Post Archive
› December 14, 2004
.mobi versus device independence
Two days ago, ICANN has approved two new sponsored top-level domains (sTLD): .travel.jobs and, more importantly, .mobi. I won't comment on the .travel.jobs sTLD; about .mobi, ITWorld says the following:
.Mobi is sponsored by Microsoft Corp., Nokia Corp., and Vodafone Group Services Ltd., which hope to target the domain specifically at mobile content and service providers as well as mobile device manufacturers, vendors and individual companies.
In a domain proposal submitted to ICANN, the sponsors predicted a boom in mobile devices by 2006, requiring a host of new Internet services and content tailored to their use.
Yes, you read that correctly: ICANN (together with Microsoft, Nokia and Vodafone) expects web developers to start differentiating between "mainstream" and mobile browsing devices in a DNS-related way, that is, by means of the .mobi sTLD, thereby completely ignoring the concept of device independence and technologies such as CC/PP.
In a device independent architecture, content is dynamically adapted to the client device making the request; this adaptation can occur anywhere in the path from server back to client, thereby taking into account the device's delivery context or CC/PP profile. Stripping out high-res images before sending content to a mobile device is an example of server side adaptation, the application of only handheld CSS rules an example of content adaptation on the client side. Note that this model makes a clear distinction between the content itself and its possible adaptations, meaning that content can be located by means of a single URI. Introducing a .mobi domain for pointing to "mobile resources" thus completely corrupts this logic and means a step back for mobile web development, if massively deployed.
In July 2004, the W3C protested against the announced introduction of a .mobi domain, but apparently was not heard. Furthermore, if you take into account the recent announcement of a formal relationship between the W3C and the Open Mobile Alliance, of which Microsoft, Nokia and Vodafone are members, the whole .mobi thing is even harder to understand.
Comments
1. December 14, 2004 03:31 PM
2. December 15, 2004 07:38 AM
Dean Jackson Posted…
The W3C/OMA liaison doesn't really have anything to do with .mobi in particular. It's more a result of the adoption of W3C technologies on mobile devices (see the position papers submitted to a recent W3C Workshop). The W3C still protests the introduction of .mobi domains, for all the reasons listed above. One Web, for everyone, on anything. [disclaimer: W3C employee]
3. December 15, 2004 09:15 AM
Andreas Posted…
Dean, thanks for the clarification :-)
And for those interested, two related links with more info:
4. December 15, 2004 09:21 AM
Charles Martin Posted…
Sorry, Dean, it was rather stupid of me to make those statements. What bugs me is that all these companies will now have a website that pretty much is a mirror of their existing one, but probably just use a different stylesheet. I just feel sorry for those companies still using tables for layouts since they'll have to now develop a separate website for the .mobi domain. Unfortunately, most companies will register the .mobi domain for their use, but then just have it point to their existing site. Just a waste really.
5. December 15, 2004 09:46 AM
Kevin Cannon Posted…
I'm still not sold on the whole device independence idea. It's a nice concept, and i think it could technically be possible, but the Interface Designer in me, thinks it's not always going to be the right solution, and it's stupid to blindly follow it.
A webpage designed for someone using a large screen, is different to a page designed to be used on a mobile phone. They're two totally different media, and whil it's nice if you can access the same content from both, you probably won't want to. Mobile content, will often be, inherently different, from 'PC-content'.
Device independence is great, but not at the expense of providing the correct content for your audience, don't you think?
6. December 15, 2004 10:37 AM
Charles Martin Posted…
I do agree that the content directed to the proper device may divurge greatly (after all, it hardly makes sense to have a graphics intensive site download to a phone). However, most companies that do target existing mobile devices either already have a domain for that purpose, or detect the device and serve different content. Why make it more difficult for the user to have to remember that they need to go to www.microsoft.mobi instead of www.microsoft.com when one site can serve multiple purposes already?
7. December 15, 2004 12:10 PM
Jonathan M. Hollin Posted…
There are very good reasons why the new ".mobi" domain is a bad idea. As described by Tim Berners-Lee in his paper: New Top Level Domains Considered Harmful. Consider the following (quoted from Berners-Lee's paper):
The Web works by reference. As an information space, it is defined by the relationship between a URI and what one gets on using that URI. The URI is passed around, written, spoken, buried in links, bookmarked, traded while Instant Messaging and through email. People look up URIs in all sorts of conditions.
It is fundamentally useful to be able to quote the URI for some information and then look up that URI in an entirely different context. For example, I may want to look up a restaurant on my laptop, bookmark it, and then, when I only have my phone, check the bookmark to have a look at the evening menu. Or, my travel agent may send me a pointer to my itinerary for a business trip. I may view the itinerary from my office on a large screen and want to see the map, or I may view it at the airport from my phone when all I want is the gate number.
Dividing the Web into information destined for different devices, or different classes of user, or different classes of information, breaks the Web in a fundamental way.
8. December 15, 2004 12:56 PM
Andreas Posted…
A webpage designed for someone using a large screen, is different to a page designed to be used on a mobile phone.
I couldn't agree more, Kevin. And so does the W3C; CC/PP is there as a solution for this kind of problem, that is, for content transformations that go beyond handheld CSS. From the WaSP's Introduction to CC/PP mentioned above:
Combining the possibility to transform XHTML content through XSLT with the flexibility and accuracy provided by CC/PP makes it possible to transform hypertext content on-the-fly beyond what style sheets already allow.
In other words, have your cake (one URL per resource) and eat it, too (device ready content).
Now, say you have to develop a site with a lot of images (using img tags) and you want to differentiate between normal and mobile browsers in a way CSS does not allow you to do—server side image compression in combination with CC/PP based user agent targeting is the answer here. Mobile browser users get lo-res images, normal browser users see hi-res ones.
Now, in case you decide that CC/PP is too scary to work with (for now), you may go with a http://your.site.com/mobile/ solution for a while. Note however that this is only a stop-gap measure, possibly leading to confusion when it comes to localizing specific resources. However, freezing a temporary http://your.site.com/mobile/ like setup by pushing for http://your.site.mobi/ to become the standard practice is a step too far and will break the web.
9. December 16, 2004 05:13 PM
John Dowdell Posted…
A webpage designed for someone using a large screen, is different to a page designed to be used on a mobile phone.
Besides visual design, this can often hold true for interaction design as well....
jd/mm
10. December 18, 2004 10:29 AM
11. December 24, 2004 01:08 AM
Ottawa Posted…
This is getting ridiculous. First they come out with with virtually useless .museum and .aero, now .mobi... .mobi is totally ambiguous too.
Charles Martin Posted…
Looks like W3C decided "if you can't beat 'em, join 'em". I wonder if they think they can influence the major players since they made no headway in cutting off the whole idea to begin with.