Live data from Hacker News

IPhone =/= Debian app

lwn.net

71–80 of 81 posts

Re: IPhone =/= Debian app

#71
post #46

While I agree with this point completely, I also think bitching about Apple's iPhone and Ipad "apps" is a moot point. 5 years from now it's all going to be html5, and all of these apps will run on any mobile device web browser. Apple now has such a head start on their app platform that pretty much the only reasonable strategy their competitors can do is collude and standardize on html5 for all apps. Then Apple has a…

Ah, 3d games and image manipulation. The last resort of all native arguments. This is going to change pretty rapidly with native client I think.

Low latency audio, hardware access, virtual memory, et al aren't handled at all or well by NaCl. So something on the scale of Photoshop with it's ability to edit 250MB+ images is nigh impossible with NaCl. Something on the scale of Logic Audio or Cubase, impossible.

Also, with a network bottleneck, mutli-GB 3D games will be unpleasant at best.

Maybe in a decade though.

Re: IPhone =/= Debian app

#72
I think that the iPhone/iPod Touch/iPad platform has taken off because the mobile Web is just painful to use right now in those tiny browsers.

Apple's fulfilling a niche that AOL did in the '90s. Where AOL provided training wheels for the Internet, Apple does the same thing for mobile computing. Unfortunately, AOL didn't figure out that users would eventually outgrow the training wheels and discover the open Internet.

I guess the same thing will happen with the mobile Web once browsers improve and people start to learn to design mobile Web sites for mobile devices. I think Apple realizes this, since they're putting effort into HTML 5.

Re: IPhone =/= Debian app

#73
post #46

While I agree with this point completely, I also think bitching about Apple's iPhone and Ipad "apps" is a moot point. 5 years from now it's all going to be html5, and all of these apps will run on any mobile device web browser. Apple now has such a head start on their app platform that pretty much the only reasonable strategy their competitors can do is collude and standardize on html5 for all apps. Then Apple has a…

Ah, 3d games and image manipulation. The last resort of all native arguments. This is going to change pretty rapidly with native client I think.

Once you've overcome the current limitations of HTML by adding native code, offline mode, hardware support, local file access, and so on, is it really meaningful to call the result a "web" app? It's more like a native app written to a cross-platform layer (huh, somebody told me those were evil) that you can easily install from a web page.

Re: IPhone =/= Debian app

#74

Earlier quoted context omitted.

I don't see what this has to do with HTML5. Could you elaborate?

I think he means that HTML5 is a hedge against the coming collapse in the number and quality of native iOS apps from 3rd parties.

"Number and quality"

What are you talking about? There can possibly be few millions more of useless apps?

Re: IPhone =/= Debian app

#75
post #11

What about the barrier to entry: $100 + a Mac + an iPhone. As compared to a computer, which pretty much everyone had by 1998. Also iPhone programming is way harder than HTML. I had a mandatory high school class that taught HTML. Maybe I'm underestimating how many people are jumping on this, but my intuition is it can't be as many as were jumping on web. Twitter/social media marketing however... :P

When I began programming in the late 1980s, we learned Turbo Pascal running on 80286s that cost around $2000. I'd call $2,000 worth of hardware in 1980s dollars a pretty high barrier to entry, wouldn't you? Now your only barrier to entry is a Mac mini and an iPod touch. Both of which can be had for less than $1,000 in 2010 dollars. The real barrier of entry is about 25% of the price we paid in the 1980s.

Mac Mini $699 + iPod Touch $199 + iPhone Developer program $99/year = $997

$997 is a lot more than 25% of $2000

Re: IPhone =/= Debian app

#76

Yes, just like the dot com boom, this growth is unsustainable. But again, just like the dot com boom, once the market has matured to blow away the chaff, what remains will have proven pretty game-changing.

The step in that process you left out was "years-long massive earthquake-level crash leading to complete economic collapse of the micro-economy under discussion", the kind where most of the casualties are all the smart non-founders in the industry. Been there, done that. The primary reason I'm involved in this community is that I was in London before the first crash. Us internet people had been there for a decade or…

You speak much truth. And yet, you should not give in to hate. That way leads to the Dark Side.

-- posting for my friend Yoda who cannot reach the keyboard

Re: IPhone =/= Debian app

#77
post #75

Earlier quoted context omitted.

When I began programming in the late 1980s, we learned Turbo Pascal running on 80286s that cost around $2000. I'd call $2,000 worth of hardware in 1980s dollars a pretty high barrier to entry, wouldn't you? Now your only barrier to entry is a Mac mini and an iPod touch. Both of which can be had for less than $1,000 in 2010 dollars. The real barrier of entry is about 25% of the price we paid in the 1980s.

Mac Mini $699 + iPod Touch $199 + iPhone Developer program $99/year = $997 $997 is a lot more than 25% of $2000

Time value of money. $2000 in late 80s is a lot more than $2000 today.

Re: IPhone =/= Debian app

#78
post #54

Earlier quoted context omitted.

Dude that makes no sense, sorry. Re-read my comment; my point is: high number of apps does not correlate negatively with importance of platform or value of whatever is created on that platform.

But it doesn't correlate positively either. The 'article' is in response to someone claiming that having X apps created in Y years obviously means that a platform is 'more quality' than another, which is rubbish logic.

I respectfully disagree.

Re: IPhone =/= Debian app

#79
post #51
post #8

Earlier quoted context omitted.

I keep wondering why Jobs keeps pushing HTML5. He too must realize the unsustainability of it all. A collapse has to happen -- and then how many "real" apps will remain for sale? And how many devs will migrate to the web where they don't have to have the approval of Apple or Google or any other hardware maker in order to create and publicize/sell something?

Jobs never liked the idea of Apps in the first place (probably because he knew it was unsustainable); he was fully prepared to go with a pure-webapp strategy until developers revolted. He's probably trying to get back out of that market as quickly as possible.

Is that the case? I thought it was more of an issue of the SDK and the APIs not being ready at the time that they wanted to launch.

I can't imagine that Jobs hates the notion of apps.

On the other hand, I don't understand why anyone involved with the Mac platform, where uniform interfaces are a big deal and microcriticism of widgets is so common, would ever champion the Web as a platform.

Re: IPhone =/= Debian app

#80

Earlier quoted context omitted.

Ah, 3d games and image manipulation. The last resort of all native arguments. This is going to change pretty rapidly with native client I think.

Once you've overcome the current limitations of HTML by adding native code, offline mode, hardware support, local file access, and so on, is it really meaningful to call the result a "web" app? It's more like a native app written to a cross-platform layer (huh, somebody told me those were evil) that you can easily install from a web page.

The names "web" app and "native" are far too coarse grained to be meaningful. The important parts are the details.

Zero install or heavy weight install? Cross platform or single platform? Offline accessible or not? Safe for anyone to use, or not? Deep linkable or not? Etc.

These attributes can be mixed and matched. And will be.

Post reply on HN