3) There's still a lot of features on the iPhone that you can only access through native apps.
He claims most of the apps he has could be web apps. But I just opened my iPhone to the Top 25 paid apps and 7 out of the top 10 could only exist as native apps (and the other three are debatable but in theory you could write them in Javascript)
Here's the list...
Voices: Requires access to the mic
JellyCar: Requires native app for soundtrack (which is actually a big deal in the game)
RedLaser: Requires access to the camera
Call of Duty: Graphics require a native app
All in One Gamebox: Graphics require a native app
Alarm Clock Pro: Has to be able to play music library
Metal Gear: Graphics require a native app
I can't really recommend that app, though. Most of it is open source, so you can go see for yourself that the code is terrible. The app is ugly, leaks abound, and so on. The guy really doesn't know objective-c. I became aware of it because I have a contracting client that wanted to use that app, discovered it was terrible, and paid me to write a better one.
I think this is an idea that is ripe for the picking, but I can't see how, really. I have most of such a browser written. Who can think of a way to monetize this?
Rhomobile has a Ruby-on-Rails inspired framework to run a scaled down Ruby 1.9 implementation that seems to act as an internal web server, allowing iPhone/Blackberry apps to be written in Ruby and HTML templates. The code you write seems to have access to native phone capabilities like GPS.
The main drawback of RhoMobile seems to be the lack of a polished feel, looking like the browser application it really is. (Wikimedia mobile uses RhoMobile for its iPhone app).
For such a critical component, youd really want something open source [and free as in beer also]... after all it needs to keep pace rapidly, be rock solid, trusted by all developers, etc.
[EDIT : correction.. Big5Lite is free, not sure what non Lite has. Big5 project is opensource it seems - http://code.google.com/p/big5/]
App developers who use it can compete while adding to this common component - it plays the role of the browser. It just happens to be a browser which gives you access to nice hardware features such as geo location, media library, contacts etc [policed by os].
Here lies the solution for some percentage of apps, between 90.0 and 99.95.
Streaming live video en-masses would likely exist* far before streaming games, given you can buffer video to compensate for latency. Games couldn't be easily buffered as they require instant feedback between button-presses and what's seen on the screen.
*Streaming video does indeed exist for cell phones, but not on a mass-scale, at least in the United States.
Six months later they cut back the free channel lineup so there's nothing interesting to watch.
Admittedly not for 3D intensive games. But maybe flash+browser will catch up with 3D. Only a year ago RaphealJS was magic.
As bad as Apple is at vetting and distributing software (when compared to the web they are bad a distributing software, at least in terms of speed), they are terrific, although expensive, payment processes.
The fact that every iPhone user necessarily has an iTunes account, and they are only one touch and a password away from paying you for your software, can't be overstated.
Apple releases a highly-portable computing platform (iPhone) and initially says "if you want to develop for it, all you have to do is write web applications". Developers rage against the irrational locking down of the system.
Google releases a highly-portable computing platform (Chrome OS) and initially says "if you want to develop for it, all you have to do is write web applications". Developers fawn over it and talk about the simplicity and openness of the system.
In other words: if writing a web app is sufficiently bad to not be worth the time on iPhone, why is it apparently generally held that it'll be sufficiently good to be worth the time on Chrome OS?
2) Your webapp for ChromeOS does not need to be specialized -- you don't have to account for fat-fingered multitouch on a small fixed screen.
It's far better to build up a real sustainable following yourself based on real merits. The app store is full of gimicky one time apps people will use once to show their friends, then never use again :/
Almost every successful startup you've heard of which publishes lots of content has an SEO strategy. Slideshare, Scribd (good golly were they aggressive with it -- good thing they got big before they got smacked), Yelp, etc etc.
As a distribution outlet, nothing (yet) can compete with the app store.
That said, I've been wanting to write a free guide to SEO aimed at HN-like companies, because I think its important. However, the response from HackerNews to anything involving SEO is so harsh, I've felt it would be a waste of my time. If that's not the case, please let me know.
The client see's it as a "iPhone app" you can host/sell it in the store, but it's really just a web app.
(Maybe there's a restriction on this, but I can't see why there would be...)
So this doesn't solve the problem.
If you think the web is popular because it's a cross platform sandboxed version of Access (which is essentially what you were suggesting) then you're the idiot. There's only so many useful applications you can make by simply storing and retrieving very simple data locally.
But giving this article any credence is giving it too much credit. It's shock-blogging and nothing more.
http://www.spreedly.com http://www.cheddargetter.com http://www.recurly.com
Currently the App Store is a benevolent dictator. I don't care how much lipstick they put on the pig, I want free speach. I don't care if that means 10% of apps will have a flag "may be spyware or inappropriate" but I want those apps distributed! EVEN if those apps help violate apple's own license terms. I want the courts to decide who takes shit down, not apple. Apple is a company whereas the courts are government.
1) With the appstore being the one authority to get apps on your iPhone, you will automatically get much less traction using the web
2) The appstore makes it VERY easy for the devs to actually get paid. If you launch your app web-only, you have to incorporate the whole payment process yourself, which sucks.
Catch 22...