April 23, 2012
I've been an eBay user since 1999, and I still frequent eBay as both buyer and seller. In that time, eBay has transformed from a place where geeks sell broken laser pointers to each other, into a global marketplace where businesses sell anything and everything to customers. If you're looking for strange or obscure items, things almost nobody sells new any more, or grey market items for cheap, eBay is still not a bad place to look.
At least for me, eBay still basically works, after all these years. But one thing hasn't changed: the eBay website has always been difficult to use and navigate. They've updated the website recently to remove some of the more egregious cruft, but it's still way too complicated. I guess I had kind of accepted old, complex websites as the status quo, because I didn't realize how bad it had gotten until I compared the experience on the eBay website with the experience of the eBay apps for mobile and tablet.
eBay Mobile App
eBay Tablet App
Unless you're some kind of super advanced eBay user, you should probably avoid the website. The tablet and mobile eBay apps are just plain simpler, easier, and faster to use than the eBay website itself. I know this intuitively from using eBay on my devices and computers, but there's also usability studies with data to prove it, too. To be fair, eBay is struggling under the massive accumulated design debt of a website originally conceived in the late 90s, whereas their mobile and tablet app experiences are recent inventions. It's not so much that the eBay apps are great, but that the eBay website is so very, very bad.
The implied lesson here is to embrace constraints. Having a limited, fixed palette of UI controls and screen space is a strength. A strength we used to have in early Mac and Windows apps, but seem to have lost somewhere along the way as applications got more powerful and complicated. And it's endemic on the web as well, where the eBay website has been slowly accreting more and more functionality since 1999. The nearly unlimited freedom that you get in a modern web browser to build whatever UI you can dream up, and assume as large or as small a page as you like, often ends up being harmful to users. It certainly is in the case of eBay.
If you're starting from scratch, you should always design the UI first, but now that we have such great mobile and tablet device options, consider designing your site for the devices that have the strictest constraints first, too. This is the thinking that led to mobile first design strategy. It helps you stay focused on a simple and uncluttered UI that you can scale up to bigger and beefier devices. Maybe eBay is just going in the wrong direction here; design simple things that scale up; not complicated things you need to scale down.
Above all else, simplify! But why stop there? If building the mobile and tablet apps first for a web property produces a better user experience – why do we need the website, again? Do great tablet and phone applications make websites obsolete?
Why are apps better than websites?
- They can be faster.
- They use simple, native UI controls.
Rather than imagineering whatever UI designers and programmers can dream up, why not pick from a well understood palette of built-in UI controls on that tablet or phone, all built for optimal utility and affordance on that particular device?
- They make better use of screen space.
Because designers have to fit just the important things on 4 inch diagonal mobile screens, or 10 inch diagonal tablet screens, they're less likely to fill the display up with a bunch of irrelevant noise or design flourishes (or, uh, advertisements). Just the important stuff, thanks!
- They work better on the go and even offline.
In a mobile world, you can't assume that the user has a super fast, totally reliable Internet connection. So you learn to design apps that download just the data they need at the time they need to display it, and have sane strategies for loading partial content and images as they arrive. That's assuming they arrive at all. You probably also build in some sort of offline mode, too, when you're on the go and you don't have connectivity.
Why are websites better than apps?
- They work on any device with a browser.
Websites are as close to universal as we may ever get in the world of software. Provided you have a HTML5 compliant browser, you can run an entire universe of "apps" on your device from day zero, just by visiting a link, exactly the same way everyone has on the Internet since 1995. You don't have to hope and pray a development community emerges and is willing to build whatever app your users need.
- They don't have to be installed.
Applications, unlike websites, can't be visited. They aren't indexed by Google. Nor do applications magically appear on your device; they must be explicitly installed. Even if installation is a one-click affair, your users will have to discover the app before they can even begin to install it. And once installed, they'll have to manage all those applications like so many Pokemon.
- They don't have to be updated.
Websites are always on the infinite version. But once you have an application installed on your device, how do you update it to add features or fix bugs? How do users even know if your app is out of date or needs updating? And why should they need to care in the first place?
- They offer a common experience.
If your app and the website behave radically differently, you're forcing users to learn two different interfaces. How many different devices and apps do you plan to build, and how consistent will they be? You now have a community divided among many different experiences, fragmenting your user base. But with a website that has a decent mobile experience baked in, you can deliver a consistent, and hopefully consistently great, experience across all devices to all your users.
I don't think there's a clear winner, only pros and cons. But apps will always need websites, if for nothing else other than a source of data, as a mothership to phone home to, and a place to host the application downloads for various devices.
And if you're obliged to build a website, why not build it out so it offers a reasonable experience on a mobile or tablet web browser, too? I have nothing against a premium experience optimized to a particular device, but shouldn't all your users have a premium experience? eBay's problem here isn't mobile or tablets per se, but that they've let their core web experience atrophy so badly. I understand that there's a lot of inertia around legacy eBay tools and long time users, so it's easy for me to propose radical changes to the website here on the outside. Maybe the only way eBay can redesign at all is on new platforms.
Will mobile and tablet apps kill websites? A few, certainly. But only the websites stupid enough to let them.
[advertisement] What's your next career move? Stack Overflow Careers has the best job listings from great companies, whether you're looking for opportunities at a startup or Fortune 500. You can search our job listings or create a profile and let employers find you.
Posted by Jeff Atwood
Apps also have platform restrictions. For example many sites will not make a iOS app because they have to give 30% of their sales to Apple.
Platforms that like to skim off the top make web sites seem more inviting to many companies.
Completely agree with your closing statement. Mobile-first web dev/design practices are becoming more popular, well explored and well documented/explained these days. Lets hope this can deal with your "Apps are better" list items #2 and #3. #4 is being dealt with (somewhat inadequately) by HTML5 storage solutions. #1 is... well, that's an arms race, and apps will likely always be ahead here being closer to bare metal BUT we may get to the point where its not very important.
And now we can go back to worrying whether robots will kill humans.
Will Apps Kill Websites?
Will Apps Kill Webapps?
Maybe, and that's a good thing.
You can use apps as long as Apple deems you fit. Ebay is a good example, I am surprised they allow it. Why don't they say ebay is circumventing their Apple payment system by allowing people to login to ebay and buy stuff with Paypal?
I think that the web will improve thanks to the apps, I've seen it at work, where just the process of creating the mobile site has brought a full re-evaluation of the desktop site.
You're right, but I think there is one more important point - especially for examples like eBay: the app was built recently, and without any legacy cruft.
I think one factor that is missed is "A good app might kill a bad website", and a "bad app won't kill a good website".
I don't know why someone would spend $100k to develop an app but not $20k to fix a web page, but it happens. Maybe people expect web pages to be stupid, cluttered, slow, and hard to use.
If a coding horror or stackoverflow app would kill the respective sites, something would be really wrong.
The Ebay website is far from unique in the design cruft that has accumulated on it. You could argue that the problem isn't so much with the web designers, or even the product people trying to shift their particular block onto the real estate, but the twin problem of a  desktop-sized display with a  pixel-accurate mouse interface. The Ebay app looks beautifully streamlined on the iPhone and the iPad, and the common factor here is the display size and the touch interface. Basically, the twin pressures of a small display coupled with a chunky touch interface (chunky in sensitivity compared to a mouse) force mobile designers to design in large, discrete blocks, which also happen to help streamline data rendering.
The proof of this for me is that mobile sites and apps often feel kind of chunky and under-designed when mapped onto a desktop monitor. I suspect it's not just the display, but the fact that the mouse interface encourages pixel-sized interaction, whereas the touch interfaces are completely in the other direction. In other words, where do you 'click' on a mobile site? This could be ameliorated by changing the nature of mouse from 'clicking' to 'touching' -- i.e., deliberately 'degrading' the sensitivity of the mouse, from a desktop perspective. I suspect the main culprit here is the mouse interface itself, and not the screen size, resolution, location, etc.
Perhaps Ebay-style websites could be redesigned along more effective principles if the nature of mouse/screen interaction was rethought?
Who wants to spend time and money to develop JS apps for all different browser/OS combinations?
It's infinitely easier and cheaper to code native apps today.
Solution: give us Dart or similar "real" language, and stop supporting obsolete browsers.
Until then, we are condemned to live in the App world.
Focus is why apps are better.
You have less real-estate so you have to focus on delivering what matters.
Another PRO for websites: They are findable in search engines. Mobile apps, not so much. There are ways around this like cross-linking between web and mobile app but that requires you to have a web app too.
Adam Nash (Greylock Partners) covers some of this in his mobile user acquisition series (detailed article links further down the page): http://blog.adamnash.com/2012/03/28/product-leaders-user-acquisition-series/
Of course, there is one simple way to encourage app-centric design on a website: take away the notion of page scrolling. The browser is pretty much the only common WIMP application that both allows and encourages scrolling of the main canvas. If you designed a website with a one-page canvas, you would be more than halfway to presenting it like an application. But in the days before AJAX and high bandwidth the concept of a 'one-page website' (in other words, 'an application') was untenable.
I very much hope that sites and [in-browser] apps will merge eventually. An 'app' will be just the 'site' adjusted to the features available on your device, with niceties which a local app can afford: precached data, integration with other parts of the platform, etc.
Of course, 'site' will always be there, as long as the 'browser' concept is around.
"Will Apps Kill Websites?"
No. Betteridge's Law of Headlines.
NO WAY. Otherwise need to have millions of apps on my iPad/iPhone.
Always we have to use google to search for website not apps.
It is much easier to remember or search and use domain names.
Only apps for Top 50 Websites as amazon may be used.
Apps should not be for websites, for other things , as calculator, games, etc.
It would also be nice if trees were made of cotton candy and rain was delicious lemonade.
Your points are valid, however I think there is room for both. Sure as the mobile devices become more prevalent the focus will switch to Apps, but I can't help but feel that sites will always make a good companion for Apps.
I'm not sure I agree with the idea the ebay website is complicated to use. I'm not an ebay regular and yet I don't find it that difficult to use. I just find it ugly. Not hard.
In any case, the whole premise of the article is sketchy. It's a given that a simplified version of a website will be... simpler. What's to tell?
How we get from there to "will apps kill websites"? The tablet and smartphone market isn't even comparable to the computer desktop market. It's just a fraction of it.
"But only the websites stupid enough to let them"
Most important part of this whole question.
With LastPass, I can give each site a very strong, unique password that I never have to remember or transcribe.
I have yet to see an app, whether mobile or desktop, that plays nice with LastPass.
wow wait a minute. As a developper I have the choice to build a web app to replace an old website and ship as a native app on multiple platform( www.phonegap.com ).
What you point out in your article is just a cleaner interface in the newest app. They will soon realize to redesign their website.
I guess I'm in the minority preferring the website.
I look at it and see a large number of features that I would and do use, which aren't present in the app.
People talk about the 80/20 rule and how everyone only uses 20% of the features. The problem is that for each person it tends to be a different 20% and cutting it down to the lowest common denominator will get a lot of people miffed. I have seen too many apps of websites I use made "simple"(dumbed down) and lose all the functionality I like in them that allows me to save clicks and time. I need to make multiple clicks to get to whatever function I want, and having to deal with a tiny screen and buttons that when next to each other are easily mispressed, the whole app experience is consistently poor for me, and I am constantly impressed by the patience of people that suffer their way through playing around on their phone when they have a pc terminal right next to them.
But now you have 3 major and 2 minor browsers. And mobile browsers. You have to support different versions of those. So you choose jQuery or some other library to keep things somehow universal, but it's another layer you have to worry about. And if you are writing software for corporations, someday they will say "Browser X is our weapon of choice. Don't spend extra dollars to support some hippie browser, we pay, and we demand your stuff to work with MSIE 8. And don't even start with chromes' continuous delivery. It's forbidden. Forbidden."
And let's face it, hypertext wasn't really meant for applications. The whole industry of webapps was based on some clever hacks and universal web browser. One browser to rule them all so to say. Ask your UX developer, it's no longer the case.
TechCrunch wrote 4 days ago about Facebook struggling with Apple and Google about future of html5 mobile web browser. http://goo.gl/jSe4d Quotation: "Apple and Google have a vested interest in seeing HTML5 lag behind native apps. " (but read the whole thing). So you won't have a web application that "fells" almost like native one anytime soon.
A web application that "fells" almost like native one. So, why the hell stay with the "almost like" and accept the imitation, where you can have the real thing?
Current native apps will see your camera, gps, gyroscope, and what have you. Your UX people will be happy doing less hacks and more real work. Your IT department will alone decide when they will publish a new version. They don't have to catch with the development of Chrome or Firefox.
Webapps will prevail thou. As simple version of much cleaner, better experience App can give. When you'll start a new relationship with some service, you'll probably will start with a browser. That's the first and second date. And if you are happy with it, and maybe want to move on to the next base (pardon me), you'll install the app. That's the sign of confidence you put in it. On the other side, full blown webapps with anything and kitchen sink will be like "we are good friends and that's why we live 30 years together, but we don't trust each other so much to have one fridge.".
What does it mean for the developers? APIs. With a good api you can power a web application with all its quirks, or you can build a set of applications for many different devices. Like netflix. Or dropbox.
Webapps (user experience) aren't sexy anymore. I see more and more developers going full time to the server stuff (scalability, clouds, apis), or programming lots of lots cheap native apps. With a little luck they can create a photo app and sell it for 1 gigadollar to Marc Z ;-) . This leaves the web front ends for someone else to do. This gap can't be easy filled.
And one personal observation at the end: with native apps i fell much less procrastinating than with the browser. Browser is one million pages waiting to be opened. Native app is one task to be done.
To be honest, I'm a bit disappointed that this generally high quality blog still stoops down to childish, sensational stuff like this. No, apps won't kill websites. Just like websites didn't kill applications. And consoles didn't kill PC gaming. And handheld devices didn't kill the desktop PC. And Linux didn't kill Windows.
Making unsubstantiated claims like that is a short sighted way to get some free extra traffic. It undermines any sense of integrity and expertise the readers place in a blog like this.
@Juhani I much prefer the website version as well, and don't find the interface especially ugly either. There is much more screen real estate and the website makes good use of it, for example by having filter settings down the sidebar instead of hiding them behind a "Refine" or "Advanced" button.
The eBay mobile apps really fall down when you go to view the listing details. Depending on the lister and their auction management software, some of them rely on Flash, and this is invisible on iDevices and doesn't seem to work very well on Android (at least in my browser it thinks I don't have Flash installed when I do). Others have a lot of custom HTML formatting which is either stripped out or rendered poorly in the mobile app. This is not the apps' fault, but it makes it impractical to use it to actually buy things on eBay.
"It's not so much that the eBay apps are great, but that the eBay website is so very, very bad."
Actually, if you are smart enough to use eBay properly, you're smart enough to realize the mobile and app versions of the site are borderline useless for discovering deals. You can't quickly and easily sort buy Newest/Buy it Now or Ending Soon/All Prices or sort by Free Shipping or Items Near Me.....instead, you're stuck viewing the "Recommended Listings" (who knows how eBay comes up with those). If you're gonna just bid on/buy what eBay recommends, you might as well go to Amazon and skip all the auction hassle.
The problem w/ non-full website experiences is that most companies cut out features they don't think are important. As a result, outside of lazy window browsing or perhaps logging into your account to see messages, they are virtually useless.
Another fantastic article and an interesting read yes the apps are far easier to use especially for selling but recently I have been using the site to sell items as I can set correct postage values.
I find that with the apps it doesn't let you set enough for postage and therefore you get stitched up with that for isntance a big text book on the app I think the maximum they will let you put for postage is for example £2 for the category yet postage for such an item is nearer £4
The eBay application lacks the most important feature I need when I'm looking for something on eBay: tabs. I usually search for an item, then I middle-click each item that seems interesting, then I switch to the tabs that have been opened in background to analyze them.
Without a tabbed interface, you have to show and item, then go back, show the next item, go back, and so on. You can't easily and quickly switch between two items to compare them.
This is a common issue with almost all the smartphone/tablet application compared with the browser-based version.
"The implied lesson here is to embrace constraints."
Oy. I pretty much stopped reading here. This is what people said during the Great Transition from client-side applications (or client-server applications) to the web.
*I* think the lesson is: embrace self discipline and good software engineering practices, in spite of the sexy new hotness of whatever latest technology or paradigm that comes along that's going to solve all your problems without the need for tedious thought.
I would add that in addition to the "Lord of the Flies" design choices that must be made when creating a mobile app that these experiences are experiences. Not only are you simply navigating through your options, but you are interacting with them. I would argue that the experience of swiping, zooming, and panning allow the user to get up-close and personal with their interactions.
That having been said - and especially in the case of eBay - it's really nice to have an auction (or search) open in one window while researching the widget of the day in another.
But apps will always need websites, if for nothing else other than a source of data, as a mothership to phone home to, and a place to host the application downloads for various devices.
Nope. All they need is an API they can talk to, which need not use HTML at all, or be visible in a browser. (And downloads? That's what a store is for.)
Now, I agree that they're not going to remove the need for websites for other reasons - but they don't depend on them, strictly.
I don't know - I think the line between apps and websites are getting pretty blurry.
I think a lot of this thinking is somewhat limited. CSS has had the ability since forever to define different representation models for different media - getting different representations for print and screen, for instance, without changing the underlying HTML.
So why not have more support for listing Android under media? Or iPhone or iPad? Something like
With appropriate API guidelines published, establishing different tags as different common UI elements that can have their visibility hidden or otherwise changed for other platforms, I see no reason why someone couldn't literally download an HTML5 app for any device. Users can be certain that the app will run regardless of what platform they run, and the enhancements will just work if you have an "officially supported" platform.
No censure in webapps.
(not everything is about code or specs)
Just in case you valore it.
"[Apps] can be faster." I agree with the premise, but it's interesting that a lot of the time, apps happen to be slower than the websites. Case in point: The Facebook app takes forever to... well, do just anything at all. On a lot of devices, it's so slow that it's downright puzzling. I know and use a lot of apps that make me wonder, on a regular basis: "Just what are you doing there all this time?"
Jörg Stöver, I have to agree with you. This used to be a well written and though provoking blog but it has turned into drivel. Tablets killing PCs because of a high res screen? Web apps killing websites? Pfffft!. Removing this from my RSS reader
Apps are really annoying imho. They sometimes come in handy, but they don't pack all the features a regular site has.
Normally, I even have trouble trying to figure out how they work.
I will add a touch by telling you thank you for this blog both entertaining and intelligent. It is a pleasure to come and browse.
Voyance gratuite immediate
Of course, then there's the middle ground, with sites like Fuelly that have an AMAZING mobile site (at least, for phones, never visited their site on a tablet) which is basic and easy, and a solid desktop site with more information, etc (and an easy way to switch between the two from your mobile device)
Nice article. I think both will continue to exist just as websites and desktop apps co-exist. I think in time as mobile devices continue to get more powerful the argument for dedicated apps will get tougher. Building a website that works every where seems like a much better idea than building a website and several different versions of an app!
Actually, with html5 related technologies it is possible to create "offline" websites, even though there are still some issues (for instance, in iOS we can't cache sound files yet). Thus, "They work better on the go and even offline" isn't really a app pro. On the other hand, apps have access to a broader set of device-specific features (for instance, tilt sensors), so that's another pro.
Everything old is new again. What was once a centralized mainframe with dumb terminal front end, evolved to the client server model, evolved to web server and browser, moves again to a client application. The more things change, the more they stay the same.