3 ms·

This is the ugly head of the post-web world. Let's for a moment assume that the web is being phased out in favor of a constellation of apps on a closed OS gated

This is the ugly head of the post-web world. Let's for a moment assume that the web is being phased out in favor of a constellation of apps on a closed OS gated by a monopolistic appstore/playstore. The end game is for a Microsoft of modern days to form, with a far stronger power than MS in the 90s. Since every player in the market knows this, there'll be fighting either for the top spot or, at least, to keep the market fragmented enough to prevent the new monopoly from forming. In either case, consumers get screwed. Either they get screwed like in the 90s, or they get screwed by systems incompatibility and fragmentation. All in all, this war is moot. Moore's law applied to bandwith, to device computation speed and to software evolution leads to this: http://xkcd.com/1367/ http://xkcd.com/1367/

pjmlp 12y ago
I feel more screwed with TTY with fancy graphics apps, than native and better OS integrated experiences.
Tyrannosaurs 12y ago
> All in all, this war is moot. Moore's law applied to bandwith, to device computation speed and to software evolution leads to this: http://xkcd.com/1367/ http://xkcd.com/1367/ Not really. If you have massively powerful devices and massively fast connections then you move to a situation where you've still got to address differences between the programming models and and the feature sets on the various devices. For a web app to have parity with a native app there are still significant technology gaps which need to be closed - offline working (moving in the right direction, still not great), native look, feel and functionality (difficult if you want it to run on multiple device types) and so on. Some of those will close over time, some may not. But harder to close off - I struggle to see how cross platform will every be much more than lowest common denominator given the consensus required on how to adopt new features as they arrive on phones. Apple launches TouchID, Samsung launch fingerprint scanning. How long before the W3C define a standard for that and the companies adopt it? How much time is that where native developers are at an advantage? The web is great and web apps will definitely be a bigger part of things going forward, but I think it's simplistic to assume that they'll take over. As for a present day Microsoft, I don't see that happening. Back then interoperability was a nightmare - if the person I wanted to work with wasn't running the same OS as me I was basically screwed. Right now I can collaborate with someone and generally speaking I don't care what they're running and it really doesn't matter.
- pekk 12y ago
  The technology gaps are going to take forever to close for two reasons. First, because the web as an app platform is a disorganized pile of bad hacks, which makes every fix harder. Second, the great point you made about how the web is bound by waiting for standards, then waiting for widespread compliance with those standards. This guarantees it's always far behind. And the pile of crap is only growing, making it more and more appealing to just leave the web behind for anything that makes it easier to get nice results.
  - sergiosgc 12y ago
    > The technology gaps are going to take forever to close for two reasons. (...) The secret is that the utility curve of technology is not flat. For example: when processors ran at 4Mhz, doubling the processor speed to 8Mhz had a much different utility value than when they doubled from 1.5Ghz to 3Ghz. As such, even if the technology gap doesn't close, ever, the utility values obtained from each technology will close, to the point of indifference. You can already observe some of this utility curve flattening, if you observe the absence of game changing feature announcements on WWDC and Google I/O.
    - Tyrannosaurs 12y ago
      We can agree or disagree about whether anything announced recently is significant (I'd suggest if you're a user then application interoperability is major and as an iPhone 5S user I find entering a PIN after using the fingerprint scanner feels positively stone age), but I'd struggle with the idea that smartphones are essentially feature complete and nothing significant will be coming down the line in the next few years.
pekk 12y ago
That xkcd comic's idea would work if people would use web apps in preference to 'native' apps. Moore's law isn't fixing that unless it somehow makes the web into a hugely better platform for apps. As it stands, we are limited to only using these lame DOM-based APIs which have been strained far past their origin in hypertext, and only using Javascript, which has been strained far past its design as a way of merely scripting a browser. These contained some good ideas but they're essentially dead. People want apps. So the original intent is warped to fit that demand, and the result is neither fish nor fowl. It's mystery meat. If the web weren't so crappy, the monopolist would not really have an opportunity. Mobile platforms bring a lot of value because the web is a bad app platform; where you have the choice, keeping HTTP but using other client technologies is the best compromise available. Mobile platform creators had that choice and went with decent languages and more appropriate environments. Where a graphics API can be just a graphics API, and so on for everything else. We can count on Firefox OS and anything like it to fail to thrive for as long as it is Mozilla against the world, defending the awkward status quo by throwing more lame hacks on the pile and awkwardly bolting on new device APIs to its browser. Don't tell me that if I can use a decent language and decent APIs rather than Frankenstein's monster, that I am "screwed by systems incompatibility and fragmentation." The web status quo sucks at a technical level and that is why it is threatened by monopolists. Doubling down on the technical suck isn't going to help. If you want a benevolent monoculture to emerge, you aren't going to get it by doubling down on the status quo and waiting until everyone gives up mobile platforms for ideological reasons, or because a cheap phone uses Javascript.
- sergiosgc 12y ago
  > The web status quo sucks at a technical level (...) I know you mean that it sucks when viewed from the perspective of an application developer, and I agree with you. However, on a technical merit perspective, the web is much more evolved, and much more a technical feat, than either Android or iOS. Android and iOS are a reiteration of what has been done cyclically through times, ever since the first OS got written. A set of libraries and associated APIs on top of a constrained set of hardware. The web is, as an application platform, a set of APIs. That's it. There's no defined hardware, no defined set of libraries. Guess what? Designing the API alone is much more difficult, a lot harder to get right, and much more of a technical feat. On one side, some libraries got written and then documented. On the other, everything is defined conceptually (well, not really; if done like this it would be really impossible instead of just difficult, but the point stands).
ZeroGravitas 12y ago
Ideally you get a "coalition of the losers" which pushes the open option, but currently in smartphones you have two strong winners (Android and iOS) who see no reason to change and a few terribly weak losers (Windows, Firefox, Tizen) who have no hope of shifting the needle. On the other hand smaller battles (like mobile browsers, open maps, open location data etc.) might be won.

pjmlp 12y ago

I feel more screwed with TTY with fancy graphics apps, than native and better OS integrated experiences.

Tyrannosaurs 12y ago

> All in all, this war is moot. Moore's law applied to bandwith, to device computation speed and to software evolution leads to this: http://xkcd.com/1367/ http://xkcd.com/1367/ Not really. If you have massively powerful devices and massively fast connections then you move to a situation where you've still got to address differences between the programming models and and the feature sets on the various devices. For a web app to have parity with a native app there are still significant technology gaps which need to be closed - offline working (moving in the right direction, still not great), native look, feel and functionality (difficult if you want it to run on multiple device types) and so on. Some of those will close over time, some may not. But harder to close off - I struggle to see how cross platform will every be much more than lowest common denominator given the consensus required on how to adopt new features as they arrive on phones. Apple launches TouchID, Samsung launch fingerprint scanning. How long before the W3C define a standard for that and the companies adopt it? How much time is that where native developers are at an advantage? The web is great and web apps will definitely be a bigger part of things going forward, but I think it's simplistic to assume that they'll take over. As for a present day Microsoft, I don't see that happening. Back then interoperability was a nightmare - if the person I wanted to work with wasn't running the same OS as me I was basically screwed. Right now I can collaborate with someone and generally speaking I don't care what they're running and it really doesn't matter.

pekk 12y ago
The technology gaps are going to take forever to close for two reasons. First, because the web as an app platform is a disorganized pile of bad hacks, which makes every fix harder. Second, the great point you made about how the web is bound by waiting for standards, then waiting for widespread compliance with those standards. This guarantees it's always far behind. And the pile of crap is only growing, making it more and more appealing to just leave the web behind for anything that makes it easier to get nice results.
- sergiosgc 12y ago
  > The technology gaps are going to take forever to close for two reasons. (...) The secret is that the utility curve of technology is not flat. For example: when processors ran at 4Mhz, doubling the processor speed to 8Mhz had a much different utility value than when they doubled from 1.5Ghz to 3Ghz. As such, even if the technology gap doesn't close, ever, the utility values obtained from each technology will close, to the point of indifference. You can already observe some of this utility curve flattening, if you observe the absence of game changing feature announcements on WWDC and Google I/O.
  - Tyrannosaurs 12y ago
    We can agree or disagree about whether anything announced recently is significant (I'd suggest if you're a user then application interoperability is major and as an iPhone 5S user I find entering a PIN after using the fingerprint scanner feels positively stone age), but I'd struggle with the idea that smartphones are essentially feature complete and nothing significant will be coming down the line in the next few years.

pekk 12y ago

That xkcd comic's idea would work if people would use web apps in preference to 'native' apps. Moore's law isn't fixing that unless it somehow makes the web into a hugely better platform for apps. As it stands, we are limited to only using these lame DOM-based APIs which have been strained far past their origin in hypertext, and only using Javascript, which has been strained far past its design as a way of merely scripting a browser. These contained some good ideas but they're essentially dead. People want apps. So the original intent is warped to fit that demand, and the result is neither fish nor fowl. It's mystery meat. If the web weren't so crappy, the monopolist would not really have an opportunity. Mobile platforms bring a lot of value because the web is a bad app platform; where you have the choice, keeping HTTP but using other client technologies is the best compromise available. Mobile platform creators had that choice and went with decent languages and more appropriate environments. Where a graphics API can be just a graphics API, and so on for everything else. We can count on Firefox OS and anything like it to fail to thrive for as long as it is Mozilla against the world, defending the awkward status quo by throwing more lame hacks on the pile and awkwardly bolting on new device APIs to its browser. Don't tell me that if I can use a decent language and decent APIs rather than Frankenstein's monster, that I am "screwed by systems incompatibility and fragmentation." The web status quo sucks at a technical level and that is why it is threatened by monopolists. Doubling down on the technical suck isn't going to help. If you want a benevolent monoculture to emerge, you aren't going to get it by doubling down on the status quo and waiting until everyone gives up mobile platforms for ideological reasons, or because a cheap phone uses Javascript.

sergiosgc 12y ago
> The web status quo sucks at a technical level (...) I know you mean that it sucks when viewed from the perspective of an application developer, and I agree with you. However, on a technical merit perspective, the web is much more evolved, and much more a technical feat, than either Android or iOS. Android and iOS are a reiteration of what has been done cyclically through times, ever since the first OS got written. A set of libraries and associated APIs on top of a constrained set of hardware. The web is, as an application platform, a set of APIs. That's it. There's no defined hardware, no defined set of libraries. Guess what? Designing the API alone is much more difficult, a lot harder to get right, and much more of a technical feat. On one side, some libraries got written and then documented. On the other, everything is defined conceptually (well, not really; if done like this it would be really impossible instead of just difficult, but the point stands).

ZeroGravitas 12y ago

Ideally you get a "coalition of the losers" which pushes the open option, but currently in smartphones you have two strong winners (Android and iOS) who see no reason to change and a few terribly weak losers (Windows, Firefox, Tizen) who have no hope of shifting the needle. On the other hand smaller battles (like mobile browsers, open maps, open location data etc.) might be won.