Mobile HTML 5 vs. Native Development: An Alternative View on What Moves the Needle -- Part 2

estrapadetubacityMobile - Wireless

Dec 10, 2013 (3 years and 8 months ago)

80 views

Mobile HTML 5 vs. Native Development: An Alternative View on What
Moves the Needle
--

Part 2

By John Jackson, Research Vice President, Mobile & Connected Platforms, IDC

In May

this year
, Apple and Google announced that they had surpassed the 50 and 48
billion cumulative app download threshold

for their iPhone and Android smartphone
platforms,

respectively. Some might argue that
what I c
haracterized as "interdependent

system
s"

in my previous blog,
work quite well, thank you very much. Indeed
,

I think tha
t
for many classes of app, developer, distribution model, and use case, they'd be right.

But we're still writing to proprietary environments


sometimes a few times per app. The
often costly and challenging issue of optimizing for various builds and code
releases is a
well understood developer pain
-
point. We at IDC are also hearing anecdotal
suggestions of a shortage of native iOS and Android developer talent.

This is ironic for two reasons. First, there is an order of magnitude greater population of
Web l
anguage developers on the planet than there
is

with either native iOS or Android
proficiency. This vast pool of Web language talent is not yet meaningfully available to
mobile app development. This will change as solutions mature and become more
broadly ad
opted.

Second, much of this talent "lives" within corporate IT departments or lines of business.
Enterprises will continue to have a voracious
,

but complex
,

appetite for applications.
With the right tooling, the same enterprises that are currently scrambli
ng to secure iOS
and Android talent might address a substantial portion of their application development
needs with talent already on the payroll.

But the thing that strikes me the most is the extent to which app distribution economics
have
not

resulted i
n a more significant investment in and utilization of HTML 5. In June
of 2011
,

the
Financial Times

(FT) now famously introduced
its

app for iOS using "pure"
HTML 5. The FT was clear about its motivations: The cost and process of porting (not
just to iOS) h
ad become onerous
,

and
the company

had little interest in sharing 30
%

of
its subscription revenue with
Apple
.

Sounds logical, right? Surely there is a small universe of premium content owners and
distribution outlets out there that are similarly motivated.

Yet to this day
,

the FT's
example is the one trotted out at any number of mobile app summit
-
style conferences.
That
'
s because, two years hence, it stands nearly alone as a mainstream example of a
"pure" HTML 5 app in the iOS and Android
-
dominated distribu
tion environment.

Why is this? There are performance "gaps" between what HTML 5 can do versus native
environments today, and these will persist. But I think technical challenges relate less to
performance versus native environments on the device itself, a
nd more to difficulty
integrating with back
-
end systems (think advertising networks, billing and payment
systems, authentication, content management).

These are things that can be, and are being, solved in the heterogeneous platform
environment of the post
-
BYOD era. To that end
,

w
hen IDC consults with IT
organizations on mobile app development, we counsel them on the importance of
tooling and solutions that can enable and accelerate multi
-
platform deployment
,

including HTML 5. Other key elements of the mobi
le app development environment
include interfaces to enterprise applications, collaborative functionality, and robust
security capabilities.

Perhaps a greater force dampening the proliferation of "standalone" HTML 5 apps
relates to monetization. Here
,

I'm

referring to the matter of "walking away" from app
store distribution and what may be established revenue streams. Publishers and
premium content owners may not like

revenue
-
share terms. But voluntarily leaving app
store distribution is an extremely risky

bet on the ability to maintain revenue continuity
and discoverability. It seems obvious that this is a bet that many will not yet entertain.

Let's revisit Clayton Christensen's articulation of interd
ependent versus modular
systems, a concept I introduced

in my previous blog.

I would argue that the route to
distribution for mobile applications looks today like two interdependent systems. These
come in the form of Apple and Google's development environments, and
,

to an extent
,

their control of storefronts (
enterprise storefronts may be viewed differently).

The mobile app is now a mainstream, familiar, and even expected engagement format.
We use apps to engage consumers, colleagues, customers, suppliers, and partners. If
the mobile app becomes the new face of

the Internet itself, it will have very different
economics from the traditional HTML
-
centric browser domain.

History

and
contemporary business theory suggest that these interdependent systems will give way
to more modular structures

(remember the early AO
L?
)
.

A mature, stable, and high
-
performing
mobile
HTML 5 environment will be
a key driver
of the

transition from interdependent to modular systems.

It's not a stretch to say that
this future is here now.
Companies such as HP and others

have created
scalabl
e and
comprehensive solutions that

let developers use open Web standards to create robust
enterprise
-
class
apps targeting multiple OS and device categories.
Of course native and
hybrid app development will remain as mainstream as ever and will be the
preferred
development approach for certain types of applications and use cases. But a
s these
solutions and their adoption grow over time,
mobile HTML 5 will begin impacting the
application development environment in new ways and in earnest.