[Discuss] 30% Apple

[me at mattgillen.net (Matthew Gillen)]

On 08/11/2011 09:38 AM, Ian Stokes-Rees wrote:
> Having just, unexpectedly, received an iPad, I've gone somewhat App
> crazy and spent more on software in the past week than I have in 5 years
> (in $0.99 increments).  It has made me painfully aware of how many apps
> don't need to be a standalone app (especially the ones that really only
> work with a network connection).  They can just be device-specific HTML
> content with JS+CSS3+HTML5+LocalStorage.  I'm sure an advantage of the
> "app" model is the device-locked nature.

I came to a similar realization regarding Android apps recently.  It is
true for many apps that are merely better interfaces to a website than
the website normally has.  However, there are some things that would be
a PITA to do in HTML5 that would work across browsers (and by that I
mean the ipad's built-in browser vs. android's built-in browser vs the
various third-party browsers).  For instance, NPR's news app has a nice
interface for scrolling between stories (swiping from side to side).
Not to mention every app developer's predilection for wanting to know
your location so they can deliver locale-specific data...

> Anyway, my prediction is that give it a year or two and everyone will
> realize how painful it is to write a different app version for every
> device and if they can manage it they will go for "responsive web
> design" (http://ethanmarcotte.com/ and
> http://www.abookapart.com/products/responsive-web-design), and you'll
> access most of todays apps through the browser.

I totally see your point, but I disagree that it's a foregone
conclusion. On the one hand you have this burgeoning
embedded-device-developer group, versus more "traditional"
web-developers.  Given that the device-developers have a nice foothold
now, they probably have a year or two to prove their worth above and
beyond what the web-developers could do.  There are a lot of input and
data sources on these devices that would just be difficult to make use
of via a browser+HTML5 (you run into the same problem web developers
always run into: someone else controls the browser and all it's
associated bugs/features).

It may be the case that IE6 and all its brokenness burned so many web
developers and IT dept. heads that they are enthusiastic about having
more control (ie not relying on a browser), because it turns out to be
cheaper for them (to not have to work-around brokenness, or to at least
have control over the brokenness).

> For my own interest, here is the list of apps I've downloaded (and
> possibly paid for) that I think would be an easy transition to the
> web-based model:

Another complicating factor is that there are several artificial market
distortions that may drive this one way or the other regardless of the
technical merits.  For instance, the rule Apple is trying to enforce
about taxing in-app purchases could significantly push certain
content-makers to prefer the HTML5+browser approach.

Matt