robotics Software
Evolution Robotics offers Robotics Software Platform and Hardware Kits (via AIlog)
"All you need is a Windows laptop and your imagination." The robot comes with easy to use software that lets you train it. Cool. I wonder how useful it will turn out to be to have little robots available this cheap (US$ 599). Will they become more than just a toy for geeks? And how much a factor will the user interface be in this?
Competing with Microsoft
Business 2.0 - Magazine Article - Beating Bill (via Bblog)
"But while AOL's subscriber growth has stagnated, MSN's subscriptions have doubled in the past year alone. Significantly, half of the new signups are AOL defectors, according to Yusuf Mehdi, vice president of MSN. "
The article describes Microsoft as a force of nature, which I guess it pretty much is for many companies...
(Via Christina) The Novus Petroleum
(Via Christina) The Novus Petroleum homepage shows an IA style sitemap and states "The site has been designed to become increasingly detailed at the deeper levels of the hierarchy (see site map). "
Let me jump on this trend and state that "The page you are viewing now has been designed to show a large amount of content that can easily be scanned and is organized in an abvious manner".
I don't like that sitemap on novuspetroleum. They should have used text links for a small site like that. (Putting all the links on the homepage is a good idea.)
How ordering links by popularity can backfire.
Small world theory of why random links matter.
"But their biggest surprise came when they started adding random links to an otherwise regular network. The average path length didn't just fall, it plummeted steeply."
Which reinforces the idea I'm getting that we need unpopular links as much as we need popular ones.
A lot of new systems are being built that create value by automatically creating related links. They often allow for automatic reinforcement of popular links (by doing things like ordering by popularity) and as such can create problems by locking users in in a small world of connected websites, while limiting serendipity.
So you could argue that high value low popularity links are a lot better than high value high popularity links, since they have the additional function of promoting serendipity.
Ah! So that's why!
Salon.com: Will Americans go for mLife? (via BlackbeltJones)
"For many skeptics, imode-type services will never take off in the U.S, for one simple reason: the car. In Japan, the ubiquitous mass transit system is often cited as a primary reason for imode's success. The transit system creates a lifestyle full of "microniches" of time. There's a lot of hanging around nearby bus, subway and train stations, usually waiting for friends or for transport. "
I never thought about that... It nicely fits in with why schoolkids love mobiles.
Small world effects through weblogs
Internet navigators think small (via IASlash)
"You can be sure, as with any technology, that once it becomes good, people will begin renaming what they have", he said. "Watch the content management vendors start claiming that they have blogging functions. That's an inevitability. There's always a deliberate attempt to confuse the issue when the public finds something it likes."
Iranian weblogs
It's now the 6th month of this, let's say, "movement" among Iranians. There are about 1000 Persian weblogs out there right now. (Scripting news)
Interesting from a cultural perspective...
Reading weblogs has made me
Reading weblogs has made me greedy for access to a constant stream of nuggets from the brains of those I admire. So it is. Via scripting: at least he posts a lot; it seems a bit quiet in the blogs I visit lately. Time for a change?
Diamonds of sugar
Episode 5: "Diamonds of Sugar": "While listening to Laura and Kana plan an evening barbecue, Hamtaro wonders what the girls are eating--a sparkling candy called "Diamonds of Sugar".
What's up with all the not too well hidden references to drugs in childrens cartoons lately? What's that about?
New evidence of Neanderthal violence
New evidence of Neanderthal violence (BBC)
"The findings indicate that the contemporaries of early modern humans were more sophisticated than their popular "caveman" image suggests.
They would have needed social skills and organised networks to take part in armed conflict. It may have been a crucial factor in the evolution of Neanderthal and human behaviour, say scientists in Switzerland and France."
So we may hae evolved to more efficiently kill each other? Makes kinda sense.
Late baby
"As more women delay having babies to finish their educations and establish their careers, a University of Michigan researcher identifies an unexpected long-term implication of waiting: likely eventual extinction of the older mother's lineage. "
Emotion Generating Technology.
My girlfriend took me to a huge toystore called FAO Schwarz (on the corner of Central Park east and Fifth Avenue) in New York. I was looking at this little robot dog called Aibo made by Sony that displays learning behaviour and plays around with kids. It recognises faces and develops its own personality over time.
A few weeks earlier I saw the movie AI again on DVD. It has a little robotic teddybear in it with real personality (Teddy), and I was thinking: if they ever really make that it will be a huge hit. Teddy has real personality, and is very loveable. It also has a little boy robot in it, which also extremely lovable but also a bit creepy sometimes.
Interestingly, in the movie, older robot models are easily discarded for newer models, a behaviour I observed in real life in the toy store as well. (see picture below)
Notice the poor little robot in the corner nobody wants anymore. It is a primitve early model, not nearly as interesting as aibo: everybody is standing around a demonstration of him.
But when I was watching a kid playing with the Sony robot dog, it didn't really treat it with a lot of emotion: it prodded it and tried to find out how it worked. It was curiosity, not emotion. Maybe emotion takes time? Then again, maybe not. When kids hug a toy, there is emotion. So it's not technological advancement that generates emotions.
Movie of kid playing with robot dog (.avi, 888K) (See also this movie of a kid playing with an older type of robot, .avi, 834K)
How can technology generate emotions?
What would make you so attached to a technology you wouldn't want to hurt its feelings? Or is that even what we want? Maybe we just want technology that can generate emotions in us but doesn't have needs of its own? Like a drug, really.
Also:
Why do people care for plants?
Our relationship with technology could be like the one we have with dogs and cats. Or like the one we have with our favourite toys. Or like the one we have with plants we care for.
A few years ago, there was a short rage of a little toy called tamagotchi (research on Tamagotchi). It was really small, hung on a keychain, and you had to regularly "feed" it and "pet" it by pressing certain buttons or it would "die". It generated the emotion anxiety: once people felt they had it in a stable state (so it wouldn't die), they tended to get bored with it. (This conclusion is based on admittedly little evidence.)
I was discussing this last Sunday with my friend Jay and he came up with a good rule for active Emotion Generating Technology: the why-did-you-do-that function. A toy like that would have to have its own boundaries, its own internal rules and behavior in order not to be boring, but it would also be important for its owner to be able to predict and understand it. The Why-did-you-do-that function allows the owner to ask that and to tell it to do something different next time.
Remember Terminator II? When the terminator first meets the boy, the boy has to teach him not to kill people. That's a why-did-you-do-that function.
EGT (Emotion Generating Technology) is not nessecarily about language recognition: dogs don't speak or understand us, and yet people really truly love them.
EGT isn't about advanced technology either. You could have relatively simple technology capable of generating emotional responses. At least, I like to think you could. Research is needed to find out what it is about technology that could generate lasting emotional responses.
There's another interesting thing going on I learned from animated movies: too much realism becomes creepy. You can get to 95% realism and be fine. But when you go over that, at say 99% realism, it gets creepy instead of cute. You are unconsciously picking up signals something is wrong, but consciously you can't tell what. So being too realistic isn't good. In Shrek, the modelers decided to make the princess Fiona character less realistic and more cartoon-like after initial tests showed the realism was too creepy.
Anyways, if you have personal stories about your experience of emotions with technology, please let me know. Links to research and other related stuff are very welcome as well.
Information architecture makes design accountable
Here's the deal: many companies still hire a Lead Designer to lead the design team (makes sense so far) that defines not only visual design but also more IA related aspects like interaction (navigation, flow, functionality) on their websites. These days, many include "familiarity with usability testing and information architecture" as a requirement, along with more tradional designer requirements like "fluent with Flash, Illustrator and Photoshop" or "an extensive visual design portfolio".
You can see which part doesn't make sense: having the lead visual designer with a focus on visual design and branding be responsible for the IA and interaction design of a website is wrong. These are different roles, that demand a different focus. Rare are the people who can (or want to) do both really well.
I was looking at one of these jobs recently, and and found myself thinking (again): they really should hire an IA instead. I thought: "Assimilate, don't complain!" and applied for the job.
But how could I pass the interview? They'd have to want to change their process. If a company has a process in place, as someone asking for a job you're in a pretty bad position to convince them to change that. So I wrote down a list of ideas, arguments and answers to potential questions to prepare for a possible interview.
The main idea in this conversation is to sell the category (IA), not the person (you). They won't hire you as an IA if they don't think IA is important. Once they do think so, guess who they will probably talk to first? You'll get a smaller slice (they might hire someone else still), but of a much bigger pie.
Businesses, especially these days, care about accountability, getting more out of existing investments, and the good old bottom line, so that's how I presented these arguments. It would be foolish to argue for example that IA makes for better visual designs, you'd be picking a fight on somebody else's turf.
Argument 1: IA deals with today's big, complex sites.
Today's websites aren't the small 500 page sites of yore anymore. Today's website are huge, complex and ever changing information delivery vehicles. The fact many websites are so hard to get around in illustrates this. Information architecture is a new role, created by these new demands. Where a traditional visual designer will tend to focus on branding and visual design issues, and information architect has the tools and methods to define websites that are easy to use, and can effectively deal with todays websites that contain huge amounts of constantly updated and changing information.
Argument 2: IA delivers user satisfaction.
IA's typically deal with large web presences, and take a systematic approach to organizing them. They have a strong research background, and base their decisions on a real understanding of users. They are familiar with using effective decision making methods like usability testing. In real life, this means development efforts are focussed on ease of use, on creating satisfied customers. User satisfaction is important for your company, and to deliver it you need a strong focus on this in your development process, it doesn't just happen.
Argument 3: IA increases efficiency through ease of use.
Good IA makes websites easier to use, and more effective. This means you can have a huge investment in a current website, with backend systems, having fought integration nightmares, branding battles, and still your website might not deliver the results you expected. IA can change that. It is very good at identifying and fixing problems that could be costing you hundreds of thousands of dollars in lost sales.
Argument 4: IA is accountable.
The core job of an IA is to create a website that reconciles business goals with user goals. An IA will always start a project with goals, and measures of success. This brings a previously unobtainable level of accountability to your web projects. You will finally know exactly which parts of your web presence work well, which development efforts are cost effective, where future efforts should be focussed.
Argument 5: IA deals with changing business goals.
In a web environment, business goals and user goals are in constant flux. You need to design a website that reflects these changing goals. So you need to reflect these goals in your design in a systematic, accountable and focussed manner. That is exactly what IA's do.
Argument 6: IA works well with designers.
The information architect typically works very well together with junior and senior designers, since she doesn't focus on visual design, so there are no turf wars: the designers just have more base material to work with. (*). The IA position is not about having great visual design skills. The IA will provide a base that the designers can build upon, without that solid base they will always have a hard time turning business requirements into successful websites.
Argument 7: IA connects visual design with business goals.
You may have noticed difficulties relating your visual design decisions to business goals. IA is the intermediate step that allows for making that link. The information architect can take over some of the roles visual designers may find tedious (sitemaps, flowcharts), basing these on business goals, while freeing visual designers to focus on design and branding. (**)
Argument 8: IA saves money.
Information architecture solidifies the processes of gathering and defining requirements in detail before implementing them. You will know exactly why a certain amount of time is being spent on a certain feature, and it won't be just because it's cool. This preciseness saves money by avoiding time spent on building features nobody will use. You will have a solid methodology for deciding on what elements to best spend your money.
Argument 9: (the evergreen of arguments) Other People Do It.
Companies like Epinions have hired information architects as Creative Directors. They realize
the unique challenges of the medium. Other companies that work with information architects
include Epinions, 3Com, AT&T, Compaq, Ernst & Young, Ford, IBM, Microsoft, Procter & Gamble, British Telecom, Fannie Mae, First USA, AOL, Square D, L.L. Bean and Hewlett-Packard, and The World Bank. (This short list was copied from here.)
Question: Won't the effort the change in process takes cancel out any potential benefits? Ie. we don't want our people to go trough the hell of adapting another process again: been there done that.
Answer: If your current process is flexible enough to deal with the constantly changing demands of the internet environment, IA will easily fit in there. Good IA makes for happier programmers, since they get better specs, happier designers, since they have more to work with, and happier project and business managers, since projects have more accountability built in and are more structured towards business goals.
I'm sure a lot more arguments could be made for hiring an IA and not just a Visual designer to lead a development team. I'd love to see your ideas.
Finally: in some cases, and especially in todays risk-avoiding business climate, trying to sell the term "information architecture" if they are not familiar with it already can be a lost battle from the start. You may be better of just being a Lead Designer but negotiating your job description and the process into something IA-like. Pick your battles, don't focus on the job description.
(*) Remember the above are arguments designed to convince someone to hire an IA instead of a Lead Designer.
(**) If this comes across as evil bashing of visual designers, I'm sorry. I love visual design. I'm talking in a limited way about visual design here, I realize that. I still think a visual design background can be for doing IA detrimental if that person takes on IA as this little extra thing they have to do, not as their core activity. Their decisions will keep being focussed on visual design, missing a lot of opportunities designing interaction. IA is important, a lot more important than visual design for 95% of websites out there in my (admittedly and understandably controversial) opinion, if you use this (limited) definition of visual design.
My cunning scheme for world domination
After a chat with Victor about my metadata idea, he pointed me to this conversation about metadata for blogs: kottke.org - may 2002 archives.
I think I may now announce my master Plan for World Domination through metadata. It will allow (I almost wrote "the blogverse"! Auch. Not enough sleep I guess) people not only to manage their metadata more cleverly than 99% of current systems, but also to exchange it. And it will be simple, unlike topicmaps. Check back in a few weeks.
Gender and Technology: A Case
Gender and Technology: A Case Study (Video feed) by Brenda Laurel: she describes ethnography as a design tool nicely as: "paying very close attention to the details in the lives of certain types of people as a strategy to design products".
Some tips on interviewing kids: interview them together with their one best friend. Advantages, they keep each other honest, and they talk a lot to each other. Focus groups with kids don't work.
They did lots of research on gender differences as it relate to designing games for girls, and came up with a hierarchy of colors and characteristics for what works for boys vs. girls. For example: pink overrides truckness. So a pink truck is still girlie, no boy wants it. Other finding: the hi-res polygon typical rendering of games is very boy-like, even if the thing that's being rendered is a pink pony, it still says "boys".
Other interesting bits: Brenda thinks there are some key biological differences between boys and girls: the biological differences are small but get enlarged by cultural narrative. (Differences like spatial awareness under time pressure (videogames anyone?)). Girls do well with Tetris, a pattern matching problem. How the problem is described matters on how well you perform: are you "helping your car down the street"?
Brenda also made some good points on ethics: use applicable ethics at the right point: when doing research, use research ethics. When developing products, use more political ethics. Don't use political ethics when doing research.
It's an interesting talk about gender issues for kids, but also about research practices - she shares lots of interesting findings. If you build websites for kids, listen to this!
Something I am thinking lately: websites aimed at very specific audiences are pretty powerful things. They serve these particular people so well that it's hard to steal the audience... Not that that had anything to do with the above post. Oh well...
VerizonWireless
My god, Verizon Wireless really sucks. I'm a pretty advanced and motivated power user, and I could not set it up to take my money automatically. Couldn't do it. It gave me fairly stupid errors that some simple testing could have easily solved and just doesn't like me.
Pay me money Verizon and I'll fix it for you. I don't like them as a company. If they hadn't so effectively locked me in, I'd go somewhere else immediately.
Refine search queries
Old news I'm sure: AltaVista Testing "Paraphrase" Tool: the tool suggest related search queries. It's about time. I have always loved the accuracy of Google's "Did you mean ...", but that is really a spell checker. The queries the new Altavista tool suggest (as far as I could find out) don't seem releant enough yet to make it worth the space they take up (especially in that space-eating altavista design, com on guys, get a designer to clean that baby up), they are simply derived from the titles and descriptions of the page one results.
But suggesting queries is definitely a brilliant idea, and the way forward for search engines, since most people don't refine queries.
By the way, I'd never seen Google do this before until today:
Petervandijck dot com and dot
Petervandijck dot com and dot org (I already had dot net) have become available, so I ordered them at my favourite registrar joker.com. (Don't use Verisign)
NYC
They: excuse me, my name is xxx, I live here down the road
Me: Hi
They: I am going to that shop over there to buy some milk for my baby, I just need a few dollars, nobody wants to talk to me
Me: (What a hard life!) Not today man.
They: I just need another dollar, what am I going to do.
Me: OK I'll give you a dollar, here you go, take care.
Next day:
They: Excuse me, I'm going to buy a turkey-cheese sandwich in that shop over there.
Me: Sorry man.
They: I'm not asking for money man, I'm hungry and I just want a sandwich.
Me: You can have my (non-eaten) bagel here.
They: No thanks, I just want a turkey cheese sandwhich to take to the shelter.
Me: (What do I do?) Next time man.
Next day:
They: So where are you from?
Me: Belgium.
They: I like Belgium! They speak French, Flemish and German, right?
Me: Yes.
They: Yeah, I visited Belgium a few times, it's a great city.
Me: It's a country!
They: Yeah I know, but I always think of it as a city - it's really one big city isn't it?
Sorry mr. Norman
Full-on rantmode.
Gratuitous Graphics and Human-centered Website Design ( jnd.org ): this is just annoying. It's not fun. If this is Don Norman's idea of good use of graphics and animation, I'm sorry.
I can appreciate Don making a much needed case for graphics and fun, but at least he could have used something that was interesting for anyone else than a complete web newbie and for longer than, say, 3 seconds. Something that didn't disctract from reading? There is a lot of good graphic stuff being developed. Sometimes the NN guys just portray this cluelessness about the web that is hard to understand from such clever people. Get permalinks Jacob! It's not hard.
What's more, IE popped up a message telling me the script was causing it to run slowly. Sigh.