On Oct 18, 2007, at 16:23, Christopher Allen wrote:

> Very interested post by Chris Messina, who was one of the organizers
> of iPhoneDevCamp
> http://factoryjoe.com/blog/2007/10/17/did-the-web-fail-the-iphone/


Folks,

Here is my response to Chris' post.

<http://www.ddg.com/~awd/blog/2007/10/19/did-the-web-fail-the-iphone/>

Andrew

____________________________________
Andrew W. Donoho
[EMAIL PROTECTED], PGP Key ID: 0x81D0F250
+1 (512) 453-6652 (o), +1 (512) 750-7596 (m)

"To take no detours from the high road of reason and social  
responsibility."
     -- Marcus Aurelius


Did the Web fail the iPhone?
October 19th, 2007
Chris Messina has written a thoughtful but, I think, wrong post about  
Apple changing course and allowing native applications on an iPhone.

The Issues:

The first major issue that we all have to get over is whether Apple  
was planning all along to allow the iPhone to run native  
applications. Because Steve Jobs has always desired closed systems,  
we can be pretty sure that his plans for AJAX development on the  
iPhone were the main plan. (Most everyone I know who has ever worked  
with “The Steve” agrees that he does not really want developers on  
any platform he’s built.) Now a technical reason that this wasn’t  
planned is that the platform is actually new. It is not a Mac nor is  
it NextStep. It may exploit that codebase but it does not claim to  
have any guarantee of compatibility with that base. Furthermore, they  
are not supporting anywhere near as many frameworks as the full Mac  
platform. Nor were they guaranteeing any kind of robustness outside  
of the APIs they use in the way they use them. In other words, their  
testing budget is restricted to their applications. As we know from  
Brooks’ “Mythical Man Month”, moving a codebase from being just a  
product to being a platform can take upwards of three times the  
effort - test and documentation alone are significantly larger  
expenses for a platform. Native apps were never ‘in plan’.

Second, Apple’s experience with the Dashboard in Tiger would lead  
anyone to believe that you can build excellent ‘native-looking’ apps  
using just web technologies. In my opinion, dashboard widgets are  
extremely impressive. I believe that supporting these widgets on the  
iPhone was always in plan and still is. Since Dashcode is fully  
supported in Leopard and, for reasons I detail later, the iPhone will  
use the Leopard kernel and runtime in early 2008, the Dashboard will  
emerge as a second programming model on the iPhone. In fact, there is  
a good chance that it could be released in the firmware update  
following Leopard’s introduction.

Third, the iPhone is truly the first phone to expect an always-on,  
effective internet connection. This is a major competitive advantage  
for the iPhone over every other phone and also gives AT&T an edge  
too. Web apps are the way to exploit this always on connection. A  
native app, while perhaps compelling for a user, does allow the 3rd  
party community to ignore this advantage of the phone.

Apple changed course in the face of customer pressure and that AT&T  
publicly disputed “The Steve’s” claim that you needed to isolate the  
network from rogue programs. AT&T has obviously had to engineer their  
network to be robust in the face of both fraud and outright attacks.

Finally, by allowing folks onto the platform in a controlled way,  
they directly remove most of the pressure to hack the phone from  
their customers. Most people don’t care about the lock-in until they  
get to the end of their 2 year contract. Then, I expect you will get  
a clamor from the iPhone hoi polloi for a formal, supported unlock.

Did the Web fail the iPhone?

Chris spends a great deal of time complaining about deficiencies of  
the Web platform - that collection of standards implemented by  
browsers. Yes, he is right. The web is deficient in may UI elements.  
Yes, the overly complex CSS environment lacks many natural layout  
patterns, such as columns. I support him in his call for web  
frameworks to redress these web weaknesses. Nonetheless, the web  
platform will move slowly because of the necessarily conflicting  
agendas of the browser developers. The browser war is still with us  
and always will be. Apple and Nokia combined with Firefox now have  
enough leverage to bring Microsoft to the web standards table but  
that doesn’t mean this will be a quick process. Furthermore, having  
actually been on multiple web standards activities (XHTML, SVG and  
XForms), I know firsthand how parochial committee members are and  
their attachment to pet features. That is part of why CSS is such an  
ornate and baroque standard. That said, the implemented subset of CSS  
features on most browsers is rather capable. It is good enough for  
almost all structured data applications and is pretty good for free- 
form data apps.

iPhone Futures:

iPhone is moving to the Leopard kernel and runtime. That is, in my  
opinion, part of why Leopard was delayed. They had to rationalize any  
differences at the low level between the platforms. We also have “The  
Steve’s” public acknowledgment that there will be an application  
’sandbox’. This is a Leopard-only security feature. If I was putting  
constraints on the iPhone runtime, I would require that all apps use  
the new Objective C 2.0 runtime and further require all apps to use  
memory garbage collection. In other environments this is a proven way  
to make applications more robust - not bullet proof but robust. Isn’t  
everyone tired of iPhone Safari crashing on them?

By the way, the rumor mill is rife with speculation that Adobe’s  
Flash Lite will appear on the iPhone with the next release. The walls  
separating the iPhone from being a full fledged internet platform are  
all falling down.

Summary:

While I understand Chris’ frustration with Apple’s decision and his  
fear that web app development on the iPhone will grind to a halt, I  
think he is too alarmist. The always on nature of the web on the  
iPhone and the switch of many web sites to rich AJAX applications  
will always bring a set of interesting features and mash-ups to the  
iPhone. (Web app folks will try to avoid special coding for any  
device.) The iPhone platform just got richer. It didn’t lose a  
constituency but gained a new market and mechanism to lock customers  
on the iPhone. Apple will enjoy added profits from this switch. And  
even more than a closed platform, “The Steve” loves profits.



--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"iPhoneWebDev" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/iphonewebdev?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to