Pragmatism in the real world

Mobile web monopoly

There’s a recent post by Peter-Paul Koch called The iPhone obsession about how Mobile Safari is being treated by web developers as the only web browser to develop for. PPK likens this to how we all used to only develop for IE6. Unfortunately, the article has lots of hyperbole and iPhone hate which significantly detracts from the actual message.

The fundamental point is that Mobile Safari is not the only web browser available on phones and if you are serious about providing mobile web to your users, then developing for the other 85% should be part of your modus operandi.

However (and there’s always a however!) it’s not easy. It’s trivial for me to have multiple web browsers and multiple operating systems on one computer for testing. It is not trivial for me to have 10 phones with different browsers and screen sizes. We don’t do enough mobile work to justify the costs.

What actually annoyed me was Rentzch’s response. Rentzch started by saying that Koch’s argument was “stupid/lazy/undisciplined” and then goes on to say:

I see a quiet revolution of mobile developers waiting for other phones to catch up to the iPhone.

and

I think mobile developers intuitively grasp that accommodating the insufficient software+hardware of non-iPhones out there will only prolong the status quo, extending needless suffering for developers and users alike.

That’s exactly what we all said when we wanted Netscape 4 to die because IE was so much better. Rentzch has made Koch’s point for for him whilst trying to disagree.

For what it’s worth, I think that any mobile device manufacturer who wants web developers to develop for their device’s browser, should provide hardware that doesn’t need a plan. It doesn’t even need to be a phone. Mobile Safari is the easiest mobile browser to test against as it’s available in the iPod Touch.

7 thoughts on “Mobile web monopoly

  1. I thought one of the main points for Safari on the iPhone is that it's a full version of Safari 4 (Without flash, etc).

    Surly the best approach for web developers is to develop standards compliant code, and let the device manufacturers sort out the rendering!?

    Are we actually going to learn from our mistakes, or are we just going to keep targeting todays "fun" platform, only to be stuck with repeating campaigns to "ditch" netscape, IE6 (7,8 & 9 ;-) ), etc?

  2. There's a world of difference between a design that works well on a greater than 800px wide screen and one that works well on a 320px wide one.

    The whole zooming in thing is "okay", but not as great as a site that's intended to work well on a small screen.

    You'd certainly hope that standards code would enable you to build one site for "small" screens that worked across all devices though!

    Letting the device manufacturer sort out the rendering is fine as long as you don't mind a sub-standard result – c.f. IE6 :)

    Regards,

    Rob…

  3. If you are doing hardcore mobile development and really need to test in multiple devices, we found a service called DeviceAnywhere (http://www.deviceanywhere.com/) that looks to provide just that. Be warned, it's not cheap at all, but it does afford you access to multiple devices on multiple networks to test against. And in the long run, it is cheaper to do that than purchase a bunch of phones and plans for testing.

    We've been going down the mobile web route for the past year, and do try to develop a user-friendly interface for 3 main device types: Web-kit phones, Smartphones, and then the old-school text-browser-only phones. We do device detection then swap layouts based on the type of device viewing the site. It has served us well so far.

  4. I fully agree with ppk and you on this point – mobiile web development != mobile safari development. And that's what I'm preaching to the marketing folks at my company.

    However a small remark. You claim, that mobile safari is the easiest to test against, because of the existence of the iPod touch. While half true, it ain't accurate. Mobile webOS safari is the easiest to test against, since you can download the VM for _free_ and have a fully blown QUERTY to type in any kind of weird URLs. Same goes for mobile Android, I'm not sure if there is an emulator for it.

  5. Yeah, I find the best part of developing for Safari is that you can develop for Android on a linux box and it will more than likely work on Safari. I develop for webkit.

Comments are closed.