july 2005 archives

Wednesday, July 27, 2005

The truth, the whole truth and nothing but...

For long, online mapping and particularly online travel directions have been under heavy scrutiny. Furthermore, directions usually come with some rather discouraging disclaimers. This has certainly been for a reason! Indeed, many drivers have been led astray following these directions. However, it seems people have started to have confidence in online mapping again. An online map recently even decided a court case!

Although Google Maps uses the same map data suppliers as for example MapQuest, Yahoo Maps or Maporama and the same routing engine as Yahoo and Rand McNally, people think that Google Maps are more accurate, according to a report from Forrester.

So if the data is the same and you count out silly things like brand affinity, what is it that makes customers feel all warm and fuzzy about Google Maps? A user interface that treats them with a little respect.

Apparently, the Google Maps UI is so convincing, it was used in a court of law! Gear Live editor Edwin Soto, fined for running a red light, opened his laptop and browsed to Google Maps to pull up a map of the junction where he was fined to prove the street he was on actually was a two-way street, not a one-way street the police officer claimed it to be. Based on this evidence, the judge had to dismiss the violation due to lack of memory of the officer.

Really, I would be very hesitant to simply decide a legal matter based on a map. Of course, it's an indication that the traffic situation may be different. However, I have really come across roads being designated a one-way street when they clearly were not and vice versa. Sometimes, the direction of the one-way street was even reversed!

But what can we take away from this? Usability helps building confidence with your web audience and enhances your online credibility. This notion has driven many companies to improve the usability of their web sites. Google Maps proves this notion can be applied to online mapping just as well! So have a look at your online mapping app and find out how your users use it to achieve their goals. Read this article to improve the usability of store locators!

UPDATE: John wrote in to point out that, although two mapping websites use the same data and same engine, one can in fact be more accurate than the other. Mapping data suppliers release updates every quarter, but you cannot assume that all mapping websites update their mapping data to this frequency too.

It doesn't only take time to update the mapping data, but the update may not always be an improvement on the current data. Therefore, updating can be postponed until fixes have been provided or even until the next release of the data! Hence, we may have to adjust the importance of the UI slightly.

Sunday, July 24, 2005

The MapQuest experience

Yesterday, driving from Gainesville to St. Augustine, I used the MapQuest travel directions in the States for the very first time (bear in mind this is only my second visit to the US). I was very surprised with the ease of use: there were just 12 directions to guide us there! Furthermore, the 80 mile journey took us just a bit longer than the 1 hour and 50 minutes calculated by MapQuest. The directions were particularly helpful, as most signposts only indicated the road numbers, not the town names these roads lead to.

Of course, this was just one observation, but during the drive to St. Augustine a few hypotheses came to mind:

Usefulness: drivers in the US have to rely more on travel directions to get to their destination than drivers in Europe, because the sign posts in the US with just road numbers are less helpful than the sign posts in Europe that give you the town names.
Clarity: travel directions in the US are easier to follow than in Europe, because the grid structure of road networks in the US result in fewer and clearer instructions, whereas the irregularity and density of the European road networks result in many convoluted directions.
Resemblance: the mathematical algorithms for generating travel directions are more suitable for the road network in the States, because the graph matches the US road network more closely. Since the road network in the States is more regular, travel speed and the length of the road segments contribute more to the travel time in the States than in Europe, where other factors play a more important role.

Of course, it is very dangerous to make any conclusions based on a few observations. However, it would be interesting to see whether there are differences in using travel directions in Europe and in the States. Another interesting question is whether the popularity of online maps and travel directions differs between the States and Europe. I'd certainly like to hear whether you think using travel directions differs in the States and in Europe.

Friday, July 01, 2005

Geo microformat

Right after having taken in the Yahootinis and the beautiful view from the 32nd floor of the hotel, I attended the geo microformat BOF session downstairs. As there are so many ways for people to refer to a location in what they publish on the web, it's about time for a structured way of tagging locations.

Hosts of the session were various people from Technorati, among others Tantek Çelik. Technorati keeps a finger on the pulse of the blogosphere as it receives notification from over seven million weblogs as they are updated. To track what topics are currently being discussed in the blogosphere, Technorati shows the 250 most popular tags. People can categorise their blog posts, photos, and links with a descriptive tag (i.e. a keyword, subject or category). But how do we know what locations are popular? How can we mark up blog post for a search engine to understand what location is being discussed? This is the question that the geo mircoformat addresses (no pun intented!). Another popular microformat is XFN, a simple way to represent human relationships using hyperlinks.

We briefly went through some of the previous work in this area, touching on GML, geotags, ICBM-tags. Eventually, opinions seemed to converge around extending the hCard microformat to include address information. hCard is a 1:1 representation of the vCard standard in XHTML

Similar discussions have taken place on the geowanking mailing list as well:

Hopefully, these discussions benefit from each other so a structure evolves that works for all.

Where2.0(05)

Been reading all the blog postings about Where 2.0? Seen all the Flickr photos shot by members of the audience or the conference photographer? Been following the IRC channel? If you're a regular visitor of webmapper, then Where 2.0 has certainly been the place to be for the last few days. It's happening: online mapping isn't anymore just for geo-scientists and for geeks. Neverthless, having a background in cartography, it was great to hear the father of the GIS industry Jack Dangermond present.

It's quite difficult to summarise all my impressions and thoughts about the conference. The preview of MSN Virtual Earth was pretty cool! But is posting map annotations to MSN Spaces really collaboration? Google Earth and KML really got the crowd's attention. Apparently, Google had to suspend Google Earth for a while because so many people were downloading and using it at the same time! But yesterday's highlight certainly was the Google map API. Already, last night people presented their first apps at the Google Mapmaking BOF session. The discussions among the attendees ranged from:

What can and can't I do with the Google Maps?
What if Google Maps is being used for evil purposes?
Do people create their pushpin mapping apps just for fun of it, or would they rather just upload their data and have a generic app plot locations on a Google map?

I really enjoyed today's panel discussions, particularly about privacy and social networking. From the discussion about privacy I took away the same message as Bill Cartwright made last year at the TeleCartography conference: people are willing to give up their privacy if they get something in return. During the social networking discussion, the point was made, that one actually don't always want to meet new people that match my profile upon entering a public space. Nathan Eagle really showed the impact of being able to track someone's location and how you may actually derive assumptions about relationships between people with similar spatial patterns. Pretty scary from a privacy point of view!

All in all, it's been a great opportunity and feel the excitement and energy, to catch up and meet a lot of the people that have an interest in this area. Finally I can put a face to so many names! Oh, and of course, I really enjoyed beautiful San Francisco! As the dust is slowly settling, further postings on topics presented at Where 2.0 may be published here in the next few weeks.