I've added a news location to my bookmarks - it's a
Microsoft blog by Jensen Harris. There's fascinating stuff on the design, and process of design, of the new Microsoft Office 12.
While we're on the subject, I wonder if they've finally made the "insert hyperlink" button smart? i.e. if I select something that has "http://" as a prefix, isn't it very obvious I just want that made straight into a link? Actually, I'd go further - tools like the Semagic blogging tool check the content of the clipboard and if that looks like a link, simply use that and don't bring up the dialog. It's incredibly effective; I always cut and paste links. I guess an alternative would be to have ctrl+V create a link on the current text instead of replacing the selection if the paste buffer contains an URL.
Fascinating article on
rounded corners here:
http://www.basement.org/archives/2005/11/why_do_we_love_rounded_corners.html
...to which I just have to say "yes". Interesting that it comes from both the nature-like approach and the practical visual properties of the things as a stronger hint to grouping.
Read. Enjoy.
There articles
popping up
all over right now about
Google Base. What surprises me is that there is a clear strategy here that people seem to be missing. Yes, you can upload content and have google host it. Fine. But many such services exist already; why is google so keen to do it? Personal ads are mentioned too, but that too is a niche business. Too small for google. So again, why?
The answer is in the extra requirements. Google base asks you to put in sets of standard information describing the thing you upload - such as "Address" = "..." or, if it's a property ad, how much the rent is, how much it costs. Or... you can just answer the questions and have google point at a website instead. Hmmm.
Tim Berners-Lee, one of the major contributors to the web as we know it, and others, have been pushing an idea called the
Semantic Web. What this provides is a common layout for rich data so that it can be understood, not just by humans, but by computers. You add it to normal webpages.
Consider something like google earth. Here is a world map that can have little signposts put on it by anyone describing items of interest at any location. With today's messy web, google can add a webpage describing, say, a property for rent, to its search engine. It doesn't have the level of understanding to associate that into google maps, though, and so relies on users keying in search terms and leafing through pages of results - or using another service.
How do you get that web data to include the computer-readable semantic content that google needs to join this all up - to get your property on its map? Well it's exactly what google base is going to ask you to type in.
The free content hosting that google base also offers is a sideshow; it's to build momentum; to get people to help build that rich, open store of information in the right format that google needs to get it moving. Once it is moving, people will start entering their data into google so that their property is found and rented quickly; or (real) estate agents will do it for you. But that too is transitional. People will soon tire of manually entering their data into google. Instead, they can just put it in their webpages and have google search find it - and that's what the semantic web effort is.
But don't think it's just about property; it's about
everything on the web.
No company other than google has the right attitude to pull this off. Yahoo and Microsoft have tried to dominate the web, but their solutions always end up closed and linking mainly to themselves. Google however continually takes an open attitude - adding value to the web ecosystem. Google base is moving in a very exciting direction. Let's hope it gets there.
Why is printing still such a backwards technology? I may just have sent a document to the printer twice, perhaps I should have walked over and checked. But why don't I have a history of what I've printed available along with an option to reprint?
