One internet or two?

By October 6, 2006Mobile

This is of course a question about mobile, and whether we should have one web which we look at from all devices or pages dedicated to display on small devices which copy those which we look at from our PCs, in effect creating two internets.

This is a post that has been forming in my mind since the Mobile Monday London event at the beginning of the week – they happen on the first Monday of every month and are well worth it if you have an interest in the mobile web (you may need a Yahoo! account to join the group on this link).  Congrats to Jo Rabin for organising this one.

The one web argument is essentially philosophical in nature.  Its proponents believe the web should be world wide and therefore only one is necessary – access from different devices should be managed by appropriate transformation.  They have on their side the practical point that maintaining two of anything and keeping both current is a pain.

The two web argument says that mobile is sufficiently different – small screen, limited input options, narrow bandwidth, not always on, and high latency, that you will not get decent usability unless you design your websites/applications specifically for that environment.

I’m a one web guy. 

I have been very influenced in my thinking by the QTEK phone I got a month or two back.  The QTEK is a smartphone with a full keyboard and push email.  I can use it to access just about any site although the navigation can be a little clumsy.  This made me realise that the distinction between PC and mobile is a little false.  Screen sizes are large on the desk and run in a more or less smooth continuum through laptops, to notebooks, to palm tops, to tablet PCs, to large smart phones, to small smart phones to mass market phones.   Trying to force these into two groups as implied by having a mobile web doesn’t make sense to me.

The other reason I am a one web guy is that the sites people want to look at on their mobiles are by and large the same as on the PC.  A number of people at MoMo this week cited studies to that effect and it rings true for me.  One of the reasons I like the QTEK so much is that I can click through links in emails and look at sites people want to show me (doing much with them is usually too hard, but most times I am just being directed towards a piece of content).  If there were a small subsection of sites that had extremely high mobile usage then there might be a case for building them differently (maybe with a .mobi URL) but I don’t see that happening much.

I am generalising here and there will be niche exceptions but I think I think what I am saying is directionally correct.  Email is a one off case which receives special treatment because of its high volume.  It even has dedicated devices.  I should also say that I am taking for granted all the difficult work that Volantis and others do in optimising sites for the myriad of devices we all use.

The reason all this interests me is that it is important in understanding where the opportunities will come in mobile.

If I am right on the one web point then that says to me that mobile isn’t a separate category for lots of applications.  So thinking about buidling mobile social networks or mobile dating sites is looking at things from the wrong perspective.  The interesting question becomes how to take MySpace to mobile, or Match.com to mobile.  Betfair is an example of where everyone has understood this intuitively – replicating the whole operation on mobile would never work because it would never get to critical mass – but mobile extensions are a great way to improve the service.  Check out their site for mobile.

This argument becomes stronger as mobile traffic is increasingly carried over IP, either via handset clients using the GPRS channel or Wifi for calls or texting (check out Hotxt), or in the network itself (our portfolio company Empower is doing some cool stuff here).

As I think about it more I realise this argument has profound consequences for much of the mobile internet value chain.  If we get to a one web world will the operator portals be there slugging it out with Yahoo! and Google?  I doubt it.

Opportunities for entrepreneurs and VCs will come (or have come) at the infrastructure level, like Volantis and tools to take apps like Betfair mobile.  Beyond that we need to look for things that are truly different on mobile – not different just because of screen size etc.  Location based services/local search is a good example.  Advertising might be another – different messages will be appropriate to people on the move and the screen size point is significant here because the advertising is secondary to the main page people are looking at and there simply won’t be enough space for it (see Google strips out ads when it optimises for mobile).

  • “Beyond that we need to look for things that are truly different on mobile – not different just because of screen size etc. Location based services/local search is a good example.”

    The unexpected development may be that mobile becomes the CORE platform for some applications e.g. music, email and starts to extend from there onto the online world. In South Africa a mobile-focused IM client (Mixit) is apparently more popular than online equivalents.

  • “Beyond that we need to look for things that are truly different on mobile – not different just because of screen size etc. Location based services/local search is a good example.”

    The unexpected development may be that mobile becomes the CORE platform for some applications e.g. music, email and starts to extend from there onto the online world. In South Africa a mobile-focused IM client (Mixit) is apparently more popular than online equivalents.

  • I’m a one web guy, the mobi tld and how it’s presented bugs me and i’ve been pretty vocal about it, just google “shawn mccollum mobi”.

    I agree that extending specific services into mobile is probably a better idea then trying to replicate all the functionality. I just hope they build it in a way I can use it to make mashups. jamescoops i’d bet your right, why would you need a desktop for music if you could buy and download music via some 4g data service right from your ipod.

    Location based services are poised to take off but I don’t think it’s going to hit critical mass until the mobile browsers allow a user to send location info to websites. Right now to take advantage of location based services you have to download a java app per service, that is if your phone is popular enough to warrant the cost of developing a version for your phone. i think the mobile web 2.0 guys should focus on mobile web browser 2.0.

    I also think phyisical hyperlinks are cool. My favorite is near field communication, it take the best of 2d barcodes and bluetooth squeelers with out most of the negative usability issues.

  • I’m a one web guy, the mobi tld and how it’s presented bugs me and i’ve been pretty vocal about it, just google “shawn mccollum mobi”.

    I agree that extending specific services into mobile is probably a better idea then trying to replicate all the functionality. I just hope they build it in a way I can use it to make mashups. jamescoops i’d bet your right, why would you need a desktop for music if you could buy and download music via some 4g data service right from your ipod.

    Location based services are poised to take off but I don’t think it’s going to hit critical mass until the mobile browsers allow a user to send location info to websites. Right now to take advantage of location based services you have to download a java app per service, that is if your phone is popular enough to warrant the cost of developing a version for your phone. i think the mobile web 2.0 guys should focus on mobile web browser 2.0.

    I also think phyisical hyperlinks are cool. My favorite is near field communication, it take the best of 2d barcodes and bluetooth squeelers with out most of the negative usability issues.

  • I completely agree that “we need to look for things that are truly different on mobile – not different just because of screen size etc.” You understand that the nature of the mobile space and in turn the expectations of its users are unique.

    In that regard, then, I’m surprised that you see supporting .mobi as supporting a second web. I don’t see people having issues with, say, amazon.com versus amazon.co.uk versus amazon.fr versus amazon.co.jp, etc.

    I believe people understand that different domains exist to serve different purposes and meet different expectations. I don’t see why .mobi is any different.

    When I’m on a mobile / low bandwidth device, I’m looking for something specific and actionable. I’m not approaching that search with the same expectations as I would on my larger-screen, higher-bandwidth PC.

    You say you see desktop versus mobile as two webs. I’d agree with you if the only other domain were .com. But it’s not … it’s one of dozens designed to meet specific user expectations. I don’t see why .mobi shouldn’t be embraced in the same way — a tool to help users reach content they know will work (in this case, on their phones).

    And despite whatever studies people may have been quoting at MoMo London, I’ve seen enough other studies to convince me that people are not using their mobiles to find a replicated version of PC-based site. Instead, I believe they’re looking for actionable, context-based information — a door into the wide-open world for mobile that you’re suggesting in your last paragraphs above.

  • I completely agree that “we need to look for things that are truly different on mobile – not different just because of screen size etc.” You understand that the nature of the mobile space and in turn the expectations of its users are unique.

    In that regard, then, I’m surprised that you see supporting .mobi as supporting a second web. I don’t see people having issues with, say, amazon.com versus amazon.co.uk versus amazon.fr versus amazon.co.jp, etc.

    I believe people understand that different domains exist to serve different purposes and meet different expectations. I don’t see why .mobi is any different.

    When I’m on a mobile / low bandwidth device, I’m looking for something specific and actionable. I’m not approaching that search with the same expectations as I would on my larger-screen, higher-bandwidth PC.

    You say you see desktop versus mobile as two webs. I’d agree with you if the only other domain were .com. But it’s not … it’s one of dozens designed to meet specific user expectations. I don’t see why .mobi shouldn’t be embraced in the same way — a tool to help users reach content they know will work (in this case, on their phones).

    And despite whatever studies people may have been quoting at MoMo London, I’ve seen enough other studies to convince me that people are not using their mobiles to find a replicated version of PC-based site. Instead, I believe they’re looking for actionable, context-based information — a door into the wide-open world for mobile that you’re suggesting in your last paragraphs above.

  • nic

    Thanks for your comments people. As bandwidth and devices improve maybe whether it is mobile or fixed will be replaced as the most important distinction by something else – for example function – to take James’s point.

    Vance – to my mind country based domains give an indication of language that is critical to every user. Adding a further distinction of .mobi starts to make the picture very complicated. If I am directing a user to a site do I need to give him two addresses so he knows where to go if he is on mobile or on desktop? And then what if he is in France – will there be a .mobi.fr??

  • nic

    Thanks for your comments people. As bandwidth and devices improve maybe whether it is mobile or fixed will be replaced as the most important distinction by something else – for example function – to take James’s point.

    Vance – to my mind country based domains give an indication of language that is critical to every user. Adding a further distinction of .mobi starts to make the picture very complicated. If I am directing a user to a site do I need to give him two addresses so he knows where to go if he is on mobile or on desktop? And then what if he is in France – will there be a .mobi.fr??

  • >

    So how about I expand my example to consider domains like a .travel or a .biz or a .aero or .coop or .museum? My point is the same as yours … the domain sets an expectation as to the content.

    >

    Do most business cards have one phone number? I usually see a main office number, a personal office number, a fax number and a mobile number. Each one sets a different expectation as to how reach someone.

    But I’m more interested in your .mobi.fr .comment … does this lead into a way to tying mobiles to network recognition for handling to location-based services so country codes become irrelevant for the mobile?

  • >

    So how about I expand my example to consider domains like a .travel or a .biz or a .aero or .coop or .museum? My point is the same as yours … the domain sets an expectation as to the content.

    >

    Do most business cards have one phone number? I usually see a main office number, a personal office number, a fax number and a mobile number. Each one sets a different expectation as to how reach someone.

    But I’m more interested in your .mobi.fr .comment … does this lead into a way to tying mobiles to network recognition for handling to location-based services so country codes become irrelevant for the mobile?