Mobile shift: You’ve probably underestimated just how big this is
trigger.io
trigger.io
1. The author of this article clearly has money on one side of the issue since he's trying to sell a cross-device app development platform.
2. The data he supplies isn't broken down by usage type, so it's a little hard to analyze. What if 50% of the 'app' time is spent playing games? Or 20% of it is spent only in a couple of apps like Facebook? He's suggesting that native apps are supplanting web usage, but the graph actually shows web usage going up. So, are apps complementing or cannibalizing web usage? Hard to tell.
Comparing Minutes spent on apps or web sites may be a bit misleading; in this case why not put time spend on Office suites or Desktops on PCs too or me watching my Lisp code. Apps such as games, contents, utilities or web front ends have all different meanings to the users and with regard to monetization (ad or not). The same is true for web sites.
a. the time I spend with my utility apps are apples and oranges comparison to the time I spend reading HN on my Android.
b. Not all 'times spend` are monetizable, or monetizable in the same rate or way;
c. Mobile apps are used slower mainly due to small screen size, finger input.
d. Games are a peculiar category on their own.
Edit: that said; mobile has expanded our connected time to new places/times; toilet, elevator, waiting (queues), couch, TV-ad pauses...
Mobile web apps like Facebook prove that you can build a non-trivial app that runs in a web browser and have the experience be so good that people don't even care about the native app version (at least everyone I know just uses Facebook right from their mobile web browser).
A good domain name is still valuable. This is more true especially as more access to phone hardware gets opened up to JavaScript APIs right from the browser. You just can't beat making a change to your app on your server and having all your users get new versions the next time they visit. It's way better than waiting weeks for Apple to review and approve your changes.
That, and an app seems to have more authority than a website. "This is Facebook, and I know it." sort of thing.
I discovered it when I was at Starbucks. My session expired and the NewsFeed displayed the Starbucks login page instead.
And from what I understand, Facebook has done a lot of significant engineering to try and improve the performance while keeping it mostly HTML5.
Mobile web apps aren't ready.
Pure speculation, but I wouldn't be surprised.
In the case of most startups, native apps are unnecessary. I already see that feeling trend my geek friends ("Ah, not another of those sites that want me to download their app. I just want to see the [movie listings, items for sale, opening hours, etc.]").
Worth noting, also, that Facebook (and Twitter) is mostly just a window with a browser (hence why the mobile sites are very similar).
I think the point the original post was trying to make was that in deciding between Web and native app, only native apps will be necessary. In other words, why spend the time developing a Web app when everyone will use it on a mobile anyway.
(I just want to point out that there's a large part of my soul that is just torn over agreeing with this post, because I would much prefer an open Web to proprietary apps. And I built Web apps for years. However, denial does not become an entrepreneur.)
Until this exact moment, it never even occurred to me to use Facebook from mobile safari. I just tried it out and it is pretty impressive; it looks (and works) very much like the app but with a slower and less responsive UI. If given the choice, I don't see why anyone would use the web version over the app though.
1) Inertia (that's the way I've been doing it for a decade, dagnammit), and
2) It's an easier context-switch. Rather than go back to the home screen, swipe over to the next app, and tap it, you just just type in the first few letters of the url, then select the site you want to go to from the drop down list.
But
1) Has had pretty poor predictive value in the last few years, and there's no reason to suggest it'll improve. Our entire industry is built around disrupting inertia.
2) Is ripe for disruption. At the end of the day, when you get right down to it, entering the first few letters of a url then selecting the correct site is both an arbitrary, and not particularly optimized workflow. The fact that we are used to it does not mean there aren't far better systems out there. Sure, Chrome on my MBP has learned to take me to my top 10 sites with cmd-l -> $url[0] -> enter, but my mom needs her buttons on the bookmark bar. I tried to remove the titles, thinking she could easily recognize the favicons, but no dice. She needs a start menu-like interface.
Similarly, I was on the bus with my friend, browsing reddit. He buys tons of apps for his iPhone, and he looks at my screen and says, "is their mobile app free?" I said, "well, yes, but most people just browse to reddit.com/.compact". he stared at me in disbelief. Why would you use the browser if you can have an app?
For those of us who live and breath the Internet, it seems only natural that a write-once-run-anywhere website would be more appealing than having to download some stupid app from some stupid store, type in your password, then have to remember which home screen it's on, etc. For those who use the internet as casual content consumers, however, apps take a lot of the frustration out of it. Want to read the economist? Tap the economist button. That's it. Want to watch a youtube video? Tap the youtube button. Need directions? Tap the map button.
For the casual user, these workflows are far more intuitive. My friend, who has spent hundreds of dollars in the App Store, rarely uses the browser on his phone. He uses the Plenty of Fish app, the Chive app, etc. For most non-technical users, this just makes sense. It might seem annoying to us, but we are not the largest market segment by far.
Now it might be that we'll see mobile web sites with a light wrapper disguised as apps; perhaps cache and a link that opens in the browser. Maybe we'll see native support for the kinds of things that are typically done in the browser. But, crucially, what I don't think we'll see is the dominance of mobile web over native clients. For the vast majority of people out there, I think they just want the app.
If you made an "app browser" that allowed better searching (type the first letter and the app appears --- tap to open it), sorting (like bookmark folders) and easy navigation between apps, you'd be presenting me a much better experience than the current process of finding and switching between apps on your phone.
That said, I still think pure mobile web could be the future. Why couldn't Apple allow mobile web apps to be sold in the app store? It's an icon that opens a web page. It could be as simple as that.
Let's take it a step further. Imagine that when you download this new type of mobile web app, Apple calls a JavaScript API that reads the HTML manifest in the 'index' file and loads the resources in the background. Apple shows the downloading bar over the app icon while this happens, just like downloading an app today. When you click it, everything is loaded and ready to rock.
The iPhone has this. Just swipe right on the home screen.
> and easy navigation between apps
On the iPhone, double tap the home button and you get a taskbar of your running apps.
> Why couldn't Apple allow mobile web apps to be sold in the app store? It's an icon that opens a web page. It could be as simple as that.
Phonegap allows you to bundle a html/javascript webapp as a native iPhone application. There are even hooks in there to access the camera, gps, etc.
This is the main reason it won't happen; app store review is one of the key parts of Apple's positioning.
Although, some 'apps' solve this by adding an icon to the home screen or an embedding a web view inside a native app container. Sadly, I don't know enough about these solutions to critique effectively.
Dropbox, pre startup, would you even have paid the $10 for it? Not sure...but now? of course.
What makes a brandable domain? For one worders like stripe/path and all, it's easy, but dropbox? How could someone have predicted it's "brandability"?
I have to question this assertion made in the very first sentence...it does not pass the bullshit test and it's hard for me to trust anything else in this post unless this assertion is proven in detail.
And by "in detail," I mean that this metric is useless unless broken up by category. Is it possible that Facebook mobile app usage is more than Facebook standard-browser usage? Possibly. What about for news sites? I would say, no way. It seems that many of us at HN, even the productive ones, spend a chunk of the work day browsing. Imagine how much the average white-collar worker spends idling away.
Also, how much of the mobile app usage includes freemium/MMORPG/spec-intensive games? This is a category that many web services will never try to compete in?
--
FWIW, here is the site that the OP links to to justify this assertion: http://blog.flurry.com/bid/63907/Mobile-Apps-Put-the-Web-in-...
And this is what that site says: > In this report, Flurry compares how daily interactive consumption has changed over the last 12 months between the web (both desktop and mobile web) and mobile native apps. For Internet consumption, we built a model using publicly available data from comScore and Alexa. For mobile application usage, we used Flurry Analytics data, now exceeding 500 million aggregated, anonymous use sessions per day across more than 85,000 applications. We estimate this accounts for approximately one third of all mobile application activity, which we scaled-up accordingly for this analysis.
Ah, a "model using publicly available data from comScore and Alexa"...right, that sounds like a totally reliable blackbox to base a completely game-changing assertion.
Competition between Android, iOS and Windows Phone already seems to be spurring a race towards openness.
But as the browser and web got more powerful and we started moving around more and having more computers (work, laptop, home) the centrality of "the cloud" because a selling point and for many applications the "local power" of the desktop was less important.
Now however we're seeing new constraints play a role.
On mobile, again power in some cases is a bit limited, we have more data/sensors to play with and HTML hasn't caught up to complicated swipe interfaces too well yet. We have location and accelerometer sensors that haven't been too well integrated yet. And your phone is always with you. So we're seeing a swing to cloud data back local apps again. The 3 mobile apps I use the most are google reader (the new site layout is space wasteful in a browser on my netbook, good luck having fun reading on a mobile browser), a movie player and a music player. On my desktop I'm just as likely to use grooveshark or some other streaming service, but with still expensive mobile bandwidth having "locally cached" mp3s wins out over streaming.
If we want to move this back all onto the web, we need an html6 that supports swipe, mobile app like layout, local storage (I know, html5 draft kinda has this... ish)
It's all about device constraints and what is available.
This is just a signal we need to develop the browser some more if we again want it to be the platform of choice in this new space.
History tells us that this "balance" is a mirage. It's a see-saw with a foundation based on thick vs. thin clients, i.e. where the horsepower is located.
It really is all very temporary anyways. Native apps took off because the web wasn't ready for mobile (most sites were horrendous, etc). I am finding more and more sites offering up excellent mobile experiences that rival or exceed what their parallel native apps offer. Just a few stories away from this the fogcreek crew say "Apps provide things we can’t get out of the web: better speed, offline support, smooth animations, push notifications, and a native look and feel". Of those the only item that legitimately is an advantage of native apps is look and feel.
The internet is going now where and this won't hurt it at all.
As for the web, the subset of the internet that is about passing webpages around? Well, maybe it will be bad for it. And possibly that's not a bad thing. There are pros and cons.
I've done a lot of "desktop" aka "client/server" programming... 5 years ago everyone declared that DEAD.
Now, we're going back... it's just now there's a few more devices to support, with tinier screens + touch.
Since the iPhone, people have had a chance to kill "dead time" which would otherwise have been spent staring into walls and windows. Twitter's microblogging and -communication, of course, has exacerbated this where we see celebrities who tweet when there is downtime on the set where they are shooting.
I think it's completely ridiculous to say that this is something particular to mobile apps rather than mobile proper.
The web experience on mobile devices is improving all the time. With all major smartphone platforms now sporting WebKit browsers, CSS3 media queries and HTML5 are widely supported in the mobile space. That enables us to create sites with one markup that function well for all screen sizes.
There isn't as much pizazz on the web as can be done with apps, but given the expense of effectively writing the same app three times (one for web, one for Android, one for iOS) responsive techniques in web design can make the web option a lot more cost effective than the alternative.
To me, the possibilities with a mobile device are far greater than what lives within the browser. As much as I love the open standards of the web, and have become a software develop in their midst, there is something that really inspires the imagination about mobile devices that goes beyond the browser.
But I can't see people switching to mobile for the bulk of their serious, work-oriented computing unless the hardware experience drastically changes. I would never program on an iPhone or iPad if a computer were available. Nor would I want to do serious video editing, or CAD. I probably wouldn't even want to type a book on an iPad unless I had an external keyboard and mouse--and at that point, how different is it from a laptop?
Granted, most of these use cases can't be done on the web today. But that may (or may not) change. And even now, there are professional tasks that I'd much prefer to do on a laptop or desktop than on a mobile device--for example, updating Basecamp to-dos.
I also wonder whether this mobile vs web distinction will persist in the long run. Imagine a future where the most popular way to access Facebook is by clicking an icon on your iPad, which launches a Facebook mobile version in Safari or a very thin native app built around UIWebView. Is this mobile, or is this the web? At that point, I think the distinction becomes meaningless.
That doesn't mean you can't use a mouse and keyboard as well. And that does make it similar to the form-factor of a laptop or a desktop but that misses the core advantages of tablet computers. The OS experience, the app. install and management experience, the portability advantages (tablets are vastly more portable than even a laptop), and the potential cost savings. Not to mention that it could easily be a different set of companies providing such systems in the future than the PC stalwarts of today (perhaps google, apple, samsung, and HTC instead of Microsoft, dell, and hp).
As you said, the fundamental difference at that point between mobile and traditional computers is the OS. But if that's the case, then I have to question how relevant the distinction is from a developer's point of view.
To cater to mouse + keyboard + iOS/Android users, we don't need to build native iOS apps. We can just as well build regular old web apps and ensure they work well for mouse + keyboard + iOS/Android. That's not very different from what we do today, except for the fact that we'd have to account for yet another set of screen sizes.
Therefore, as far as serious/business/professional apps are concerned, I don't foresee a major paradigm shift. More like a pressure to learn some new web dev techniques, such as responsive design.
Again, though, I don't claim this would hold true for consumer apps. Many of those will likely be used from an iPhone or an iPad without any external devices.
I'm not sure it makes sense to think of it in terms of number of 'useful' website vs mobile apps today by some definition of 'useful' though. Even if you expect the ratio of useful / not-useful to decline at 40M Android apps vs 400K you'd still expect that most of the top useful 100 apps in 3 years' time haven't been conceived yet and will be very different from today's.
For example, the app I'm building right now has an entirely iOS-based UI, but the brains are hosted on the public Web via API. The app talks to the back-end via Websockets for anything that requires brains.
This type of architecture allows your application to participate in the UI-du-jour (native apps) while at the same time participating in the open Web. And you can simply write another client when another type of device (I dunno, Kinect?) becomes the consumer's preferred method of consuming. I think it will become much more popular in the coming years.
Some people are definitely concerned this leaves us in a less open world, but IMO, we'll be just fine. The most important part to have open is the data/logic layer, and that has happened through the explosion of APIs.
No question that apps are big. But there's a lot of data out there that people regularly used less than 10 apps in the last 30 days. Almost all of those are social or built-in apps. They used a lot more than 10 websites in the past 30 days.
The raw number of apps and websites available is a complete red herring. Mobile apps have existed (in a numerically significant form) for < 4 years. The web is closer to 20. Expect the amount of noise in the app space to follow a similar growth curve except to date app discovery is worse. With the control vendors have over app store environments, this may get even worse and not better. Or it might get a lot better. Who knows. One thing is for sure -- if you want to be found right now, you're better off with a good web site.
We discussed this a while back in a blog post based on other data from Flurry. Also discussed by Skyfire.
Mobify post: http://blog.mobify.com/2011/06/23/have-apps-really-passed-mo...
Skyfire post: http://www.businessinsider.com/why-flurry-got-it-wrong-on-ap...
Ed: fixed link.
Nevertheless, I would like to raise a few points/questions with the reasoning in this analysis, somewhat in defense of the web:
1. Mobile apps provide a different mix of functionality when compared to the web (particularly on our PCs). Quite a few mobile apps are simply games, various time no-value-add time sinks and other similar non-connected activities. Hence, I am unsure about the merit of comparing 60 minutes inside of a web browser to 60 minutes on a mobile app. (On the other hand, perhaps facebook usage more than compensates for all this)
2. One reason mobile usage is so high is that our smartphones are always with us and many of us chronically play with them when we have more than a 15 second pause. Again, I am unsure about the "quality" of the interaction during these times.
3. It seems that many mobile apps - the huge hits notwithstanding - solve small, yet meaningful problems in a cool way. This is great, but it eventually yields a lackluster impact - both financially and in terms of the effect on society - and it makes for a real tough proposition to develop mobile apps as a team (read multiple people) and expect to sustainably "grow" a business. Yes, the barren solution space is huge, but the bang for the time spent is also smaller on average. In contrast, the web now provides really large flexibility in solving a range of problems at various depths.
4. It also seems that several impressive startups are leveraging the mobile platform as an add-on to their web-based presence, another channel to solve problems on or a value enhancer to the tech. There is 0 doubt in my mind that mobile is here to grow in this sense.
All in all, applications that leverage mobile have a large empty space to fill. However, people should not just jump into mobile because it seems to be the "rage".
Fifteen years ago we were bitching about all the startups abandoning 16 bits worth of open and unfettered TCP port address space for the walled gardens of port 80 and this new-fangled HTTP protocol.
Viva la (r)evoluzione!
If you're building an app nowadays, ideally you'd have a web, mobile web, iOS and Android version at launch. Because you want distribution and need to go where your users are.
Thing is, the web is very crowded so it's hard to get attention whereas relatively, it's still wide open when it comes to the native app galleries.
But the main driver will be supply. There just are'nt enough Android, iOS developers. Web devs are legion.
And no cross-device solution will be able to keep pace with webkit & HTML/CSS.
If you are going for the consumer market and some parts of the B2B market then yes mobile will keep advancing.
But you aren't going to do the design, programming, your taxes or any other labor intensive work or creation on your mobile.
So yes we probably underestimate it, but if we believe in that, then we are most probably also underestimating how much both desktop apps and web apps are going to matter many years ahead of us.
Should I remind you that the chip in the that SIm card is in fact the same cpu chip that was used in Apple IIs?
In next ten years you will see design,programming, etc done on tablets
We could do much smaller cellphones than we do. But we don't because there is a certain size where things start to break down.
I need the space when I work. Just like I need a desk of a certain size.
The need for big screens wont go away any time soon.
For a mobile device to replace my desktop/laptop user interfaces + form factors are going to need some rapid advances.
A phone sized interface will probably always feel constricting and confining when trying to tackle anything complicated. (And a good part of mobile UX is preventing and alleviating this feeling of claustrophobia, IMO.) It's extraordinarily difficult to cram the full neuro-kinesthetic experience of a complex task into a playing card sized screen; we're more comfortable interacting with things that are "body sized".
But whatever the backend is (and remember, your users don't give a hoot where the computing is done), certainly large sized interfaces are not going to disappear. What is most likely going on is a re-centering of the bell curve: more people will have a smaller screen as their primary interface to the net/web, and larger interfaces will accommodate more complex behaviors and tasks as well as group activities.
It's true that there's a mobile shift in _startup_. Most startups want to hit the mobile goldmine and become the next [enter a company here]. However, we fail to understand that those important minutes spent are spent on either games, which are consumed in matter of days (There's not many games that sustain growth for over a period of a few months), and services that, for the most part, already had a web presence.
Here's my rationale: If your startup/internet service company wants to be successful, it will have to have a mobile presence, if it doesn't, opportunity from competitor will arise and you might get into trouble.
However, considering the cost of a native mobile app (iOS or Android), I wouldn't be surprised if within 2-3 years, startup would go back to focus on web first and if traction occurs, expanding to the mobile area (Except for startup that leverage the mobile hardware, ie camera, geo-location, etc)
Also, does it happen to be any cached/alternative sources for that content that's not hosted there so I can read the article? Normally I could use an anonymous proxy but it seems like too much of a hassle (it's not that easy to find an anonymous one that hasn't been already banned internally as well). Thank you!
Also, I can casually browse dozens of sites and completely forget about them afterwards... I don't casually install dozens of apps.
Really? No shit.
Show me that people are using their mobile apps more than they're using their desktop/laptop browsers and you'll have something to write about.
Is the argument simply that the app market is less crowded? Is a comparison of the number of websites on the internet(apples) to the number of android applications(oranges) that compelling?
Do they mean mobile apps vs. desktop Internet use? That would make more sense.
Similar to the way that different desktop browsers each roll out new feature support, and often times support features in completely different ways (CSS transparencies/gradients anyone?) mobile browsers have even slower and more varied support for features and interaction control. This problem with desktop browsers has kept most web developers and firms I work with from integrating the latest and greatest HTML5/CSS3 features in all their sites, because there is a huge handicap in trying to optimize for so many different platforms as each evolves on its own timeline.
Even though writing the same app to work well in both iOS and Android is a HUGE undertaking, the advantage is you are only writing your application to meet two different sets of API specifications. Even better yet since mobile devs are typically either iOS OR Android experts you can potentially have two different teams of devs working on the platform they know best instead of having one html team constantly striving to make your application work well (or at all) in a dozen different commonly used mobile web browsers.
I applaud Facebook for their efforts to build out a really great looking mobile web app that does look an awful like their mobile apps, but has anyone watched someone who doesn't know (or care) about the difference between a native app and mobile web app use it? I got a fairly tech-savvy family member a kindle fire recently and it comes pre-installed with a shortcut to the Facebook mobile web app because the native app is not available in the amazon marketplace. Watching over their shoulder I was amazed to see how frustrating the experience was for them when the querystring gets corrupted at certain points and all they are left with is a PHP whitescreen of death. No loading symbol, no ability to get out of the corrupted state the browser got itself into, and generally the user was left with no idea why this wasn't working as nicely as their Facebook app on iOS/Android that looks exactly the same.
At the end of the day I now lean more toward the side of mobile apps continuing to be more efficient and effective for apps that require lots of interaction with the user even if you could technically implement all those interactions in a browser. The relative simplicity of only having to code for two platforms instead of 12, and simultaneously providing your users with a more responsive experience is definitely worth it.
http://gs.statcounter.com/#mobile_vs_desktop-ww-monthly-2010...
What about adding a 3rd metric for how much time the consumers browse on their desktops and laptops? That's going to eclipse the time stated in the article.
The reason we're seeing such high app numbers in Flurry is that mobile devices are not as convenient to browse on compared to laptops and desktops, so apps like games take over the time spent on such devices.
Many of those will be in India, where RIM is actually growing like crazy. I anticipate Windows on Nokia will be a serious player, and Android phones will drop to prices the masses can afford.
The race is not over and the "mobile web" may yet turn out to be the dominant platform.