We are running some remote usability tests over Webex. The participant is given control of the mouse (webex lets you do that), and clicks through, while we’re all on a conference call. It actually works pretty well, I’m getting good feedback. A good option if you have remote participants.
Worldchanging points me to How smart does your bed have to be, before you are afraid to go to sleep at night? What a title!
PeterM questions the value of research departments like Microsofts, saying they don’t produce real products, whereas Apple (who killed their research department a while back) does keep coming up with innovative UIs.
The Google toolbar has a Valentine’s day theme - it’s kinda annoying since I can’t turn it off. Tim Jarret also noticed it. Oh, and it’s friday the 13th!
I am hosting a site on a shared server and I get the Ensim control panel. The site started to give strange mysql errors (corrupted tables), and it took me a while to figure it out. I am using all of my available space, so mysql runs out of space to write to.
Now the html + images is about 30 megs, the databases are about 30 megs as well. I checked and found that the apache error logs are over 200 megs (other logs are fine - 8 megs).
I changed the settings in the ENSIM panel, but it doesn’t seem to change the error log, only the other logs. Any ideas?
I’m also confused: I have my email on that server, but nothing else really. Where is all my storage space going? Any tips for optimizing that?
Jeff Lash spots a trend: increased usage of the site footer.
He refers an early article of mine where I analyzed clickthroughs from a sitemap at the bottom of each page compared to the usual site navigation. Jeff implemented the sitemap on every page concept at XPlane - it’s gone now. I had it for a few years on my Colombia site, but then I swiched to using Drupal as CMS and I haven’t had the coding skills to re-instate it. I’d like to test it again at some point.
I’m happy to report that loggnig the documentary happens pretty much at live speed. I just run the tape and log as it goes. So 20 hours of tape means 20 hours of logging. Not too bad.
I created a table in Word. The first column has the timecode (0:25:41), the second the scene (”plane”), the third details (”old man looking”), the fourth comments (”nice image - bad sound”). I only use the comments for 1 out of every 5 rows, more or less. A 1 hour tape turns into about 2 pages in Word. I just run the video on the TV and type in my Word doc as it goes.
I wish I had time to play around with StarLogo.
I should spend some time checking out this OWL thing. RDF and OWL Recommendations: “OWL builds on RDF and RDF Schema and adds more vocabulary for describing properties and classes: among others, relations between classes (e.g. disjointness), cardinality (e.g. “exactly one”), equality, richer typing of properties, characteristics of properties (e.g. symmetry), and enumerated classes.”
I wrote an artikel about IA for the Dutch publication Naar Voren: “Informatie-architectuur is een van die modewoorden die regelmatig rondvliegen in de wereld van de webbouwer.” It was surprisingly difficult to talk about these concepts in my mother tongue, for things like controlled vocabulary I had to reverse to English. Not that the word doesn’t exist in Dutch, but I don’t know it.
Ludicorp is the kind of name I’d give my company - if I had one. It has that hard-to-qualify Lex Luthor feel to it.
I’ve returned from filming in Colombia and now I have 22 tapes (about 16 hours) of video to edit.
The first step is to get an idea of what you have. I fast forwarded through most of the tapes, and used my digital camera to make screenshots of the TV. I then printed out the screenshots (using Windows XP’s printing wizard makes it very easy) per scene. I laid out all the scenes to have an easy way of discussing them and getting an overview. Think of it as a high-level sitemap.
Next I will have to watch all 16 hours and make a detailed log: at this timestamp, a man with donkey passes. This timestamp, the interviewee talks about X. I expect this to take 2 weeks, I’ll report back. And yes, it’s like a content inventory. Yey.
Jon is homing in on a beautiful approach to adding semantics to plain old XHTML elements, for searching and styling. It turns out you can specify multiple class elements like class=”personquote Jon”. In this example, you can apply styles separately for personquote and for Jon. Sweet.
For my reading/watching list: Future Salon : The revolution will not be televised. It will be streamed.
If blogs are conversations, then aggregators like the excellent Bloglines should look for FOAF files and display a picture and the name of the writer(s) (if they can find it) with a feed. I’d like to see who’s talking, and a name+picture makes it easier to do that.
(Bloglines could then add a function to see who this blogger knows and let me subscribe to feeds those people produce.)
I updated my feed to be called “Peter Van Dijck’s Guide to Ease”. Now I just have to add a picture. I always resisted having pictures of me floating around on the web (try to find one), but I think maybe I should give in.
Moderation continues to be a tricky art: craigblog: Flagging issues on craigslist: “Hey, we’re trying hard to turn over a lot of control over craigslist to the community. That’s proving to be a very democratic, effective, and fast way to run things.”
(via Catalogablog) Controlling your Language - Links to Metadata Vocabularies: a long list of links to CV’s and such, with explanations on how and why to use them.
Flash demo: interesting approach to providing dense information on a real-life map.
Wired News: Iran’s Most Wanted: Filmmakers: “Mantini, who works 60-hour weeks at a brick factory, spends weekends and holidays making movies. Each film is totally handcrafted. And each is completely illegal.”
Ten Mistakes Writers Don’t See (But Can Easily Fix When They Do) Gotta keep those in mind. A particularly nice example of flat writing: “He wanted to know but couldn’t understand what she had to say, so he waited until she was ready to tell him before asking what she meant.”
Language Log: schild en vriend: “Those suspected of being Walloons or Frenchmen who could speak Dutch were asked to say schild en vriend “shield and friend”, an expression regarded as particularly difficult for those who were not native speakers. Those who did not pronounce it correctly were determined to be the enemy and killed.”
Indeed. And the test has stood the test of time. Native French speakers still can’t pronounce “schild en vriend” correctly. Neither can native Spanish or English speakers.
To categorize or not: “My aim is to create a “loose” categorization of the services growing up around “social networking” guidelines, and not necessarily all “social software” offerings.”
Jon Udell is doing some experimentation with attributes to add more semantics to common HTML elements. He’s using these attributes to do searches and generate “dynamic categories”.
Automatic Organization for Digital Photographs with Geographic Coordinates (PDF, via Catalogablog): “We describe PhotoCompas, a system that utilizes the time and location information embedded in digital photographs to automatically organize a personal photo collection. PhotoCompas produces browseable location and event hierarchies for the collection. This organization is created using algorithms that interleave time and location to produce an organization that mimics the way people think about their photo collections. In addition, our algorithm annotates the generated hierarchy with geographical names. We tested our approach on several real-world collections and verifed that the results are meaningful and useful for the collection owners.”
I’m looking for examples of websites that raise awareness on certain issues (refugees for example), and analysis or research on that. Thanks.
Categorization of people you know is turning out to be a challenge for social network websites.
See Orkut
Follies and Clear…
precise and problematic: “The precision shows up in the digital choices we’re given: Is Phil your friend or not? If he is, is he one-star, two-star or three-star sexy? Choices you are not given
include: (i) Sort of sexy. (ii) Could be sexy if he dressed better.
(iii) If I were a woman, I think I’d find him sort of sexy if I went for that type and if he dressed better. So, exactly how many stars does that work out to?”
A nice categorization challenge: how do you categorize the people you want to add to a social networking service?
Here’s my take: for most sites, the basic connection should mean that you know who this person is. Not “Friend” but “People I know”. Doc Searls seems to think the same.
Further categories would really depend on the site. If I had the time (or someone paid me) I’d work out a categorization framework. But there are taxes to do.
I’ve always wondered how well Google’s translation approach works. They use volunteers to translate their pages in various languages. I checked out the new Dutch toolbar page today, and found some dodgy translations. Not exactyly wrong, but not particularly nice either. Here are a few edits. The English translations are my translations - I couldn’t find the relevant English page (maybe it’s been changed meanwhile).
- “Meer mogelijkheden van Google” (More possibilities with Google): “Meer mogelijkheden met Google”
- “de installatie vergt slechts enkele seconden” (the installation only takes a few seconds): “de installatie duurt slechts enkele seconden”
There are more edits I could make, but those fall in the realm of good writing. When you translate stuff, good writing often gets lost. I’ve seen the same in translations between Spanish and English. The translator should really be a writer as well.
World66 looks cool. It’s kinda like a huge wiki for travel, but more structured. And they know how to attract people like me: I’ve visited 22 countries in the world so far (red in the image below). A lot more to go.
Dave on the Dean campaign: “For a fraction of the cost of a state campaign, they could have deployed an information system for voters that would have made history. We’d be raving about how the Internet has made it possible to be a responsible voter for the first time. $40 million isn’t a big drop in an ocean of television ads, but in the space of public information systems it’s a virtually infinite amount of money.”
One interaction problem with having users assign categories is: how do you get them to select the category? If the categories are stable and the user has more or less learned them, James’ Autofill can come in handy. It also allows you to define new categories on the fly.
Nice.
(via Boing BoingThings): “An anonymous reader sez, “The BBC made a unique deal with Real Networks which disposes of their spyware tactics. Basically, if a user clicks on a link to download Real Player from a BBC website, the referrer script sends them to a page where they can download an expiry-free, spyware-free and nuicance-free version of the player. It’s because the BBC have such a stringent public service remit, that it was offensive to charge people a license fee for BBC content, then make them pay all over again for the facility to view/listen to it.”
So you can download an adware-free, expiry-free version of Realplayer at the BBC.
Statistics on the amount of words people use in searches. Nothing unusual.
Nice to see a good technical writer being recognized: Simon Willison started a blog on client side scripting at Sitepoint, and is being paid for it. Similarly, Mark Pilgrim started writing for O’Reilly a while back - it looks like good technical writers are being discovered through their blogs.
Interesting: Mitshubishi play a commercial on TV that ends in the middle of the action. Then they send you to www.seewhathappens.com.
Something feels all wrong with the way these social network sites work. (I only joined LinkedIn). I’m not sure what. The interesting part is: you can probably find out what it is by reading the right blogs. I don’t think they’ve nailed it yet, but they probably will be the first to to so. A few years back, I developed user requirements for a coupon site (yep) by looking at lots of epinions reviews of other sites like it. The exercise was very useful - the reviews were obviously written by vocal and interested users, and I came up with a lot of good stuff the market research had missed.