Categories
APIs Languages

Websites vs Web apps: What the experts think

websites-vs-webapps

Definitions of web sites vs. apps

Web sites are so deeply embedded into our daily culture that it is impossible to imagine life without them. Even as a developer, I find it hard to remember the times from my childhood when my chubby little hands didn’t yet know how to type. In the last two decades, the Internet has grown, expanded, exploded and became impossible to ignore, making any keyboard without an Internet connection pretty much useless.

In the last few years, the web brought with it a new term that can be exciting and confusing at the same time: “web app”. But what is a “web app”, how does it differentiate from a “web site” and why does it matter?

Understanding this difference ultimately makes us better users or developers? Is a business going to blossom just by marketing its online presence as a “web app” instead of a “web site”?

To figure out the boundaries between websites and web apps, I interviewed several prominent figures in the web technology domain who contributed with their experience and professionalism to help guide the debate: Dominique Hazael-Massieux (Mobile Web Initiative Activity Lead at World Wide Web Consortium), James Pearce (Head of Developer Advocacy at Facebook), Michael Mullany (CEO at Sencha), Christian Heilmann (Principal Developer Evangelist – HTML5/Open Web – at Mozilla Corporation) and Stephen Pinches (Head of Learning Technologies – ELT at Pearson plc and Group Product Manager – Mobile & Emerging Platforms at Financial Times). In this article I pieced together their expert input to help answer the web site vs web app debate.

The difference between Web sites and Web apps

In the pre app store era, the word “applications” had been applied to Web sites that provided advanced user interactions and capabilities previously available only through installable software. Early examples of web applications include Webmail, Google Maps and Google Docs. Compared to the classic web, i.e. blogs and news sites, web apps provided a richer user experience and access to advanced browser capabilities.

Today single-page web sites might still be referred to as web apps, but it’s more about the task focus than the technology itself. From this perspective, as Christian Heilmann explains, “The use case of an application is always to DO something with it”.

The task centricity of web apps is easier to understand if you think of smartphones or tablets: an app’s purpose is to achieve a specific task, like making a call, checking your email or finding a taxi nearby.

Some may argue that we can simply classify Web sites as being read-only and Web apps as being read-write. That certainly seems simple enough: Web sites are for consumption what Web apps are for creation. Does it sound right?

For developers, it is easier to draw the line between web sites and web apps if we think of the technical distinctions. Web apps have some defining attributes that bring them closer to their native counterparts:

  • self-contained
  • rich/interactive user interface, possibly mimicking the native UI of the device,
  • using advanced device capabilities – like geolocation, camera integration, or other technologies that the W3C Device APIs and Policy Working Group is developing,
  • action oriented rather than information oriented
  • not relying heavily on (or hiding when possible) the browser chrome (back button, reload button, address bar),
  • working off-line, for example using HTML5 ApplicationCache, localStorage, or indexed database.

Mozilla’s Christian Heilmann argues that the offline attribute is not a technical necessity in terms of definition, but rather a crucial usability distinction:

“Seeing how flaky our connections are – I am writing this on a plane – our apps should make people as effective as possible and this means we shouldn’t be dependent on a connection. The interface should be usable whilst we are off the grid and sync as soon as we go online”.

But how can we explain the difference to non-technical users? And, do we need to?

According to Dominique Hazael-Massieux, a Web site can be presented as a Web app as long as users consume it in a similar way they do a native app. If it’s exposed as an iconified app and used for a specific task, it shouldn’t matter whether it’s contained in the browser or installed via an app store. Facebook’s James Pearce outlined a few possible vectors that need to be considered when differentiating between Web sites and Web apps. I‘ve summed up his arguments:

Creation versus Consumption

Pearce asserts that read-only interaction should be classified as a site, but this criteria is not sufficient to distinguish between web sites and web apps. We still have cases like Flipboard (clearly oriented towards consumption) or Twitter and Facebook (with entirely user-generated content) that do not fit in any box.

Linkability

Since both web sites and web apps can be launched by entering a URL into a browser or from a home-screen icon, this is clearly “not a reliable way to distinguish between web apps and web sites” according to Pearce.

User Experience

Visual pizzazz is an important argument, one that users might particularly relate to, but is also a fuzzy boundary. What if my site displays a fixed toolbar, but no back button? What if my list appears as hyperlinks instead of ‘tappable’ items? What if I use plain scrolling instead of smooth fancy bars?

Architecture

In the case of single page webapps, is SEO the price to pay when choosing to give the browser far more autonomy and responsibility and take advantage of its HTML5 APIs like storage? Do Web sites have SEO capabilities while Web apps don’t? We are back to explaining the differences between the two by using technical terms.

Should you be building web apps or web sites?

This question might be regarded as a technicality with a pinch of marketing to spice it up. This reminds me of the “HTML5 is ready” contest by Sencha that was announced a few months back, encouraging developers to draw inspiration from native apps and create similar web apps that show off the capabilities of HTML5.

The creators of the competition correctly argued that “the mobile web is the most fertile ground for leading edge web development because it doesn’t have the legacy of the older internet explorers that the desktop does. You can start your development with the assumption that your app or your content will be used in a fairly recent browser, so you can take advantage of a whole host of features like Canvas, inline SVG, HTML5 video, CSS3 styling etc. that bring the experience alive for the user”, as Sencha’s Michael Mullany explains.

Would it be safe to argue in favour of building web apps instead of web sites especially on mobile? Mobile users perform specific tasks on their devices, so a web app that offers the same experience as a specialised native app might gain more interest compared to a regular website.

Long term the distinction should not matter. According to FT’s Stephen Pinches, it really doesn’t make any sense, on the long term, to speak about the future of the mobile web: “there shouldn’t be “mobile” and “desktop” but simply good, user-centered design, which adapts and responds to the screen size and features of the device upon which it is displayed. However, on short to medium term, there is a need to differentiate and ensure the user experience is as good as possible on a given device.”

The ‘app-ification’ of the Web

Whatever your preference may be, there is an increasing number of mobile developers targeting web apps. Based on VisionMobile’s latest Developer Economics survey of 6,000+ developers, already 23% of HTML5 mobile developers develop web apps, compared to 38% who develop mobile websites.

With browsers increasing support for device APIs, and with a growing number of developers going direct to native with PhoneGap, Icenium or Appcelerator, or even with the recently launched Firefox OS, the web world is clearly moving in the direction of apps.

As Sir Tim Berners-Lee said in 2012, “the solution is in your hands: develop web apps!”

Categories
Languages

Is HTML5 the 3rd horse in the race?

Biggest developer survey

We’re thrilled to announce that the Q2 Developer Economics survey we conducted throughout April was the most successful to date, zooming past the 6,000 respondents mark, making it the biggest developer survey globally.

UPDATE: The survey results have now been published – download the free report here.

We broke through the 6,000 developer mark mainly thanks to the help of our 48 Marketing and Regional partners. Together we reached developers from an unprecedented 115 countries, from mature markets, like the US and Western Europe, to emerging markets, like Brazil, Russia, India and China. To reach developers on a global scale, we translated the survey in 10 languages (Arabic, Chinese, French, German, Japanese, Korean, Portuguese, Russian, Spanish, and Swedish), aided by our local partners, who helped us reach the local dev communities. Thanks to a partnership with Mobile Monday, we also promoted through over 20 local MoMo chapters in Asia and Oceania.

In the next two months we ‘ll be diving into the results of the survey. The Developer Economics state of the developer nation report will be launched in July, as a free download thanks to the sponsorship by BlackBerry, Mozilla, Intel and Telefonica. This 5th incarnation of the Developer Economics report will feature the latest market trends, including Developer Mindshare and Intentshare, platform selection criteria, revenue models, revenues per app and many more. To whet your appetite until the July launch, you can read the previous editions of the Developer Economics report.

To be the first get the Developer Economics 5th Edition report, sign up for our mailing list!

Sneak peek on Mindshare: Android, iOS duopoly entrenched – with HTML closely behind

Our early results from the Q2 app developer survey are starting to come in – starting with the Developer Mindshare Index 2Q13, i.e. the percentage of mobile developers using each app platform.

As you can see in the graph, the use of Android and iOS is still predominant, with a few percentage points of change for both platforms when compared to our 4Q12 survey. You’ll also notice the continued growth of HMTL as the third horse in the platform race, slowly creeping up on iOS. These trends have been steady over the past year – but what do they mean?

Developer Mindshare Index 2Q13

The continued positioning of Android and iOS as the top two platforms is a no-brainer: Android has the largest installed base and iOS enjoys the highest revenue potential overall – so why does HTML5 continue to grow?

HTML5 grows in popularity as large groups of web developers are leaping over the ever-shrinking chasm from desktop to mobile apps. Moreover, HTML5 allows for the development and deployment of apps that work across different platforms, usually at a lower cost of developing HTML apps, and for most app categories. About two thirds of developers targeting HTML mobile develop web sites or web apps while just under a third are using PhoneGap. Stay tuned for more analysis on the route to market for HTML5 apps in the full report.

HTML5 has wide industry backing across telcos, handset makers and platforms (Firefox OS, BlackBerry WebWorks and Tizen) going for it. At the same time, there are certain key disadvantages, namely access to native platform APIs, as well as the lack of a unified development environment and quality debugging tools.

HTML5 is now challenging the duopoly as a development or deployment platform – with the route to market varying across browsers, hybrid apps (e.g. PhoneGap), JavaScript converters (Appcelerator) and dedicated platform frameworks (BlackBerry WebWorks). We still see a growing diversity in the go-to-market approaches for HTML5 developers, and one which we believe will continue to expand. We‘ll be analyzing the HTML vs. native tradeoffs in a future report, but in the meantime – what’s your take on the HTML vs. native debate?

Sneak peek: Windows 8 and BB 10 are gaining traction

As you can see from the early Developer Mindshare graph, Windows 8 and BlackBerry 10 have already attracted a reasonable amount of developer attention. What’s important here is that BlackBerry developers have been quick to migrate from the old legacy (5,6,7) platforms and adopt the latest, BB10 platform.

BlackBerry Mindshare

What’s interesting to note in the graph above, comparing the use of BB platforms between the two latest surveys (4Q12 vs. 2Q13) is the fact that the BB 5,6,7 platforms are quickly fading into oblivion, with BB10 mushrooming to a substantial 15% mindshare in just 6 months. The mindshare of BB10 is slightly less than that of BB 5,6,7 six months ago, but the platform is still gaining in strength, as our data for the platforms that developers plan to adopt seem to suggest, so there’s room for growth. The extent to which the new BlackBerry platform can grow in Developer Mindshare depends primarily on the volume of devices that BlackBerry will manage to sell in the coming months, given that reach is the primary reason for platform selection.

Competing against Windows Phone and BlackBerry 10, new entrants Firefox OS and Tizen are slowly gaining support from a few handset OEMs and network operators. Another open question is whether the HTML5 platform proponents – Tizen, Firefox OS and BlackBerry WebWorks – should band together towards a single HTML5 implementation or keep pursuing independent and conflicting strategies. What’s your take?

Full report available in July

We’ll be stopping our sneak peek here – stay tuned for the full report for more (out in July)! There, you’ll find an in-depth analysis of major trends, such as the shifting balance of power between the top platforms, devices vs. tablets, revenue models, as well as the main factors affecting app monetization. If you haven’t already done so, subscribe to our mailing list to receive word of the report publication.

Until next time,
– Matos (@visionmobile)

Categories
Languages

HTML5 Adoption and the Importance of Independence

Last year was quite a bad one for HTML5 in terms of developer mindshare. At the end of 2011, developer sentiment seemed to favour a shift away from native and towards HTML5 for a large range of application categories. As the year went on, there were more horror stories than successes and the tide of opinion swept the other way as Facebook publicly declared that HTML5 wasn’t good enough for their mobile apps. With a title declaring the importance of independence you’d be forgiven for thinking this article would be about a need to reverse that trend to get away from the tyranny of walled garden app stores. Nothing of the sort we promise.

Look out for sampling bias

Independent surveys and statistics are the important thing referenced in the title. HTML5 adoption just happens to be the subject of one of the best bad stats examples in the midst of last year’s shift. Apparently last summer, just after Mark Zuckerberg’s revelation that betting too heavily on HTML5 for mobile apps was the biggest mistake he’d made at Facebook, 94% of app developers were betting on HTML5 winning. Of course this survey came from Kendo UI – a vendor of HTML5-based tools for mobile app development. It’s unlikely they set out to create a useless survey but they did want some data to support their tools. So they asked web developers if they were using, or planning to use web technologies – amazingly most of them said yes! This is clear from the fact that the number one reason for using web technologies in the survey was “familiarity of languages”. Such a high proportion of developers working with web technologies should be excellent news for upcoming web-app only platforms like Firefox OS and Tizen, however, the huge number of native applications being created across all the platforms suggests the real figures are nothing like this. It’s a clear case of sampling bias. Kendo UI recently published another survey in which a more realistic 50% of developers built some apps with HTML5 last year but a rather less credible 90% were planning to use it in 2013. Contrast this with our latest mindshare and intentshare data, which agrees with 50% use last year but sees only 15% of those not already using HTML planning to adopt it.

Seek transparency and independence

In our developer economics survey we make a big effort to ensure we collect data from a wide range of developers and we publish the breakdown of platforms developers are working with along with the results of all the other questions. Where appropriate we weight or normalize results according the the proportion of different groups in the survey. Of course it’s not possible to avoid all bias in the sample and there is undoubtedly an element of self-selection – developers with an interest in the commercial aspects of app development are much more likely to answer a survey entitled “developer economics”. To a certain extent, that’s deliberate – serious developers who are trying to build businesses that involve apps want to know what other like-minded folks are thinking and doing.

If you’re looking for reliable information on the app market, particularly if you want to make business decisions using it, you need the most independent and transparent sources. We’re doing our best to be one of those sources.

Categories
Languages

HTML5 vs Native – What are the tradeoffs?

In our latest developer survey we asked developers who use or plan to adopt HTML5 why they do so and also what the technology needs to compete with native alternatives. The results show a tradeoff of increased portability and lower development cost against capability, in the form of reduced API access and a poorer development environment. In this scenario, the key to success with web technologies is taking advantage of their strengths in areas where their weaknesses are less of a handicap.

Developer Economics 2013 - HTML5 trades off native optimisation for portability and cost

HTML5 is becoming a viable alternative to native development across a number of app categories. We found that HTML developers mainly focus on specific app categories such as Business & Productivity (42% of HTML developers), Enterprise (32%) and Media apps (28%). On the other hand, Games are not a common category among HTML developers (12%).

We asked developers that use or are planning to use HTML about the reasons for platform selection. The majority indicated code portability as the main incentive for using HTML5. Low cost development is the second driving force for HTML5 adoption, highlighted by 51% of developers. HTML is still an “extension platform” in that only 26% of developers who use it consider it their main platform. We asked developers that use, have used or are planning to use HTML what they think HTML5 needs to compete with native platforms. Access to native APIs is a top challenge with 35% of developers indicating this as a critical success factor. HTML5 will always be a step behind in support for native APIs, given that cross platform tools and browser vendors will always have to implement support for a new API after it is released to developers by the platform vendor. In addition, the HTML5 development experience is subpar, with developers indicating that a better development environment (34%) and better debugging support (22%) are needed. More importantly, optimised HTML5 devices were not seen as important as the native API access or dev environment. This leads us to conclude that HTML proponents such as Facebook, Mozilla and Google should focus on cross platform tools and development environments on at least equal levels as they focus on full platform efforts like Facebook Platform, Firefox OS and Chrome OS.

[doritos_report location=’DE13 Article – HTML5 vs Native’]