The Mobile Browser Is Dead, Long Live The App
forbes.com
forbes.com
Time spent is a poor way for anyone to evaluate a market or even a single product. It could mean anything from "wow this product is amazing, let me stay on it all day" to "wow this product sucks, and I can't find anything but I need to so I'm going to spend 3 hours on it finding what I need".
Its an even worse choice to then extrapolate that metric into high level prognostications about what does/doesn't matter in the space.
Mobile Web has a role for consumers and companies. Mobile apps, like packaged software of the past, also have a role for consumers and companies. The value is in understanding why they have different use cases and building software that delivers value to users for what they need right now.
But hey, if this poorly written article resonates with you, feel free to listen. And the rest of us will laugh at you all the way to the bank.
However, they aren't the future, because:
- Web apps are always up-to-date, no updating necessary
- Web apps are stored in the cloud, not on your device
- Web apps work on any device/platform, apps need to be written for each
- Web apps are there on demand, instantly, while vendors cannot expect you to download an app for every website, company and product.
Look at what Google is doing with the Chromebook and look at devices like the Synology where the best feature is the Web app interface served from the device itself. This is the future, the "Internet of Things".
Mobile Apps are the future because:
- Mobile apps allow you to keep the version you have, updating when desired
- Mobile apps are stored on your device
- Mobile apps are customized to your device and can take advantage of all its features
- Mobile devices provide a browser for things that shouldn't ever be an app to being with.
I hope web based apps are the future but I think it will be for different reasons.
- The promise of Java, they won't be written in Java but just like Java the idea that you can write portable applications is very powerful
- Not being locked into an app store. App stores will still exist for discovery and monetization reasons but being able to run an app by visiting a URL will allow for all the things that companies like Apple won't allow.
Do you really believe that the user is actually in control of the upgrades ?
> Mobile apps are customized to your device and can take advantage of all its features
Having to code specifically for each device sounds like a benefit to you ?
On iOS at least, it's (by default) a manual process to update an app.
Having to code specifically for each device sounds like a benefit to you ?
Benefit for the user, yes.
I would agree with that if everyone (companies) actually did that, but they don't. Often they pick the popular platforms, or choose frameworks that allow for a single codebase and output a often subpar UX App, or just don't make the app to begin with.
Making it harder for a company to give the end user the product, even if an arguably better product, is not always a win for the user.
If you can be successful on iOS or Android first, you'll probably then have the resources available to develop for the other platforms as well.
To me as a developer? It isn't, it is a pain in the ass. But to a user if your web app doesn't support random cool hardware feature of their phone they'd rather have a native version (see samsung s-pen stuff)
> Web apps are stored in the cloud, not on your device
Put another way, web apps can only be run when you have connectivity. Of course, there are kludges that allow for some functionality offline, but they're not great just yet.
OTOH, some apps cede their advantage here by unnecessarily relying on network connectivity, even when they could avoid the dependency.
> Web apps work on any device/platform, apps need to be written for each
Except where they don't because of browser quirks, and thus need custom code written to work around those quirks. Is it better than native apps for each platform, assuming the functionality fits your needs? Sure, it can be, in much the same way that Java is for client side apps.
Bottom line, web apps vs. native apps is not black and white. Suitability of either approach to a task varies now, and the line where one makes more sense than the other is a moving target.
The line only gets more fuzzy when hybrid apps are considered.
But strangely in Windows 8.1 (not phone yet) you can make Windows store apps using WebGL and JavaScript. FireFox OS also support WebGL.
Until web apps can run in a first-class embedded Javascript runtime with equal access to the OS's native frameworks, web apps are going to continue being second class citizens. Part of me thinks they're only popular now because developers like working on them.
If they have equal access then they are not webapps anymore,just native apps written with web technologies.
A HTML/JS app that doesnt live in the browser is not a web app.
"Web apps" are websites, that take advantage of web apis(ajax,...). Web apis can only be found in browsers.
Devs cant have it both ways,they cant get native feel,performance and portability. Something portable means something trade-offs regarding the native api.
The problem exposed with the op is,do users like web apps? especially on phones,where resources are limited,users impatients and used to the os features and where poor performances are not tolerated.
I believe if one really cares about user experience ,one should write native apps,where the core UI isnt a webview. now if your "app" can be a mobile website,no need for an "app" at first place.
I think what people are ignoring is that a lot of what we use has moved from document type sites to application type sites. I browse HN fine with mobile browser, but for almost anything else, I want the app version.
Open services promote integration and availability of things you want to do.. for example I can't instapaper an article in my safari browser, I am sure I could've fixed that if there'd been a layer of script language between OS apps and web apps. It is only a sliver but it has been annoying me quite a bit. If the mobile ecosystem will be scriptable and integratable then one can make web be a first class citizen on the mobile. For now browser is a pretty crummy way to do web on a mobile device. Screen is too small, value added features are only available to apps only. Maybe for a good reason - technological one, battery and security(those are the problems to solve). Yet I would like to program my alarm clock from the web and have it send me some news for the time when I got up ... something like that (I am brain storming here)
2c
I'm sure there are masochistic people who write code for browsers who'd love to maintain the browser-API abstraction over N OS API implementations. I'm not so sure there is a sufficient number of them to get such a task off the ground. It's not some trivial matter of simply implementing something like an HTTP-based API.
I prefer this mindset instead:
Love that quote, and the signature even more!!
The remaining 32% is split between 14% in the browser and 18% in apps. That's not the same thing as the mobile browser being dead.
> but I understand the utility of having applications. But they contribute to a tunnel vision of what a smartphone can do. They provide a good user experience, but poor integration. A smartphone is smart if it helps users day to day
He didn't mean it to contrast native applications vs browser based ones. He meant it to contrast standalone / non integrated native applications vs integrated experience where the user doesn't see it as "working with application A", but sees it as "doing task A". Native vs Web wise, Jolla's Sailfish is actually heavy on the native side.
> The methodology of Android and iOS is the dominant viewpoint.
It's a not a reason to copycat it and to be a follower rather than a disruptor. Sailfish and FirefoxOS don't follow a lot of aspects of the established iOS and Android approaches and it's a good, not a bad thing.
Apps work best because they do not carry baggage of the desktop world with them. People have the freedom to build specifically for mobile with mobile tools.
There are no two bigger movements that contribute to this than: Responsive Design and the Mobile first.
Responsive Design implies that mobile is just different from desktop based on the display size and that you can design a single interface that can work for both. Both of these ideas are completely wrong.
Mobile First implies designing for mobile and then desktop is the best approach but this can never work as both platforms will suffer from inferior implementations.
----
Setting the ship back on course would take us building "Mobile Only" with mobile only frameworks, by dropping the ones that were created in the desktop era.
It's been clear for a while that monetising apps is easier than the web, though the discovery problem is intense.
What I've not seen, and would be interesting, is how different this is for iOS and Android users. i.e. do iOS users spend more time on apps vs in browser than Android users?
Actually I prefer the HTML5 approach, but as you said, monetising apps is easier than the web. I believe this is the key issue here.
They both have their up-sides and down-sides, but which is best usually depends on the context in which they are being used, and if nothing else, user preference.
I don’t see why one has to “win” per-say … it seems to me like the future is big enough for both to thrive :)
mobile versions of websites are pretty much a joke. I can't stand what amazon does when I try to visit with opera (forces it to mobile). Makes it pretty unusable.