Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
Competition in the Mobile
Ecosystem  Browsers &
Web Apps
A response to the NTIA request for comment
VERSION 1.0
Open Web Advocacy
contactus@open-web-advocacy.org
Page 1 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
1. Table of Contents
1. Table of Contents
2. Introduction
3. Proposed Remedies
3.1. Rationale
3.1.1. Third Party Browsers (iOS)
3.1.2. First Class Web App Support (iOS / Android)
3.1.3. In-App Browser Abuse Protection (iOS / Android)
3.1.4. Browser/Web App Equality
3.1.5. App Store Support for Web Apps (iOS / Android)
3.1.6. Safari Core Functionality (iOS)
3.1.7. No Chrome Preferencing Android)
3.1.8. Website Transparency Obligations (All)
4. Detailed Breakdown
5. Direct Responses to Questions
5.1. Definitions and Statistics
5.2. Software and Support for Developers
5.3. Avenues for App Distribution
5.4. App Users
5.5. Other Factors
Page 2 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
5.6. Potential Actions To Increase Competition
6. Brighter Future for Mobile Ecosystems
7. References
8. Open Web Advocacy
Page 3 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
2. Introduction
OWA would like to thank the NTIA for soliciting public feedback on the state of the competition
in the mobile ecosystem and for the thoughtful nature of the questions asked.
App stores are the primary way that users install apps on mobile devices, and for iOS are the
only way for users to install capable apps. Mobile operating systems and hardware
gatekeepers exert extreme control over users' devices long after they are sold. The app stores
and device restrictions allow these gatekeepers to extract extensive revenue from users not by
merit or user choice but by blocking the user from installing software from third party
providers that compete with their own offerings. They can further leverage this power to
charge the user additional fees (up to 42.8% on software and services from third party
providers.
The software market is highly competitive, it is not possible to add 42.8% fees and not have the
majority of costs be passed to consumers. OWA views these fees as an unreasonable tax on
the user.
If a developer for a given number of users requires $7 per a user to support and
maintain their product and the gatekeeper demands a 30% cut, in order to receive $7
the developer must increase their price to $10, thus increasing the price the consumer
pays by 42.8% 1/1.3  1.428.
Various regulatory agencies around the world (including in the US have proposed forcing the
mobile operating system gatekeepers to allow third party app stores. While this would
definitely go a significant way to alleviating these issues, app stores would still wield
considerable power as users would likely only install one or two. Additionally, depending on
implementation, Native Apps would still have to be written in different programming languages
and with different APIs which are bespoke to each operating system. This means that Native
Apps would not be interoperable between systems and would still require significant cost to
build and maintain multiple versions of the same App while providing the gatekeeper an
additional form of lock-in via non-transferrable investment and expertise in that ecosystem.
Web Apps are an untapped source of interoperable competition with the app stores.
In a world of capable browsers, Gatekeepers are not able to block their installation and would
thus have no leverage to add additional fees or block Apps that compete with their own
offerings.
Web Apps are applications installed from the browser that have the potential to provide users
an experience on par with Native apps. In competent, modern browsers they are interoperable
between operating systems, work offline, have a superior security and privacy model. They are,
in short, capable of amazing things.
Investment into the Web platform is highly desirable for future competition as it is both open
and free and allows any company to build upon it as a platform. Web Apps are increasingly
becoming the primary choice for companies on desktop operating systems. Commercial users
Page 4 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
now spend greater than 60% of their time within a browser and many companies such as
Adobe and Microsoft are heavily invested in adopting interoperable Web Applications Word,
Powerpoint, Photoshop, Illustrator) because they allow these businesses to produce higher
quality, maintainable applications which can target multiple platforms with near-zero friction to
users and low costs of operation for developers.
Apple’s original vision for applications on iOS was Web Apps, and today they still claim Web
Apps are a viable alternative to the App Store. Apple CEO Tim Cook made a similar claim in
Congressional testimony when he suggested the Web offers a viable alternative distribution
channel to the iOS App Store. They also claimed this during a court case in Australia with Epic.
Despite this Web Apps are not currently a viable competitor on iOS for three reasons:
1. Browser Ban
Apple’s Safari is the only browser on iOS as all other browsers have been effectively
banned.
2. Missing Features
Apple has refused to implement key features (some for more than 10 years) that would
allow Web Apps to compete with Native Apps from the App Store, both on iOS and in
Safari.
3. Bugs
Apple's iOS Safari is buggy and unstable compared with rivals, rendering it unviable as
a platform for applications.
iOS is a significant player in the mobile ecosystem landscape with 46% in the US. iOS users
enjoy higher average incomes, creating a market that can not be ignored. This goes a long way
to nullifying Web Apps development cost and interoperability advantages; not only for iOS but
also for Android. Companies that cannot imagine using the web to address the wealthiest, most
valuable users are forced to accept the costly logic of re-developing their services for each
native ecosystem.
We have written about this extensively in our paper “Bringing Competition to Walled Gardens”.
Regulators worldwide have acknowledged this anti-competitive behaviour and have proposed
legislative remedies to counter it.
The latest revision of the EU’s Digital Markets Acts notes:
"When gatekeepers operate and impose web browser engines, they are in a position to
determine the functionality and standards that will apply not only to their own web
browsers, but also to competing web browsers and, in turn, to web software
applications." (emphasis added)
The UKs Competition and Market Authority have highlighted this accurately and extensively in
their mobile ecosystems market study interim report and write (emphasis added):
Page 5 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
"As a result of the WebKit restriction, there is no competition in browser engines on iOS
and Apple effectively dictates the features that browsers on iOS can offer (to the extent
that they are governed by the browser engine as opposed to by the UI."
“Importantly, due to the WebKit restriction, Apple makes decisions on whether to
support features not only for its own browser, but for all browsers on iOS. This not only
restricts competition (as it materially limits the potential for rival browsers to
differentiate themselves from Safari on factors such as speed and functionality) but
also limits the capability of all browsers on iOS devices, depriving iOS users of useful
innovations they might otherwise benefit from.
They note that Apple has a multiple incentives to hold back Webkit, hindering Web Apps ability
to compete with the iOS App store (emphasis added):
"First, Apple receives significant revenue from Google by setting Google Search as the
default search engine on Safari, and therefore benefits financially from high usage of
Safari. Safari has a strong advantage on iOS over other browsers because it is
pre-installed and set as the default browser. The WebKit restriction may help to
entrench this position by limiting the scope for other browsers on iOS to differentiate
themselves from Safari (for example being less able to accelerate the speed of page
loading and not being able to display videos in formats not supported by WebKit). As a
result, it is less likely that users will choose other browsers over Safari, which in turn
secures Apple’s revenues from Google."
“Second, and as discussed in Competition in the distribution of native apps, Apple
generates revenue through its App Store, both by charging developers for access to the
App Store and by taking a commission for payments made via Apple IAP. Apple
therefore benefits from higher usage of native apps on iOS. By requiring all browsers on
iOS to use the WebKit browser engine, Apple is able to exert control over the maximum
functionality of all browsers on iOS and, as a consequence, hold up the development
and use of web apps. This limits the competitive constraint that web apps pose on
native apps, which in turn protects and benefits Apple’s App Store revenues.
They propose banning Apple from blocking third party competitors from porting their own
browsers to iOS, complete with competing engines.
The UK CMA found Apple’s security arguments justifying its ban on competitors unconvincing
(emphasis added):
Apple’s restrictions on competing browser engines: Apple does not permit the use of
third-party browser engines within its mobile ecosystem – all browsers are required to
use its browser engine, WebKit. We have not identified compelling evidence to date that
suggests that, for dedicated browser apps, the potential impacts on competition or
consumers from Apple’s WebKit restriction are justified on security grounds. We are
therefore seeking to assess the merits of a requirement for Apple to allow alternative
browser engines on iOS, at least for dedicated browser apps. This could be
implemented by requiring Apple to permit third-party browser engines to interoperate
with its iOS operating system, subject to those browser engines meeting conditions that
would address any risks that might arise from a greater choice of browser engines (for
example, complying with appropriate quality and security standards).
Page 6 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
Apple collected $72.3 billion USD in App Store fees in 2020. While it has not published the
costs of App Store review, payment processing, refund handling etc, it has been estimated that
the iOS App Store has a nearly 80% profit margin. Industries with healthy competition feature
leading firms with profit margins between 5 and 20 percent. This imbalance strongly implies
that Apple’s removal of functional competition in the App Store and beyond have distorted the
mobile phone market for software and services for about half of the US’s consumers, and a
substantially higher fraction of mobile commerce.
OWA believes that gatekeepers of mobile hardware/operating systems should compete to offer
additional services and software to customers on merit and user choice, not by blocking
applications that compete with their own or by charging consumers additional fees on
applications/services from competitors. OWA proposes a number of remedies (or desired
outcomes of remedies) that we believe are vital to restore competition and user choice to the
mobile ecosystem.
Only if regulatory or legislative action is taken can the bright future of software competition be
saved, not just for the US but for the world. Apple and Google were each started in garages by
two developers, we need to ensure this dream holds true for the next generation of
entrepreneurs. Competition is the key to unlocking maximum benefit for consumers.
Page 7 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
3. Proposed Remedies
In this section we outline goals that remedies should achieve, in priority order, to restore
meaningful competition.
We acknowledge that regulatory progress is often long, and that legal text needs to be written
in ways that are broad but still anticipate legal challenges and attempts to delay progress by
the gatekeepers. It is our hope that by stating exactly what we believe are the solutions to the
problems facing competition in the software industry, it may provide some insight into how to
construct statutes and regulations so that they can avoid expensive and (more importantly)
slow appeals processes by the gatekeepers. An important aim of proposals should be to create
clarity for all parties.
From most to least urgent, we hope that market interventions achieve:
1. Third Party Browsers (iOS)
Reversing Apple’s ban on competing browsers and browser engines by requiring Apple
to allow third party browser engines.
2. First Class Web App Support (iOS / Android)
Web Apps integrated into Operating Systems to the same level as Native Apps.
3. In-App Browser Abuse Protection (iOS / Android)
Preventing In-App Browsers from hijacking a user’s choice of browser.
4. Browser/Web App Equality (iOS / Android)
Ensuring Browsers and Web Apps are capable by having the same level of access to
hardware/software as Native Apps, the gatekeeper’s browser or operating system
services.
5. App Store Support for Web Apps (iOS / Android / Windows)
The ability to submit Web Apps to each of the app stores easily without having to
purchase a specific device (i.e. without a Mac).
6. Safari Core Functionality (iOS)
Ensure that Apple (as a primary gatekeeper) provides core/basic functionality for Web
Apps within Safari to enable them to compete against Native Apps.
7. No Chrome Preferencing Android)
Ensure that Google can not unfairly provide any preference to Chrome via licensing
agreements or app/operating system functionality.
8. Website Transparency Obligations (All)
Ensure Websites have to publish reasons to users for either not supporting a particular
browser or if prompting them to use another browser.
Page 8 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
3.1. Rationale
3.1.1. Third Party Browsers (iOS)
Reversing Apple’s ban on competing browsers and browser engines by requiring Apple to allow
competitive third party browser engines is key to unlocking competition with Apple’s walled
garden from an open, safe, and interoperable platform (the web).
Rationale:
The lack of third party browsers on iOS
Removes competitive pressure on Apple to invest in their own browser.
Incentivises Apple to not implement support for web functionalities, in order to
protect App Store revenue.
Provides no incentive to respond to developer needs or ensure that Safari is
reliable/not buggy.
Protects Apple’s Google Search revenue but deprives other browsers of that
revenue.
Ending this ban will:
Enable the ability to create functional Web Apps.
Reduce mobile application development, distribution and maintenance cost for
business, fostering innovation and lowering prices for end users.
Encourage the development of interoperable Web Apps, easing the switch
between mobile operating systems, and fostering the emergence of new ones.
Protect the future against anti-competitive behaviour and remove incentives to
stall progress.
IMPORTANT
Making a browser work on a new platform is a complicated and expensive task for any
browser vendor. To ensure competition is restored as quickly as possible, browser
vendors such as Google, Microsoft and Mozilla need to be given the clear signals by
the NTIA and other bodies that a goal of regulation is explicitly to enable such “ports”
and to create a clear expectation of Apple to facilitate these ports. Apple must allow
responsible browser vendors For example those that have the “browser” entitlement
that Apple provides today) to access all system features necessary to bring the best
Page 9 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
and most complete browsers that they can. This must include access to currently
private APIs that Apple does not provide access to. Apple must not be allowed to
obstruct other browser vendors with any onerous requirements.
Apple has given nonsensical security arguments to delay compliance in response to
other regulatory bodies. Care in messaging should be taken to ensure the plain
meaning of regulatory intent is not subverted by unfounded claims. No responsible
party will object to any reasonable security policies (backed by convincing evidence) to
protect users. Security policies need to balance utility and security, and this should
always prioritise the users interest over the Gatekeeper or Developers.
All operating system vendors delegate some of the responsibility regarding security to
browsers, including Apple. It is important to note that it is not the gatekeepers sole
responsibility to protect users. There will be many instances where providing low level
system access to competitors with strong security track records and dedicated security
teams is massively in the consumers interest, even though this means delegating
protecting the user to these competitors. These competitors can abide by reasonable,
narrow-scope, non self-preferencial security/privacy rules backed back by compelling
evidence.
3.1.2. First Class Web App Support (iOS / Android)
To create credible competition with gatekeeper app stores and platforms, it isn’t enough that
competing browsers be given the ability to access increased capabilities when users visit sites
in a browser. All modern mobile operating systems also support (closed) mechanisms for
“installing” web sites as apps (a.k.a. Web Apps / PWAs).
All browsers on iOS and Android must be granted the ability to create and manage this class of
applications. They must also appear fully integrated with the operating system, and provide the
user control over their operation without preference to gatekeeper controlled and taxed Native
Applications. On Android this would mean providing WebAPK minting to third party browsers.
Apple may need to provide new APIs for competing browsers to support this use-case, or
access to Apple’s internal mechanisms for delivering this functionality through Safari today.
Rationale:
First Class Web Apps on Android and iOS will:
Ensure that Native Apps do not receive any artificial preferential treatment over
Web Apps.
Provide users a method of controlling the permissions (i.e. notifications,
bluetooth, etc.) of Web Apps in the same manner they control Native Apps.
Ensure that the web ecosystem that is interoperable between all devices and
free and open can compete with proprietary platforms that only work for
specific ecosystems.
Page 10 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
Substantially decrease costs of developing safe, cross-platform apps by
providing a standards-based alternative.
Place pressure on native ecosystems to bring fees in line with value provided to
developers, reducing deadweight economic losses. This is accomplished by
providing a credible alternative that developers can switch to if the fees are
unreasonable.
Enable companies and developers to build “apps” on the web. To the extent that
Web and Native Apps are indistinguishable, it will increase competition and
finally make true Apple’s claim that Web Apps are suitable substitutes.
3.1.3. In-App Browser Abuse Protection (iOS / Android)
Apps should not be able to undermine a user’s choice of browser or otherwise intercept,
monitor or control a user’s web browsing to third-party sites. Apps that wish to act as a
browser should explicitly compete
as
browsers. A user’s choice of browser along with
preferences, security settings and their extensions (for example privacy enhancing extensions)
needs to be respected by every app that loads non-collaborating web content (e.g., excluding
first-party pages and ads). This should be enforced by the operating system.
The test to apply should be “
what should happen when a user taps a link inside an app?
Rationale:
Improves browser competition by ensuring that apps that act as browsers
explicitly compete with other browsers.
Ensures that Web Apps can not be broken, or prevented from being installed
from within a custom In-App Browser.
Ensures that a user’s choice (and thus competition) is respected.
Improves user privacy by ensuring their preferences, security settings and
extensions are respected and by not letting apps intercept user browser traffic.
This is covered in detail in our regulatory submission “
In-App Browsers  Subverting
User Privacy, Competition and Choice
.
Regulatory pressure needs to be applied to ensure that operating systems require that
apps respect user choice when it comes to browsers.
Developers should also be given explicit ways to mark their content as being only
loadable in the user’s default browser. Non-conformance with this should be subject to
monetary penalties.
Page 11 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
3.1.4. Browser/Web App Equality
Browsers and their Web Apps need sufficient operating system access to provide equivalent
functionality to gatekeeper’s own apps, browsers, system provided functions, and apps
distributed via the gatekeeper’s online stores. Gatekeepers should also provide full access to
hardware APIs, including NFC, Bluetooth, USB, serial, HID, accelerometers, proximity sensors,
elevation sensors, temperature sensors, light sensors, hardware buttons, location sensors, and
other sensors available to Native Applications.
The aim of this remedy is to remove any artificial restrictions on what browsers or Web Apps
can do either via software or hardware, owing to the superior untrusted-by-default security
model that modern browsers enforce regarding these capabilities. Operating systems
restrictions are therefore redundant when content is loaded in a sufficiently secure, modern
browser.
The EU’s Digital Markets Act contains interesting language in this regard (emphasis added):
“If dual roles are used in a manner that prevents alternative service and hardware
providers from having access under equal conditions to the same operating system,
hardware or software features that are available or used by the gatekeeper in the
provision of its own complementary or supporting services or hardware, this could
significantly undermine innovation by such alternative providers, as well as choice for
end users. The gatekeepers should, therefore, be required to ensure, free of charge,
effective interoperability with, and access for the purposes of interoperability to, the
same operating system, hardware or software features that are available or used in the
provision of its own complementary and supporting services and hardware. Such
access can equally be required by software applications related to the relevant services
provided together with or in support of the core platform service in order to effectively
develop and provide functionalities interoperable with those provided by gatekeepers.
The aim of the obligations is to allow competing third parties to interconnect through
interfaces or similar solutions to the respective features as effectively as the
gatekeeper’s own services or hardware.
Draft EU Digital Markets Act 11 May 2022  Page 46
We recommend the US adopt similar language in legislation that would enable browsers and
Web Apps to perform any function that either the software or hardware enables. Note that we
are not suggesting this be true for every class of program on iOS; just browsers, and only
browsers with a strong security track record.
The word available is important as there may be important functionality that is “available”
either in the hardware or in the software that is not used by the gatekeeper but which they still
prevent access to. It’s important to open the door for Browsers and Web Apps to access this
functionality as this will spur innovation.
Page 12 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
3.1.5. App Store Support for Web Apps (iOS / Android)
It should be possible to submit Web Apps to app stores without the developer being required to
purchase or own a specific device (i.e. a Mac with Xcode installed). This would need to include
both a web-based process to submit the Web App and the removal of any rules preventing
Web Apps from being submitted to the App Store.
Gatekeeper’s should remove strictly unnecessary steps or requirements from the process of
submitting an app.
Rationale:
Ensures that firms with limited resources do not have to choose between building
Native Apps for discovery in stores vs. a website for discovery on the open web.
Increases the likelihood that developers will choose open technologies that will work
across multiple ecosystems, lowering costs and improving competition.
Fosters investment and expertise into the free, open, and interoperable web ecosystem
instead of siloed native ecosystems. This directly lowers costs to businesses and users.
Reduces the cost of developing for the app stores by not requiring the developer to
purchase expensive hardware or maintain expensive, per-target-OS application
wrappers.
Page 13 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
3.1.6. Safari Core Functionality (iOS)
As an operating system gatekeeper with control over the default browser, Apple has the ability
to directly influence the uptake of Web Apps. Due to their ownership and collection of fees
from iOS’s App Store, Apple has strong incentives to limit the viability of Web Apps as an
interoperable and rent-free replacement for native iOS applications which can only be
discovered within its rent-extracting App Store.
Although competition between browsers is the most important and primary driver of
functionality, its essential that Safari, as the default browser, should be required to implement
basic functionality needed for Web Apps to ensure they are competitive including:
Install Prompts/Installability
At a minimum, Web Apps should be as easy to discover and require as few steps to
install as Native Apps. This includes the ability to display a banner and prompt the user
to install an app from Safari. As described in
Bringing Competition to Walled Gardens
sections 5.4.5 and 5.4.3.1 the Web App install procedure is deeply obscured, whereas
this is not the case on any other operating system.
Notifications
Push Notifications are essential for a wide range of applications
FullScreen, Badging, Deep Links, Screen Orientation Lock
Bluetooth and NFC
Without these open-standards-based APIs, entire categories of apps cannot be built
and developers cannot develop Web Apps that interact with hardware. This forces
developers to build Native Apps if they require this functionality. Apple has used
proprietary APIs over these otherwise interoperable protocols, and made them
exclusive to apps from the App Store which is subject to Apple’s taxation.
Apple has not yet published detailed justification of all their reasons on why blocking
Web Apps from using Bluetooth and NFC while providing those functionality to Native
Apps is justified on security/privacy grounds. This inhibits Web Apps ability to compete
with Native Apps and given their concern, Apple's implementation of Core Bluetooth
API for Native Apps is surprisingly permissive. Prior to iOS 13, (late 2019 applications
did not even need to ask for Bluetooth permission at all. Even now Apple has failed to
adequately lockdown Bluetooth for Native Apps.
It's revealing that Apple is not implementing Web Bluetooth over security/privacy
concerns and, in effect, forcing users to download a Native App with far broader
powers when they don't appear to have adopted equivalently strong protections within
their Native App ecosystem.
Page 14 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
Rejecting Web Bluetooth on these grounds is profitable, if nonsensical.
Rationale:
Apple benefits from poor uptake of Web Apps and can significantly affect how
successful they are by being the default installed browser on iOS with
significant market share. To prevent this, a minimum set of viable functionality
must be made available to all Web Apps across each platform.
Page 15 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
3.1.7. No Chrome Preferencing Android)
Google should not use their control over the operating system to provide unfair preference to
their own browser, Chrome, either through the operating system or agreements with partners
(i.e. device manufacturers).
This should include:
“Google Search App” (a.k.a.AGA/”AGSA” Must Respect Default Browser Choice
Google forces manufacturers to place the Google Search App on the first (primary)
homescreen of most Android devices. This search box drives an
enormous
amount of
web traffic and fails to respect user’s choice in default browsers, instead causing
Chrome to be invoked whenever users tap on search links that would otherwise take
users to their default browser.
This undermines browser competition on Android and is technically unjustifiable.
Google can know if the user’s default browser is not Chrome and must be made to
respect this choice especially with what is arguably the single most valuable browser
integration in any program or operating system.
Mobile Application Distribution Agreement MADA should not require Original
Equipment Manufacturers OEMs) to prefer Chrome, as for example requiring Chrome to
be the system default and on the first homescreen.
Users must have a way to disable “In-App Browsers” system-wide and gatekeeper rules
should enforce this policy.
Rationale:
For healthy browser competition Chrome should win market share through user choice
and merit, not control of the operating system, dark patterns or via contracts with
device partners.
Page 16 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
3.1.8. Website Transparency Obligations (All)
Gatekeepers such as Apple, Google and Microsoft control high profile websites. If some of
these Web Applications do not work in a specific browser, that can cause users to choose
another browser. This has been reported many times as an issue by multiple browser vendors,
including Opera and Mozilla.
Imagine the fictional scenario where Google decided not to support any browser other than
Chrome for Youtube. This would have a dramatic negative effect on browser competition.
There are, however, legitimate reasons why some applications can’t work in some browsers
including serious bugs and missing features.
OWA suggests that where a Gatekeeper’s website does not support a browser which has above
a 2% market share, they be compelled to publish a document containing detailed reasoning
that prevents support of certain engines. A convention could be established for these files to
improve their discovery, e.g. a compat.txt file hosted on the website root (or in the
./well-known/ directory). However provided, they should include details regarding:
Bugs
If the browser contains speed, bugs or stability issues that affect a user then the
document must contain a detailed description of these bugs with links to bug tickets
filed against engine projects.
Functionality
If the application requires specific functionality that the browser does not support it
must provide links to the missing functionality with a rationale why the
application/website requires that feature. For example the current version of photoshop
does not yet support browsers other than Chrome, but for legitimate reasons.
If a gatekeeper’s website wishes to display a “switch to a better browser” dialog then this must
also include specifics as to why it is better in the browser they are suggesting which needs to
be accurate. In the banner the gatekeeper should provide a link to a user readable description
of these reasons.
Gatekeepers must also provide “try anyway” links in these UIs to ensure that users of other
browsers that
may
be compatible are able to proceed to their applications, even if they appear
slightly broken.
It would also be worth considering applying these rules to all of the largest companies rather
than just gatekeepers.
These obligations will help to protect competition, by providing transparency to browser
vendors/developers and end users.
Page 17 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
4. Detailed Breakdown
This section is designed to contain the granular detail about each desired effect with specific
points that can be used to determine if the remedy is successful.
1. Third Party Browsers (iOS)
1.1. With own Rendering Engine
Browser’s must be allowed to choose and ship their own rendering engines
without restriction to how the rendering engine should work or behave. The
rendering engine is critical for pushing web capabilities forward.
1.2. With own Javascript Engine
Browser’s must be allowed to ship their own javascript engine or for any other
programming language and have access to any hardware functionality required
to make the engine performant and secure.
1.2.1. Including JIT
Javascript Engines with JIT Just in Time Compilation) must be allowed
for performance and compatibility reasons.
1.3. No App Store Restrictions
App Store restrictions must be minimal. For competition its essential that
browsers should be able to bring their vision of the web and apps without
interference from the operating system except on narrow scope and heavily
justified security issues which are in the end users interest. It is important that
app stores' rules impose no restrictions related to the user Interface,
functionality or technology.
1.4. Quick Update
Browsers need the ability to push updates very quickly and should either be
automatically accepted or given maximum priority for App Store review subject
to strict service level agreements from the gatekeeper. This is critical for
minimising the end user’s window of vulnerability (i.e. the length of time an end
user is exposed for) when bugs need to be patched.
Page 18 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
1.5. Ability to Install Web Apps
Browsers need the ability to install Web Apps and for those Web Apps to be
deeply integrated into the operating system.
An installed Web App should be indistinguishable from a Native App.
It should be technically possible (but not required) for a browser to create a
Web App that also has the ability to install other Web Apps to act as app stores.
1.5.1 No User/Operating System Interaction
The browser must be given full control to be able to install and update a
Web App by itself without the operating system interrupting the
process. Browsers need the ability to install Web Apps without
interaction from the user for specific use cases such as syncing Web
Apps installed between devices, and when restoring from backups.
2. First Class Web App Support (iOS / Android)
Web Apps once installed should have the same level of integration into the operating
system as Native Apps.
This includes:
2.1. Control over Web App Settings
The installed Web App should appear as an app on the Settings.app page, and
on each of the privacy menus (where relevant).
2.1.1. Custom Settings per Browser
Each browser can add individual settings per to its installed Web Apps.
2.1.2. Custom Settings per App
Each app can add custom settings to its Settings page.
2.2. No Double Permission Prompts
Only the Web App should need permission to perform an action. For example if
the Web App has permission to send notifications and its installing browser app
does not, then the Web App can still post a notification.
2.3. Equivalent Controls/Integration to Native Applications
2.3.1 Context Menu on Home Screen
The long-hold side menu needs to be supported.
2.3.2 Integration with VoiceAssistants
Browsers and Web Apps should be able to fully integrate with Voice
Assistants the same way Native Apps can.
2.4. Appears in all areas Native Applications are listed
Page 19 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
Browsers and their installed Web Apps should be given sufficient integration
into the OS that they can be listed anywhere Native Apps are.
2.5. Storage Control
Users should be able to control a Web Apps storage if they want too. Web Apps
should appear on the iOS “iPhone Storage” list as separate entities.
2.5.1 Offload App
Consideration should be given as to whether Offload App should be
implemented for Web Apps.
2.6. Reliable/Permanent Storage
The operating system can not delete Web App storage even under storage
pressure, unless specifically allowed by the browser that installed it or by
specific user request either via the installing browser or by operating system
storage controls such as “Delete App”.
1.5.2. WebAPK Minting or Equivalent Android)
Android must provide other browsers the ability to mint WebAPKs or an
equivalent mechanism of installing fully integrated Web Apps.
Longer term, there is room for debate as to whether Google should be allowed
to continue having exclusive access to mint WebAPKs from their server
infrastructure and whether they should delegate some of this control to
competing browser vendors similar to the trust delegation process used by SSL
certificate chains.
1.5.2.1 WebAPK Minting must be fast
If Google is to retain exclusive control over the minting service and
provides no alternative for browsers to be able to install Web Apps
locally, they should be responsible for minimising the time taken to
install a Web App. Currently there is a delay of several seconds while
waiting for Google’s minting server to respond which could adversely
affect all Web Apps.
Page 20 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
3. In-App Browser Abuse Protection (iOS / Android)
3.1. Third-Party Content only open via Users Default Browser
Operating systems need to protect users and competition by ensuring that if a
user visits a web page that it opens that web page using the user’s default
browser. The only exception should be for first-party content so that apps can
use a WebView to render their own content or from co-operating third parties
such as ad providers.
3.2. User can choose Remote Tab In-App Browser IAB or their Default
Browser Per App
Web content should always use their default browser to render content
although a user should be able to choose whether web content should open
within the app or whether it should open externally in another browser.
3.2.1 Default Behaviour
The user should have easy control to set the default behaviour of all
apps to either open in an IAB or in their default browser.
3.3. First/Second Party Content Opt-In Mechanism
A mechanism to ensure that WebViews can only load first and second party
content would need to be implemented to ensure they can strictly only open
first and second party content.
3.4. Only actual Browsers can be a Default Browser
It is important for competition that popular apps defer browsing the web to a
user’s default browser. Only apps that are actual browsers should be able to
receive a browser entitlement. For example, popular social media apps and
messaging services should not be able to receive a browser entitlement.
3.5. Users must be able to install Web Apps directly from within IABs
When a user is browsing the web from within an In-App Browser they must be
able to directly install a Web App.
Page 21 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
4. Browser/Web App Equality on (iOS / Android)
Browser’s should be able to access the following functionality. Web Apps should be able
to access that functionality as allowed via the Web Apps installing browser.
4.1. General
4.1.1 Hardware
Browsers and Web Apps should be able to make general use of any
hardware the device provides including functions enabled by the
hardware but disabled at the software layer.
4.1.2. Same access as Gatekeeper’s Browser
Browsers should receive (but not limited too) all the same abilities/the
gatekeepers browser receives.
4.1.3. Same access as Native Apps
Browsers should be able to (but not be limited to) use/access any
function Native Apps use.
4.1.4 Same access to System Apps and Functionality
Subject to narrow scope security concerns, browsers should have
access to (but not be limited to) the same functionality that System
Apps use.
4.2. Communication Protocols
NFC
Bluetooth
USB
Serial
HID
Networking
Low Level Networking TCP/UDP
Firewalls
VPNs (including Apple’s Private Relay feature)
4.3. Sensors
Location/GPS
Accelerometer
Step Counters
Proximity Sensors
Elevation Sensors
Temperature Sensors
Light Sensors
Other
Page 22 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
4.4. Health Related Sensors
Heart Rate
Glucose Monitoring
Other
4.5. Hardware Buttons
Volume Up/Down
Off/On
Ring/Silent
Back Tap
4.6. Lower Level Functionality
CPU Functionality
GPU Functionality
Memory Related Functionality
Disk Related Functionality
Security / Cryptographic Related Functionality
Video/Audio/Image Encoding Functionality
Managing Processes Spawn/Terminate, etc.)
4.7. Telecommunications
Baseband/Radio/5G/4G Related Hardware Functionality
Telephony Related Functionality Receiving/Making Calls)
SMS/MMS Related Functionality
Wake from Sleep
Ability to swap about the default phone call application
4.8. Device Management (iCloud Replacement)
Device Backup
Device Restore
App Data Backup
App Data Restore
Device Fleet Management
Managing fleets of devices, with the ability to apply settings/policies
remotely to those devices.
Locate Devices
The ability to run an interoperable device finding and remote
management service (i.e. Find your Android/iOS/Windows/Linux/mac
devices from the same service)
Remote Lock
Remote Alarm
4.9. Filesystem Access
Page 23 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
4.10. Software Access Equality
4.10.1 Media Player/Music Library
The ability to interact with or replace the system default media player
including on the lock screen (i.e. Apple Music).
4.10.2. Payment Services
The ability for browsers to integrate with the default payment service
(i.e. Apple Pay).
4.10.3. Voice Assistants
The ability to interact/integrate and replace voice assistants.
4.10.4. Time Tracking/Focus Management
The ability to track application use and manage application time limits
and other restrictions.
4.10.5. Widgets
The ability to create and place home screen widgets
4.10.6. Control Center (iOS)
The ability to add icons and widgets to control center
4.11. AirDrop (iOS and Android Equivalent)
Although not covered by browsers or Web Apps, OWA believes consumers and
competition would benefit from an interoperable standard for Apple’s AirDrop,
and its Android equivalents.
4.12. AirPlay (iOS and Android Equivalent)
As above but for Apple’s AirDrop its Android equivalents.
4.13. Future Devices Apple / Meta / Google)
Future devices such as AR/VR glasses and smartwatches should enable the
installation of third party browsers and Web Apps in order to prevent yet
another gated ecosystem.
Page 24 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
5. App Store Support for Web Apps (iOS / Android)
5.1 External Web Based Service
The app store should accept Web App Submission via a Public Web Based
Service with a goal of ensuring specific hardware or operating systems are not
required and to enable automated deployment tools that can submit to multiple
app stores from different providers reducing the effort and cost for developers.
5.2 No Specific Operating System or Hardware Required
Developers are not required to use any operating system to submit the Web App
to the app stores.
No Mac required.
No XCode required.
5.3 Minimal Steps
The steps and complexity of submitting a Web App are reduced to the minimum
required. Note that signing up for a developer account on an app store can be a
separate process.
6. Safari Core Functionality (iOS)
6.1 Install Prompts Installability)
6.1.1 Minimum Bar is Equivalent to Native
The install procedure for Web Apps is at least as easy and requires equal
or less steps as the procedure for installing Native.
6.1.2. No Regression in Installation Speed
Any changes should not noticeably decrease the speed it takes to install
a Web App
6.1.3 In-App Browser Installation
All In-App Browsers must support easy and seamless installation of Web
Apps
6.2 Notifications
6.2.1. Must be Free
6.2.2. Must not Require an Apple Developer Account
6.2.3. Delivery time equal to Native Apps.
6.2.4. Users can easily enable/disable them.
6.2.5. Full support of the Push API specification.
6.3. Fullscreen for <canvas> and other non-video content.
6.4. Badging
6.5. Deep Links
Page 25 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
6.6. Screen Orientation Lock
6.7. Bluetooth
6.8. NFC
7. No Chrome Preferencing Android)
7.1 “Google Search App” Must Respect Default Browser.
7.2 Mobile Application Distribution Agreement MADA should not require
Original Equipment Manufacturers OEMs) to prefer Chrome
8. Website Transparency Obligations
8.1 Gatekeepers must have a compat.txt with detailed explanation for
browsers with a market share of greater than 2% that the website does not
support including:
Bugs
With links to bug tickets
Functionality
With links to missing functionality
8.2. “Try Anyway” button must always be available
8.3. “Switch to a browser” must describe the benefits to the user including in
terms of speed, performance / functionality
8.4.Consider applying all large companies
Page 26 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
5. Direct Responses to Questions
Note: NTIA Questions are Listed in Dark Blue
We have attempted to answer all questions that relate to Browsers and Web App competition.
For a more detailed and in depth discussion, please see our paper "Bringing Competition to
Walled Gardens".
5.1. Definitions and Statistics
1. How should we measure whether the app ecosystem is competitive?
A key problem in the analysis of app ecosystem competitiveness is the challenge of measuring
deadweight losses from anti-competitive behaviour. To counteract this difficulty, we suggest a
focus on the structural costs and competition for services necessary to provide essential
functionality within an app ecosystem.
For instance, the web ecosystem enjoys healthy competition among many aspects of the stack
(all of the various services required to deliver a website or app) and many hosting providers
allow developers to create apps on their infrastructure.
Many “CDNs” Content Delivery Networks) compete on cost and features to optimise delivery
of Web Apps. There is broad and diverse competition in DNS and TLS certificate provisioning
services. Lastly, there is competition for users between browsers (the platform or “runtime” for
Web Apps) and for payment processing services for those apps (e.g., Stripe, PayPal, Trustly,
and Google Pay).
App ecosystems that force these services (and many others) to be fused into a single “bundle”
by gatekeepers remove competition from the ecosystem. Identifying these bundled services
and charting the ways in which interventions will increase competition and deliver lower prices
seems like a productive way to consider this measurement challenge.
Item 4 in the detailed breakdown in Section 4 can provide a good starting point to identify
these bundled services.
1a. How should the “success” of an app be measured? 
Developers have a wide variety of interests and will therefore define success differently based
on the positioning of their business in the market. In many cases, “apps” aren’t the goal, but are
instead fronts or outlets to
services
that developers maintain across many operating systems
and form-factors (device types, desktop/mobile etc). There isn’t a single, clear line between an
app and a service, but in the modern era, nearly all commercially important “apps” require
functionality that requires an internet connection and the ability to call a provider’s service
APIs.
Page 27 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
In general, we can think of successive steps in success for a publisher as:
The ability to produce a front-end (“an app”) with intended local and remote
functionality without burdensome or unrelated costs.
Low-friction discovery by users who may be interested in a service, often through
competitively-priced ads.
The ability to monetize services provided to users with competitive transaction
overhead / fees.
Access to tools and techniques that enable profitable services to reach new users over
time without gatekeeper interference (beyond core platform security responsibilities).
1c. Does the reported total of the number of apps available at any one time in an app store
have bearing on the state of competition among apps or particular categories of apps?  22
No.
There are plenty of low-effort/low-quality apps that are either unmaintained and in general not
of merchantable quality on both the iOS AppStore and on Google Play. This includes apps who
rely on an intrusive level of advertising or are basic/simple clones of other apps.
Measuring the volume of high-quality apps is difficult, and not automatable. If it were
straightforward, presumably Gatekeeper’s app stores would include fewer low-quality apps.
2. Are there any important and specific entities (or categories of entities) such that it would
be a mistake to omit—or improperly include—them by defining the “mobile app ecosystem”
to focus on mobile devices, such as phones and tablets?
From a technical perspective mobile devices such as phones and tablets are not significantly
different from any other personal computing device such as a Laptop or Desktop Computer.
Some differences are:
Their primary input is touch rather than a keyboard/mouse.
The volume of private data that resides on the device (although in some cases people’s
Laptop/Desktop computers contain as much or more private data).
A greater usage of the device for payments (although other devices are still used for
payments)
That its carried by the person for the majority of the day.
However the key difference is the mobile app ecosystem operating systems are significantly
more restricted than desktop operating systems.
Page 28 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
Rules should apply without distinction to general purpose consumer computers (including
desktop/laptop/tablet/phone/vr headsets) although a special focus is needed on the mobile
ecosystems as that is where the majority of the current anti-competitive issues are.
If Apple were to release a Macbook with iOS instead of macOS (technically feasible today), the
same anti-competitive issues would be in play.
2a. If so, how should this study be scoped so that it is optimal but feasible?
OWA recommends that NTIA study the practical substitutability of Web Apps and Native Apps
under the current competition regimes constructed by gatekeeper policies. Apple has claimed
in court that Web Apps can be adopted in lieu of its App Store terms and conditions, but
developers report that this is (charitably) a misleading claim.
To the extent that Web App substitution requires true competition between browser engines,
and fair terms for browser competition, we encourage NTIAs study to include these concerns
as a way to emphasise effective, available, and low-cost structural reforms rather than small
interventions which may have limited effects in practice.
2b. For example, should mobile apps offered specifically for enterprise use (
e.g.,
for use by
businesses, not for consumers) be considered in this study?
Both Google and Apple provide enterprises (businesses that own devices issued to their
employees) with alternative policy mechanisms for operating and managing those devices,
including the ability to construct custom applications not subject to app store policies.
If NTIA considers these policies, OWA suggests that they be taken as evidence of the broad
leeway gatekeepers have in creating tiers of trust that enable certain categories of apps that
enjoy different rules of the road. That flexibility will sit at the heart of any new policy regime
regarding browsers, and we note that it is not new or unexplored.
3. Apps are not all the same. For example, some have different technical features and
capabilities (
e.g.,
location-based apps compared to messaging apps), while others are
bound by specific regulatory guardrails (
e.g.,
banking apps or children's apps). In the
context of framing competitiveness within the ecosystem, how should we categorize types
of apps so that they are grouped by distinguishable barriers and other significant factors?
Are there ways to best categorize or segment the market to diagnose specific market
barriers, such as those that could impact app developers, or consumers?
To the extent that OWA is concerned about app categorization, we note that Apple (and to a
lesser extent Google) have
already
created a specific category of app – the browser – that is
special and historically distinct from a policy enforcement point of view.
Instead of applying its long-running policy of removing apps that Apple has viewed as
“duplicating functionalityof iOS, Apple has instead
forced
makers of other browsers to deliver
fully identical features in many areas of browsing, thanks to Apple’s own restriction on browser
Page 29 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
engine choice. Yet Apple has, for more than a decade, allowed those browsers to exist and has
recently cemented this category with a new entitlement.
Apple thereby admits that browsers are a unique category with special powers relative to other
apps. You can see this from Apple’s own practices.
We recommend that regulators ratify this category. Browsers
are
special. They require
enhanced powers to keep users safe and to deliver advanced functionality as a competing
platform / runtime to the Native App platforms that gatekeepers tax developers and service
providers to use today. Browser makers intentionally accept this responsibility and should be
rewarded for their care.
We are certain that NTIA will consider other categories and factors in its final draft, but insofar
as we believe the web can help repair the mobile app ecosystem, browsers must be
empowered to be capable enough to deliver competitive apps.
3a. Should distinctions be made based on type of content and app functionality?
OWA is narrowly interested in questions related to Web App substitutability with Native Apps,
and the enabling factors related to browser competition that can enable or preclude the web
from being a true substitute for native.
To the extent that web browsers are already understood to be available within stores without
regard to the content they display or enable, we believe that their existing categorical
differences to other apps listed in stores are justified. Should gatekeepers attempt to preclude
the distribution of Browsers through app stores on the basis of the content they can display,
we believe this would be anti-competitive and adverse to the interests of users and publishers.
3b. Should distinctions be made based on the level of hardware or operating system
integration required for the app to function? For example, categories might include apps that
access location data, special-purpose hardware (
e.g.,
near field communications), secure
elements for payment, or other credentials.
OWA is neutral on questions of functionality in most classes of apps. However, we do believe
that the existing, expanded levels of system access that modern browsers enjoy relative to the
constraints that Apple (e.g.) places on other apps are justified. To facilitate effective browser
competition, safe, sandboxing browsers with a strong track record of security must be granted
deep system access to deliver competitive features.
Page 30 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
4. How should web apps (browser-based) or other apps that operate on a mobile
middleware layer be categorised?
Web Apps run on a middleware layer (the browser) that is relatively unique in the world of
software in that a large fraction of the UI, security, and privacy choices related to the
applications themselves are subject to choices users can make about their browser, without
requiring expensive or difficult replacement of hardware or operating system.
In a pure sense, the web and the practice of effective browser choice is unique, in part,
because it represents the pinnacle of US policy choices that promote the legal and regulatory
basis for industry standardisation, and through it, interoperability. The emergence of a
computing platform that facilitates user choice in platforms/runtimes without large
expenditures of money is a shining example of US industrial policy delivering lower costs and
consumer benefits.
Apple’s iOS and Google’s Android arrived into a world already transformed by the web, but have
succeeded in closing over a commons that had been tenuously opened through competition
and interoperable standards.
This reduction in the ability of browsers to credibly satisfy user and developer needs coincided
with early smartphone system constraints, but those constraints no longer dominate. Nearly
every smartphone sold in the US contains more than 2GB of RAM, and all iOS devices feature
astonishingly fast CPUs. In this context, the browsers and Web Apps could easily provide a
strong substitute for Native Apps, as they do on desktop operating systems today, where
browsers (including on Apple’s macOS and Google’s ChromeOS are able to provide compelling
feature-sets that enable developers to provide high-end experiences without jeopardising user
safety.
NTIA may wonder if or how other middleware layers should facilitate app delivery, but OWA is
agnostic on this point. We only note that browser makers have a unique track record of
delivering high security for their middleware, often leading the industry in technologies such as
sandboxing, hardware-enabled protections, and transport security. Perhaps other communities
could provide similar value, but we believe that its enough to admit (as Apple does by
continuing to ship Safari to its iOS users) that browsers and web are a unique category. We
don’t need a complete theory of middleware to assert that high-quality browsers are more than
safe and capable enough – at least when operating system vendors and gatekeepers do not
hobble them.
Page 31 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
5. There are some indicators that there is a difference in kind between some apps that
generate large amounts of money or are downloaded often and most other apps. For
example, one industry analyst reported that 97% of publishers that monetize through the
Apple App Store earned less than $1 million per annum in 2021, compared to other reports of
more than $1 billion earned by the top 13 apps (including games) on both Apple and Google
platforms. 23 What is the best way to assess the competition environment for less popular
apps and start-ups?
App stores are heavily centralising and exacerbate “winner take all” dynamics in tech,
particularly when coupled with gatekeeper’s highly opinionated and arguably self-serving
policies regarding implementation technologies.
Small publishers and startups are least able to afford custom, per-OS development efforts. This
helps to ensure that large, well-capitalised firms which can afford multiple development teams
enjoy an advantage over smaller firms that may only be able to reach a fraction of users
through app stores, or may not be able to maintain high quality apps thanks to the added costs
of having to re-develop identical features in multiple disparate technology stacks.
OWA notes that the web has levelled this difference for several decades on traditional PCs, and
we believe that if mobile gatekeepers are forced to facilitate true browser choice on iOS and
Android, a similar lowering of prices for new entrants and smaller players will come to mobile.
Its worth dwelling briefly on the question of why mobile gatekeepers have benefited from this
oligopolistic situation. Would they not be better suited to richer product catalogues in their app
stores? It isn’t clear that they would. Built in barriers to competitive interoperable Apps due to
the required duplication of development effort Apps must be rewritten for each operating
system in different programing languages) combined with the ability to block any App from
their app store and the ability to give "sweetheart" exceptions to app store rules gives
oligopolists additional leverage and rent extraction opportunities. Finally, as these firms may
also be potential acquirers of startups in the technology industry, added leverage over them
helps their bargaining position when acquiring talent or pricing in potential merger and
acquisition scenarios.
5a. Can any potential harms, such as deficiencies in data security and privacy protections,
be traced back to the current market imbalance?
Yes. For the Native App ecosystems a user has a choice either to purchase an Android device
or an iOS device, from that point they can have marginal control over how the apps operate.
Web Apps by comparison are installed from the user's chosen browser. Every browser by
default is significantly more sandboxed and has greater privacy and security than any Native
App, requiring individual permissions to access potentially privacy breaching functionality.
Users can use many methods to control security and privacy using the web where no
alternative exists in Native Apps.
Native apps do not provide the user any agency in controlling the app. There is no equivalent to
browser extensions for Native Apps. User’s can make choices about which browser they want
Page 32 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
to use and that facilitates both privacy and security choices that then apply to every app they
install. Operating System gatekeepers have not provided any means of achieving this through
Native Apps.
5b. Is there evidence to suggest that consumers are less likely to avoid or stop using a
particular app even if they would prefer a more privacy enhancing environment because of a
lack of competitors offering similar services?
Application developers have perverse incentives to move users to Native Apps and away from
the web, even when they are delivering nominally identical services. This is most true regarding
advertising and tracking. The set of permissions and capabilities provided without active user
consent to Native Apps has created a huge market for privacy-invasive mobile ads networks.
Those ads platforms can monetize better than web ads (in many cases) because the tracking
that underlies the ability to accurately target ads at specific users is much easier in the less
privacy-preserving Native App context.
As an example of this, it is worth considering that recent drama over Facebook and Apple
tussling over “Ad IDs” relates to a capability that has never been exposed to the web, out of
(legitimate) fear regarding the use of such an identifier for tracking purposes. That Apple and
Facebook are arguing over moving back to a position that is only as privacy-preserving as the
web has been signals the gap in potential value to advertisers over this tracking. See also Eric
Seuferts two-part series “Apple Robbed the Mob’s Bank” [Part 1, Part 2].
Page 33 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
5.2. Software and Support for Developers
6. What unique factors, including advantages and obstacles, are there generally for app
development — especially start-ups — that are relevant for competition?  24
Startups face higher hurdles in a world dominated by app stores.
Unlike the previous generation of startups in the “Web 2.0” era, modern startups must factor in
heavy advertising costs related to user acquisition in the app store. Because the Native App
model is more cumbersome (users cannot just click a link, they must download a full application
before they can begin to experience a service), strong advantages attach to incumbents that
have already convinced users to download their apps.
This winner-take-all pressure exists not only regarding device space and acquisition costs, but
also in the ways that app store operators create opaque and heavily favouritism-based “hits”.
Developers of Native Apps have reported for many years that you need to “know someone” in
Cupertino to arrive in the coveted “featured” list within the App Store. For small businesses, this
can be make-or-break, and instead of a transparent market that responds efficiently,
gatekeepers wield extreme power over small firms that may be potential acquisition targets – or
future competitors.
The NTIA will no doubt be aware of many examples of Apple gating publication of apps that
compete with Apple services into its store on capricious and opaque grounds. While OWA
believes that reform of these publication rules is overdue, we are more interested in systemic
fixes to this entire class of market power imbalance; namely the ability of browsers to provide
effective, capable web-based substitutes for most classes of applications, providing
developers with a credible alternative to “knowing a guy” in the San Francisco Bay Area.
The unbridled power of todays’ gatekeepers harms users in multiple ways. First, it reduces
choice and increases switching costs, driving up potential margins and locking users into
single-vendor ecosystems where their transactions are taxed. Next, it reduces data portability.
And lastly, because of the policies put in place to render the web less competitive, users suffer
the downsizes in security related to a mandated monoculture.
None of these are signs of a healthy market or effective choice.
7. Are there particular obstacles preventing more development from different communities,
such as by location/region, ethnicity/race, language, or gender?  27
Apps on Android and iOS for the app stores are written in different programming languages.
This means to have an App work on multiple operating systems often means writing the App
several times, this is prohibitively expensive. Web Apps could allow developers to write a single
App (one code base) and have it work on all systems. However we do not believe this is
currently viable due to rules imposed by Apple.
We have written extensively about why Web Apps are not able to effectively compete on iOS in
"Bringing Competition to Walled Gardens".
Page 34 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
Making apps that are accessible to people with disabilities is a solved problem on a mature
platform like the Web (which is not to say that every web developer is competent at
implementing these solutions). Because of the prohibitive cost of making one single-platform
app for iOS and another for Android, many organisations choose to use a ‘middleware’ language
such as React Native, which can then generate a version for iOS and one for Android. However,
React Native is an immature product, with 1,900 accessibility issues logged by developers but
not yet fixed by its vendor, Facebook.
This also negatively affects people with lower income. Android users on average have lower
incomes than iOS users. Android devices are typically cheaper.
As iOS users are on average wealthier it is not surprising that they spend more. Developers are
driven by revenue and financial constraints. As in many cases developing/maintaining the same
Application twice in two or more different programming languages is prohibitively expensive
and Web Apps (due to constraint imposed by Apple) are not viable, the developer may choose
to develop a single App for a single operating system. As iOS has the wealthy users this is
typically the operating system chosen. Many Apps are iOS exclusives as a direct result of these
rules.
This is clearly in Apple's favour and a disincentive to either allowing effective competition of
third party browsers on iOS or allowing Web Apps to effectively compete with the iOS App
Store.
This means less wealthy users on Android are deprived of useful Apps that remain locked in
iOS by these barriers.
8. Are there studies or specific examples of the costs or advantages for app developers to
build apps for either, or both, of the main operating systems, iOS and Android (which have
different requirements)?  (28
8a. What are the challenges specific to multi-platform development and how can they be
mitigated?
The primary concern in cross-platform development is access to capabilities and compatibility.
To the extent that one platform in a list of supported systems lacks a specific feature,
developers can struggle to adapt.
When cross-platform development also includes multiple runtimes, the potential for differing
features between those runtimes can also add an additional tax to developers attempting to
build from a single codebase.
Web developers have become adept at adapting to many gaps, but there are limits, and
additional difficulties caused by a single platform or runtime inevitably add expense to projects.
These are key reasons why Apple’s Safari browser is thought of poorly in the web development
community today. Apple policy mandates that Safari’s WebKit is the only available engine on
iOS. Therefore, it is impossible for a developer to recommend another browser to gain
Page 35 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
additional features or reduce exposure to showstopping bugs. Combined with the slow pace of
feature development, and generally lower quality of Apple’s implementation, developers
attempting to deliver cross-platform experiences on the web struggle with basics.
The alternative is to submit to the logic (and taxes) of the App Store. This is often vexing, but
Apple’s low-quality web implementation ensures that developers understand that they will have
superior access to features and support using Apple’s proprietary tools.
Some developers try to bridge these gaps, using libraries like Cordova or React Native within
their App Store-listed Native Apps, but infusing them with small aspects of web attributes
(faster development, ability to update out-of-band), however developers understand that Apple
may use policy excuses to deny them the ability to publish new versions of their apps if they
are discovered to have become too “web-like”.
8b. What are the costs and advantages of developing standalone apps for these platforms
relative to other means of providing the same services or content, such as web apps, which
can operate across platforms?
Web Apps (if the current barriers were removed) offer a number of significant advantages over
Native Apps.
Interoperability
For Web Apps interoperability between operating systems becomes almost automatic.
Provided the full version of each major browser (complete with their engines) are allowed and
they are given sufficient access to the operating system to provide the features that Web Apps
need, browsers provide developers a consistent and stable platform to develop on.
Multi-operating system projects can be built with a single code base and do not need to rewrite
the App multiple times in different programming languages.
Lower Cost
Native iOS Apps have to be written in Apple created languages Swift/Objective C and system
APIs that are specific to iOS. You can not run an iOS Application on an Android device (typically
written in Java/Kotlin). Web Apps however only have to be written once, in one language and
then can run on any device.
In order to support multiple platforms for their application (without using Web Apps) a business
will need to produce separate applications for iOS, Android and Windows. This typically
involves expanding the team and having multiple code bases effectively multiplying the
workload. Keeping multiple code bases in sync is also difficult and time consuming even for
large businesses.
This can increase by 25 times the development and maintenance cost. These costs must be
passed onto consumers. For some smaller businesses it can cause the product not to be
produced at all.
Page 36 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
Higher Quality
Users also suffer lower quality applications because companies have to split their resources for
developing and maintaining applications between two or three platforms, while they could have
focused them on only one application.
No additional Gatekeeper Fees
Additionally the business must then pay 1530% of their revenue to Apple (further increasing
costs by up-to another 42.8% for users).
For example, imagine you require $10 per user to cover the costs of developing, publishing and
maintaining an Application. The app store that you are selling your App in decides to add a 30%
fee. In order to still receive $10, you now need to charge $14.4, which is a 42.8% price increase
for the end user. However the actual price increase will be higher as when you increase the
price by 42.8% you will lose a percentage of users as you move to a higher position on the
demand curve, causing the equilibrium price to be even higher.
Gatekeeper can not block competitors
"there’s no safety, security or privacy issue  Apple just doesn’t like those apps.
Benedict Evans  Technology Writer
"It should not surprise you to know that Apple’s interpretation of its text often seems
capricious at best and at worst seems like it’s motivated by self-dealing."
Dieter Bohn  The Verge
Gatekeepers like Apple effectively ban certain categories of Applications they don't like or that
potentially compete with their own offerings.
No Arbitrary App Store Review Process
"there are endless horror stories around curation of the store. Apps are rejected in
arbitrary, capricious, irrational and inconsistent ways, often for breaking completely
unwritten rules."
Benedict Evans  Technology Writer
“There's a lot of talk about the 30% tax that Apple takes from every app on the App
Store. The time tax on their developers to deal with this unfriendly behemoth of a
system is just as bad if not worse”
Samantha John  CEO Hopscotch
Page 37 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
The App Store review process can be an extremely stressful and chaotic experience for
businesses and developers despite problems with actually stopping malware.
9. What role does interoperability play in supporting and advancing a competitive mobile
app ecosystem?
Interoperability, as accomplished through web standards and browser competition, has been
crucial in improving outcomes for developers building applications on the web aimed at
desktop PCs.
Compatibility is somewhat less relevant when critical capabilities are missing. Until a platform
can meet the basic needs of a developer, the ability to “port” that experience to other contexts
is not a material concern.
Apple scuppers both capability and compatibility, harming the web’s ability to lower costs and
increase competition on mobile. Apple’s Safari and WebKit engine lack key capabilities, driving
many apps off the web entirely and into proprietary app stores.
Web Developers whose apps can in theory work on the web even with Safari’s constraints often
report extreme difficulty in achieving interoperability and standards conformance. This adds
costs and is a large disincentive to web development. These costs and unpredictability create
pressure to move important work into proprietary platforms where it will be better supported by
the runtimes.
OWA believes that the best solution to this problem is true browser engine choice on iOS and
effective browser choice on all mobile operating systems. True browser competition places
competitive pressure on lagging engines otherwise they lose market share through developers
recommending alternatives. It also ensures that developers can use tested, cross-platform
runtimes that are not dictated by gatekeepers with an incentive to keep the web from
competing with the local Native App platform.
Both of these effects are crucial to creating true competition in the mobile app ecosystem.
9a. What are the key characteristics of interoperability as it relates to the mobile app
ecosystem?
For users:
1. Being able to install the same app on different devices with different operating systems.
This lowers switching costs and creates a larger library of available software.
2. Unified billing in services across devices further reduces switching costs and creates a
richer market for quality software.
For developers:
Page 38 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
1. Interoperability reduces costs (“write once, test everywhere”), allowing teams to focus
on delivering features rather than rewriting them multiple times.
2. Web interoperability creates a gatekeeper-free, safe distribution mechanism that allows
developers to publish updates to app functionality and content without requiring costly
app review processes. Because Web Apps run on a safer, more mediated/secure
middleware runtime, trust decisions can be made dynamically by the underlying
browser, which allows browsers to worry less about the dynamic nature of the content.
9b. What other barriers (
e.g.,
legal, technical, market, pricing of interface access such as
Application Programing Interfaces APIs]) exist, if any, in fostering effective interoperability
in this ecosystem?
Apple, Google, and Facebook have created unique barriers within their ecosystems to effective
interoperability and competition.
As documented in our paper “Bringing Competition to Walled Gardens” and Alex Russell’s blog
posts on the topic, each of these players has created severe challenges for developers
attempting to rely on browsers as their primary mechanism for delivering compelling
experiences.
Apple
Apple’s policies against browser competition and antipathy towards web content in its
App Store are infamous. Combined with mandated use of an engine that trails the
industry on nearly ever metric (compatibility, features, performance, and security),
Apple bears chief responsibility for the low prevalence of interoperable software in
todays mobile ecosystem. The list of missing APIs and features is too long for us to list
here. Please consult our paper for a fuller explanation.
Google
The Google Search App for Android fundamentally undermines browser choice, harming
competing browsers and reducing the potential of developers to trust in the web as a
reliable platform. Users that change their default browser away from Chrome will begin
to experience the web as a fragmented, amnesiac (different logins / cookies etc)
experience. This reduces success for both developers and users as sessions, logins,
and permissions are no longer coherent.
Facebook Meta)
Meta’s apps (including the Facebook apps on iOS and Android) ignore browser choice
and instead inject their own (broken) “in app browser”. This provides Meta a privileged
position for collecting user data, but harms third-party developers and users that may
have configured different privacy and security policies.
Meta’s “IABs” lack access to baseline Web App APIs, even on platforms where they
would be available Android).
Page 39 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
9c. How are these barriers different or similar than those present in other ecosystems?
No other general purpose operating system other than iOS bans third-party browsers or their
engines. Even ChromeOS provides the ability for Mozilla and others to bring their own “real”
browsers to users though the Play Store. Every other operating system also allows browsers to
provide the ability for Web Apps to be installed as if they were Native Apps.
Lack of functionality in Safari is less critical where alternate browsers are allowed.
Note: Safari is only available on two operating systems (iOS/MacOS as Apple has chosen not
to port it to Windows, Android, ChromeOS, or Linux. Competing browsers such as Chrome,
Edge, and Firefox are available on all of those platforms using their own engines.
9d. How does data portability, or lack thereof, factor into consumers keeping the same app if
they switch from one operating system (iOS or Android) to another?  29
This is a significant concern.
If a user has many apps in one native ecosystem, moving to another proprietary,
non-interoperable ecosystem may force them to repurchase apps they already owned (due to
differences in mandated payment platforms), worry about gaps in app availability, or bear
additional costs related to data migration.
Web applications are, generally, a solution to this problem as data is synchronised with cloud
services by default. This is what allows transparent access, for example, to one’s webmail
across devices, regardless of operating system. Apple and Google have incentive to undermine
this model of application development in order to create higher switching costs related to
their native ecosystems – the surest “upgrade” user is the one already locked into “your”
ecosystems, after all.
Page 40 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
5.3. Avenues for App Distribution
13. Some mobile apps are pre-loaded on mobile devices or set as default apps, while others
are only available through an app store, through a browser (web apps), or, for devices using
the Android system, by sideloading. Is there data comparing these mechanisms and their
effect on app distribution?
Pre-loaded Apps provide a significant advantage to the gatekeeper as many users will stick
with the default provided, however this is far less significant than the gatekeeper being able to
either entirely block Apps from third party providers, being able to extract additional fees by
blocking the ability to install Apps or being able to block the installation of Apps that compete
with their own Apps or services. Additionally we believe that removing barriers to
interoperability (for example via Web Apps) is a key method of reducing the ability of operating
system gatekeepers from engaging in this style of anti-competitive behaviour.
The EUs Digital Markets Acts latest revision noted:
"When gatekeepers operate and impose web browser engines, they are in a position to
determine the functionality and standards that will apply not only to their own web
browsers, but also to competing web browsers and, in turn, to web software
applications."
The UKs Competition and Market Authority have highlighted this accurately and extensively in
their mobile ecosystems market study interim report and write:
"As a result of the WebKit restriction, there is no competition in browser engines on iOS
and Apple effectively dictates the features that browsers on iOS can offer (to the extent
that they are governed by the browser engine as opposed to by the UI ."
“Importantly, due to the WebKit restriction, Apple makes decisions on whether to
support features not only for its own browser, but for all browsers on iOS. This not only
restricts competition (as it materially limits the potential for rival browsers to
differentiate themselves from Safari on factors such as speed and functionality) but
also limits the capability of all browsers on iOS devices, depriving iOS users of useful
innovations they might otherwise benefit from.
Apple's ban on third browser engines allows it to limit the functionality of all browsers and all
Web Apps (note browsers other than Safari on iOS and Chrome on Android can not even
effectively install Web Apps). We have listed a number of remedies that we believe would
restore competition both between browsers and between Web Apps and Native Apps.
Additionally this would lead to interoperable Apps that would work on both Android and iOS,
lowering consumer costs and reducing lock-in.
As such, while arguments could potentially be made for remedies to address pre-loaded apps
and services (i.e choice screens), in the context of browsers we have not included it, as in the
absence of allowing browser engine competition and other remedies, we do not believe this
would be effective.
Page 41 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
15. How do, or might, alternative app stores (other than Google Play or the Apple App Store),
affect competition in the mobile app ecosystem?
Allowing alternate app stores has significant benefits to end users. It greatly strengthens
competition by:
Reducing how much gatekeepers can charge for access to users.
Reducing the ability of gatekeepers to block apps that compete with their own.
Provides direct competition for the app stores themselves without forcing the user to
abandon the operating system (and all the apps they have purchased on it) and buy a
new phone with a different operating system, or harming their security.
If alternative app stores were available, then the default app store would have to compete with
the alternate app stores on:
Payment processing
Refunds
Escrow
Moderation and content policies
Discovery
Reviews
Developer experience
Cut of developer revenue
Much of gatekeepers' ability to add additional fees to third parties selling software to
consumers comes not from the value of the services they provide (listed above) but from their
stranglehold on App installation. Forcing them to compete would greatly limit this, thus in a
competitive market will lead to significant reduction in App prices for end consumers.
That said, while this does appear to have considerable advantages to users and significantly
reduces the ability of gatekeepers to add additional fees for access to users, this has a number
of potential limitations.
Depending on implementation Native Apps are likely still not interoperable, as they have to be
written in different programming languages and the system APIs are distinct. Users are likely to
Page 42 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
install only 1 or 2 app stores meaning the app stores will still wield considerable bargaining
power. With higher bargaining power they can extract more money leading to higher costs and
less high-quality apps for end users.
Web Apps have a number of distinct advantages. They are automatically fully interoperable.
The gatekeepers have no leverage to extract any money from them. They are significantly
cheaper to develop as there is only a single code base. For large companies (where
development budget is less of an issue) they can achieve higher quality as writing/maintaining
the same app multiple times in different programming languages and keeping it in sync across
multiple development teams is extremely hard.
15b. What unique barriers are there affecting each of the main operating systems Android,
iOS that might prevent web apps or—to the extent allowed on Android system—alternative
app stores and sideloading, from gaining more popularity with users and app developers
than they currently have?
Apple has effectively banned rival browsers from iOS and has made it impossible for third
parties to provide the capabilities Web Apps need in order to effectively compete with the iOS
App Store. Thus any browser engine functionality missing from iOS Safari is missing from all
browsers on iOS.
Apple faces little effective competitive pressure to improve the quality of their iOS Safari
browser and has incentives to inhibit it from competing with native. Thus Apple’s decade long
prohibition on competition for Safari on iOS has a compounding anti-competitive effect as
companies sink money into non-interoperable native iOS applications instead of Web Apps.
We write about this extensively in our paper “Bringing Competition to Walled Gardens”.
15c. Is there analysis comparing competition on iOS ecosystem (where app distribution is
limited) to that of alternative distribution mechanisms on Android operating systems?
While it is possible to install Native Apps on Android directly (rather than through the official
app store), it is not possible for users to replace the app store with one from another reputable
provider. As such users have to jump through a number of awkward scary security hoops to
install each application directly. Unsurprisingly adoption by end users is low (only 1.1% of
Android devices had at least one sideloaded app).
Web Apps have the advantage of being interoperable between Android and iOS but that
advantage is severely stunted if they are not viable on the other major platform iOS. The
previous question and our paper outline in detail our arguments on this.
16. What evidence is there to assess whether an app store model is necessary for mobile
devices, instead of the general-purpose model used for desktop computing applications?
The biggest argument against the app store model for mobile ecosystems is the power it allows
the gatekeeper to exert over users' devices after they have purchased them both in terms of
charging additional fees on every sale of third party software and in terms of blocking (either
legally or technically) software that competes with their software services (for example
Page 43 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
payment services, app stores, Safari/Webkit, Google Drive, iCloud, Apple Arcade). While there
are legitimate security threats that mobile users face, a strong case could be made these
limitations are primarily made to enrich the gatekeeper, not to benefit the end user.
Even Apple executives appear to be aware only their stranglehold on iOS installation is allowing
their 30% tax on revenue, something they can not achieve on Mac OS.
"Neither is on the store because they don't have to be. They can be on Mac and
distribute to users without sharing the revenue with us"
Philip Schiller Apple Upper Management On the Mac App Store
Web Apps are interoperable and run in the secure sandboxed browser environment. We believe
that they offer a strong and complementary alternative to the app store model. They are
significantly cheaper to write as there is only one code base across multiple operating systems.
Gatekeepers have no leverage to charge these apps additional fees for access to users and
they can not block apps that compete with their own services. Currently there are a number of
barriers that are blocking this from being viable. We discuss this extensively in Bringing
Competition to Walled Gardens” and propose a number of remedies.
It is clear from recently uncovered emails showing Apple's upper management's line of thinking
on iMessage that they do not see interoperability as desirable and would prefer where possible
to lock users into their platform:
“The #1 most difficult [reason] to leave the Apple universe app is iMessage ... iMessage
amounts to serious lock-in”
Unnamed Apple Employee
“iMessage on Android would simply serve to remove [an] obstacle to iPhone families
giving their kids Android phones ... moving iMessage to Android will hurt us more than
help us, this email illustrates why.
Craig Federighi Apple's Senior Vice President of Software Engineering
Page 44 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
17. Mobile app stores act as initial screeners and responders for concerns about mobile app
content, such as fraudulent apps and malware. 33 Similar issues for screening and
responding exist in other contexts, such as website hosting and search engine retrieval.
What empirical data is there analyzing any unique content screening issues related to
mobile app stores that affect competition?
NTIA is correct in noting that app stores combined multiple, traditionally disaggregated,
functions:
Discovery and search quality
Developer reputation management
Application binary delivery
Payment processing
Security screening
Post-hoc security response
In other environments and in traditional desktop operating systems, these functions are either
handled by different parties at various stages, or delegated to browser runtimes.
The concentrated power of mobile app store gatekeepers has created a culture of fear
regarding many aspects of the “mobile deal”. Startups or parties who object to terms of one
aspect of the system are hostage to the power wielded by gatekeepers, justified by other
bundled aspects.
This is played out repeatedly as leaks, objections, and blog posts documented app store
rejections for opaque reasons. But these are anecdotes and not data. NTIA should view the
lack of objective data not as a sign that many parties aren’t affected, but rather a display of the
incredible market power of the oligopolists. We are more than a decade into the consequences
of harsh chilling effects.
Nearly every technology business today relies on these gatekeepers continuing to agree to
carry their content, in stark contrast with the broad availability of services on the open,
interoperable web enjoyed by desktop users.
17a. Is there evidence of legitimate apps being rejected from app stores or otherwise
blocked from mobile devices? Is there evidence that this is a common occurrence or
happens to significant numbers of apps?
“Legitimate” is the key word in this question. Apple will view anyone who runs afoul of any wide
interpretation of its (unreviewable) decisions as being illegitimate. Others will note that these
Page 45 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
rules are applied capariously and without consistency and that many large parties enjoy
“sweetheart” deals in lieu of lawsuits between similarly armed legal departments.
OWA does not take a view as to which camp is right. We do think that app stores have a right to
carry what content they like and edit their inventories as they see fit. Where we agree with
Apple less comfortably is in support of Apple’s assertion that Web Apps should be a substitute
for Native Apps for developers who do not wish to be subject to Apple’s rules and regulations.
To make this true in practice, OWA believes that it is necessary to regulate true browser
competition into reality on mobile operating systems. Apple, Google, and Facebook have in
different ways precluded this possibility from becoming reality, and so there is no true option
outside of app stores for developers to fall back on. This harms developers and consumers.
17b. What assessments are there of their effectiveness, or lack therefore, on security and
privacy of end users?  34
App stores propose a model that OWA and browser vendors believe to be particularly weak
with regards to privacy and security.
Native App containers and sandboxing systems tend to provide significantly more capability to
developers without user consent beyond installation. This gap in consent creates greater
opportunities for tracking and the lack of choice in application runtimes reduces the ability of
the runtime to be more proactive in moderating the behaviour of developers.
Users on the web, however, can switch their browsers and install extensions to modify negative
behaviours of developers.
It is difficult to get an objective sense of app store effectiveness beyond these structural
factors, however we will note that malware producers are not particularly impressed with the
weak static analysis tools employed by Apple.
17c. Are there disincentives or unique barriers affecting the degree of security and privacy
protections offered by alternative app stores?
OWA does not have a strong view on the potential for multiple app stores.
Instead, we favour an outcome in which competing web browsers are empowered to offer
installation of Web Applications on fair and non-discriminatory terms. Today, this is not the
case. Apple limits the “Add to Homescreen” capability to Safari, and Google fails to provide
access to the WebAPK back-end to competitors.
Both parties deny competitors the ability to provide safe, portable, interoperable,
heavily-sandboxed Web Applications for easily installation via installable to their own browsers.
Opening this feature to competing default browsers may allow for safer alternatives for
application distribution than a “wild west” of competing Native App stores which also enjoy
access to the overpowered default capabilities of Native Apps.
Page 46 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
That is, we imagine a future where “alternative apps stores” need not introduce new or
expanded risks due to the insecure nature of Native Apps, but can instead be provided by
browsers directly on top of their more isolated, secure and privacy-preserving runtimes.
Even if NTIA agrees with gatekeepers that competing app stores are dangerous, we hope and
trust that NTIA will consider browsers to be a separate class, and acknowledge the privacy and
security benefits of expanding their role in delivering interoperable software.
19. How does the existence of imposter and other fraudulent apps affect developer
incentives or legitimate app lifecycles?
All developers must worry about how to keep their users secure from imposter/fraudulent
Apps.
Web Apps rely on the locked down and untrusting nature of the browser environment and
correct identification of the website the Web App is installed from using the domain name and
https.
Native Apps installed from the app store rely on the vetting process of the app store. Typically
this involves an automated code check and a brief click around by an employee of the app
store.
App stores (including the iOS App Store) have struggled to keep imposter/fraudulent Apps at
bay. Additionally their advertising has lulled users into a false sense of security leading the
users to fail to do basic checks such as looking at the App publishers name.
We believe that Web Apps provide a more secure and private mechanism of installing Apps on
mobile devices.
Page 47 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
5.4. App Users
21. How do most consumers find and make decisions to use apps? Is there data to show
whether the usage of an app or any other relevant metric for performance is tied to existing
brand visibility outside of the mobile app ecosystem? Is there data about how often people
use the search feature in an app store, search engines through browsers, or particular
ranking lists of popular apps or app storefronts? Is there empirical data that examines how
app rankings, app reviews, or other objective measures of apps (for example, popularity,
quality, or number of downloads) are used (or manipulated) to influence consumer choices?
It is very likely that the largest operating system gatekeepers Apple and Google keep detailed
aggregate data on whether an App is installed via a link to the app store from the Web (App
Clips, App Install Links, Smart App Banners) or via the users directly searching on the app
store.
For the purpose of this study it may be useful to compel Apple and Google to provide the NTIA
with this data. This should include any data they have that involves the interactions between
websites and the app store. Additionally third party ad campaign systems for Apps may have
useful data they could provide to the NTIA.
The amount of effort that Apple has put into developing App Clips and App Banners (while
refusing to implement the equivalent for Web Apps) suggests that App discovery via the Web is
material to the App Store’s business.
We have written extensively about Web App, App Banners and App Clips in our paper "Bringing
Competition to Walled Gardens".
22. The E.O. asks the Department to explore ways to maximize “user benefit” with regard to
competition in the mobile app ecosystem. How should we measure or consider user benefit?
22a. What is the appropriate scope of users for consideration? Should it include developers?
Developers should not be considered users in this context. Benefits to consumers should be
the primary consideration.
However, in a competitive market, many benefits to developers directly translate into benefits
for end users.
For example Web Apps have the potential to drastically reduce costs for multi-platform Apps
(i.e the developer only has one code-base and uses the browser environment as a consistent
platform). It has the potential to improve quality for large companies as keeping multiple
versions of the same App written in different programming languages is difficult and time
consuming. Additionally operating system gatekeepers are not able to charge Web Apps
1530% rent for the privilege of accessing users on their platform. While these are clearly
benefits to the developer they translate to higher quality and significantly lower cost Apps for
the end user.
Page 48 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
If an action benefits developers but on balance harms the interests of end users then in our
view that action should not be taken.
22b. If there are conflicts between end-user and developer interests, how does this affect
the assessment of user benefit?
In general with careful consideration of the context, end users' interests should prevail over
developers interests.
22c. How might convergence of end-users and developers—through low-code
environments, for example—affect this dynamic moving forward?
In this context, if those users are providing functionality to other users via low-code they are
developers.
23. Do apps that are developed for, or used by, certain communities (such as by income,
ethnicity/race, or gender) face significantly different competitive challenges? What are the
challenges?
Please see the answer to question 7.
Page 49 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
5.5. Other Factors
24. Some apps make use, or would like to make use, of additional mobile device components
beyond those that are more commonly accessible (
e.g.,
camera, microphone, contacts) in
order to offer an innovative product or service, but the operating system or device provider
does not allow such access. 36 Similarly, for some apps, it might be essential to be able to
interconnect to other hardware and services, such as cloud services. What are the valid
security concerns and technical limitations on what device functionality an app can access?
Gatekeepers of operating systems impose many rules on applications in order to protect user
security. Many of these rules are reasonable and in the users best interest. Additionally what is
necessary may change over time.
We are not arguing that there be no rules imposed by the operating system on applications that
run on it. Rather we are arguing that these rules be necessary, proportional, have strong
publicly available evidence and have the user’s interest in mind, not the gatekeepers.
24a. What factors should be considered in striking a balance between encouraging
companies to ensure proper security measures, while allowing third parties to access the
protected features that might allow for further innovation and competition?
The key point here is whether or not there is strong evidence for each security rule that:
It is narrow scoped and targeted
It is necessary
It could not be achieved by less onerous methods
The unmitigatable security harms prevented by this rule are significantly and
provably worse than the harm to competition caused by the rule
It is applied on a consistent basis to the gatekeepers own apps/services and their
business partners
It is not used as a tool to damage competition in favour of the gatekeepers own
apps/services and their business partners
Finally it is not the gatekeepers sole responsibility to protect users. There will be many
instances where providing low level system access to competitors with strong security track
records and dedicated security teams is massively in the consumers interest, even though this
means delegating protecting the user to these competitors. Major browser vendors are a prime
example of this in relation to features such as creation of apps Web Apps), process spawning
and JIT.
Page 50 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
We believe that the onus of proof should be on the gatekeeper to justify these rules and that
they must do so publicly and free of charge.
This style of language is already present in both the US's Open App Markets Act and the EU's
Digital Markets Act Section 4.
24b. Are there specific unnecessary (
e.g.,
technical) constraints placed on this ability of
app developers to make use of device capabilities, whether by device-makers, service
providers or operating system providers, that impact competition?
The Apple App Store Review Guidelines contain the following rule:
“2.5.6 Apps that browse the web must use the appropriate WebKit framework and
WebKit Javascript.
Webkit is the engine that powers Safari and several Linux browsers. Apple also produces a
“WebKit framework” that is included in its operating systems (macOS, iOS, iPadOS, tvOS, and
watchOS.
In practice, Section 2.5.6 is a requirement that iOS and iPadOS browsers from Google,
Microsoft, Mozilla, Samsung, Opera cannot use their own engines the way they do everywhere
else. These engines take hundreds of thousands of engineer hours to develop, and are
excluded from Apple’s most successful consumer operating systems. Competing browser
vendors are only allowed to produce shells around a very specific, unaltered version of Safari’s
WebView; a component whose features Apple dictates.
This means that Apple can dictate the feature set of all browsers on iOS and importantly the
feature set of all Web Apps on iOS.
In the remedies section we have included a list of legal and technical constraints that we
believe need be removed, and a number of obligations that need to be imposed on operating
system gatekeepers to restore competition to the mobile app market. Web Apps are an
interoperable and rent-free competitor to the app stores that were it allowed to compete would
provide significant benefits to consumers and important competitive pressure.
These remedies all have a consistent theme forcing the gatekeeper to allow sufficient access
to the operating system to competitors that they can provide services that replace and
compete with those provided by the gatekeeper and its app store. In particular allowing both
rival browsers and their Web Apps a chance to compete with the gatekeepers browser and app
store. This access can be constrained by necessary and evidence based rules for
security/privacy/legal reasons as described in 24.a).
Page 51 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
24c. Are there other means or factors to consider for mitigating specific risks that would not
inhibit competition?
As described in 24.a) there are many such rules that can mitigate common security risks. The
key point is they must be justified with strong evidence to prevent gaming by gatekeepers
aiming to prevent competition with their own apps and services on platforms they control.
25. What unique challenges, if any, do software updates pose for app competition, including
updates driven by the app developers and those necessitated by other ecosystem changes,
such as operating system updates? How does this impact security and costs for those apps,
products, and services?
Web Browsers on traditional operating systems and websites both enjoy high paces of
software delivery versus native mobile apps today.
Based on the lessons learned from decades of remediating security issues, modern web
browsers have moved to a “silent auto-update” model in which users frequently encounter new
versions of the app simply by restarting it. This enables low latency for browser vendors to
deliver fixes for security issues and is one of the key mechanisms underpinning modern
cybersecurity (small “patch gaps”).
Web developers likewise have enjoyed the ability to update their content quickly and without
the intervention of any gatekeepers. This has spawned industry-wide disciplines of A/B testing
on live traffic, allowing the entire world of Web Apps to be more agile and lower-risk for
developers than traditional software development.
In both regards, mobile operating systems today represent a regression in security and
productivity versus the desktop operating systems status quo.
It is important to note here that browser updates on iOS for all browsers are tied to iOS system
updates as each iOS update brings with it a specific version of Webkit (the engine that powers
iOS Safari) that all browsers on iOS must use without modification. This is an antiquated
practice that all other browsers have abandoned due to the issues it causes in delaying users
from receiving vital security patches and bug fixes. This means to update the browser, users
have to update the entire operating system.
iOS users remain vulnerable to known bugs in Safari longer than users of alternative browsers
on every other operating system. This picture is made even darker by operating systems
update rates. Since Safari requires a full operating system update, further hassle (and
attendant delay) is introduced in getting patches into user’s hands than if the browser updated
like a “normal app” (the standard on all other modern operating systems). Safari requires the
user to update their entire operating system, a process that makes the device unusable for
up-to 20 minutes.
According to the metrics regarding bugs filed by Google’s Project Zero team, Safari is the
slowest major engine to fix issues and is significantly slower than others as delivering fixes
through software updates.
Page 52 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
We have written about the security problem this causes in our paper “Bringing Competition to
Walled Gardens” in section “8.2 Security”.
Page 53 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
5.6. Potential Actions To Increase Competition
27. What specific measures might the federal government take to foster healthy
competition—especially for nascent app innovation—in the mobile app ecosystem?
In the remedies section we have outlined the remedies in priority order that OWA believe are
necessary to restore meaningful competition.
We have outlined a brief ratitional for each of the remedies. For a far more in depth and
detailed discussion please see our paper "Bringing Competition to Walled Gardens".
28. What specific actions could the private sector and civil society take to ensure and
promote healthy app competition (such as technical standards development or
monitoring)?  (37
Creating software that is interoperable between devices is incredibly important for both
lowering the costs of software development and increasing competition between device
manufacturers by allowing software to automatically work on all devices. The Web currently
stands as the only viable cross-platform application platform.
Much work has been done on improving the capabilities of Web Apps in the last 5 years. Please
see the section “The Web Is Capable” from our paper “Bringing Competition to Walled Gardens”.
There is significant demand for this functionality but in the mobile ecosystem it is being held
back by the anti-competitive practices of a few companies. These are gatekeepers who are
incentivized to block competition and innovation, not to benefit users but rather to allow them
to use their control of the operating system to extract revenue from the software layer.
Were these anti-competitive forces to be removed we believe that the Web and all of the
companies involved in it (browser vendors, framework developers, game engines, the wider
software community) would invest significantly in improving the Web's ability to provide high
quality interoperable Web Apps to consumers.
Page 54 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
6. Brighter Future for Mobile Ecosystems
We would like to thank the NDIA again for considering these important issues.
Mobile ecosystem app stores wield a lot of power and lock-in over consumers. This allows
them to both block competitors and charge consumers additional fees to install Apps from
competitors.
Web Apps could be an interoperable source of competition for app stores however they are not
currently viable on iOS for a number of reasons we have outlined in our answers and our paper.
We have proposed a number of remedies that we believe are vital to restore competition and
user choice to the mobile ecosystem.
We believe that gatekeepers of mobile hardware/operating systems should compete to offer
additional services and software to customers on merit and user choice, not by blocking
applications that compete with their own or by charging consumers additional fees on
applications/services from competitors.
US regulators have a chance to blunt the anti-competitive power of incumbent gatekeepers
and create a level and competitive playing field for the future of mobile App development.
These changes will lead to the following benefits to consumers:
More competition
Interoperable Apps
Higher quality Apps
Lower cost Apps
Less lock-in into particular mobile operating systems
Competition not walled gardens leads to the best outcomes for consumers
Page 55 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
7. References
Definition of Web Apps
https://developer.mozilla.org/en-US/docs/Web/Progressive_web_apps
W3 Web Standards) Patent Policy
https://www.w3.org/Consortium/Patent-Policy-20200915/
Commercial users now spend greater than 60% of their time within a browser
https://www.microsoft.com/en-us/microsoft-365/blog/2020/08/04/revisit-idea-modern-brows
er/
Apple claiming Web Apps are a viable alternative to the iOS App Store
https://www.accc.gov.au/system/files/Apple%20Pty%20Limited%20%2810%20February%2020
21%29.pdf
Tim Cook claiming to congress that Web Apps are a viable alternative to the iOS App Store
https://www.youtube.com/watch?v=H6eYLCxxQdA&t=306s
Apple vs Epic - claiming to congress that Web Apps are a viable alternative to the iOS App
Store
https://9to5mac.com/2021/03/25/bypass-the-app-store-says-apple/
Latest version of the EUs Digital Markets Act
https://www.consilium.europa.eu/media/56086/st08722-xx22.pdf
UK Competition and Markets Authority  Mobile Ecosystems Market Study Interim Report
https://www.gov.uk/government/publications/mobile-ecosystems-market-study-interim-report
/interim-report
Apple collected $72.3 billion USD in App Store fees in 2020
https://appleinsider.com/articles/21/01/05/app-store-earns-723-billion-in-2020-almost-double
-google-play-revenues
iOS App Store has a nearly 80% profit margin
https://www.marketwatch.com/story/how-profitable-is-apples-app-store-even-a-landmark-ant
itrust-trial-couldnt-tell-us-11622224506
Typical profit margin in competitive markets
https://www.brex.com/blog/what-is-a-good-profit-margin
Alex Russell  Respecting Default Browser Choice
https://infrequently.org/2021/07/hobsons-browser/
Definition of JIT Just in time compilation)
https://en.wikipedia.org/wiki/Just-in-time_compilation
Apples policy of removing apps that Apple has viewed as “duplicating functionality” of iOS
https://en.wikipedia.org/wiki/IOS_app_approvals#Functional_restrictions
Page 56 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
iOS  browser entitlement
https://developer.apple.com/documentation/xcode/preparing-your-app-to-be-the-default-bro
wser-or-email-client
Photoshop ported to the Web
https://web.dev/ps-on-the-web/
Apple Robbed the Mob’s Bank Part 1
https://mobiledevmemo.com/apple-robbed-the-mobs-bank/
Apple Robbed the Mob’s Bank Part 2
https://mobiledevmemo.com/apple-robbed-the-mobs-bank-part-2/
React Native has unfixed 1900 accessibility issues
https://github.com/facebook/react-native/projects/15
iOS users on average spend more than Android users
https://www.entrepreneurshiplife.com/why-iphone-users-spend-more-money-than-android-us
ers/
Benedict Evans  Technology Writer  On Apple banning Apps they don't like
https://www.ben-evans.com/benedictevans/2020/8/18/app-stores
Dieter Bohn  The Verge  Apple's self dealing App Store policies
https://www.theverge.com/2020/6/17/21293813/apple-app-store-policies-hey-30-percent-dev
elopers-the-trial-by-franz-kafka
Apple ban entire categories of App
https://www.theverge.com/2020/9/18/20912689/apple-cloud-gaming-streaming-xcloud-stadia
-app-store-guidelines-rules
Apple ban Apps that compete with their own offerings
https://www.theverge.com/2021/9/16/22676706/apple-watch-swipe-keyboard-flicktype-lawsui
t-kosta-eleftheriou
Apple App Store Review is chaotic, irrational and stressful
https://www.ben-evans.com/benedictevans/2020/8/18/app-stores
Apple App Store Review - an unfriendly behemoth of a system
https://twitter.com/samj0hn/status/1431001795904561160
Apple struggles to stop malware on the iOS App Store
https://habr.com/en/post/580272/
iOS App Store is littered with malicious apps
https://www.xanjero.com/news/developer-kosta-eleftheriou-claims-apples-app-store-is-littere
d-with-malicious-apps/
Alex Russell  Browser Choice Must Matter
https://infrequently.org/series/browser-choice-must-matter/
Page 57 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
Only 1.1% of Android devices had at least one sideloaded app
https://www.techtarget.com/searchmobilecomputing/opinion/What-did-2019-see-for-mobile-s
ecurity-More-Punycode-phishing-and-jailbreaking-returns
Philip Schiller Apple Upper Management On the Mac App Store
https://applescoop.org/story/apple-execs-discuss-why-the-mac-app-store-has-not-been-suc
cessful-in-internal-email
Apple Management on why imessage is not interoperable with android
https://www.theverge.com/2021/4/9/22375128/apple-imessage-android-ecosystem-lock-in-ep
ic-games-filings-app-store-dispute
Apple vs Hey Monopoly Power)
https://www.hey.com/apple/
iOS App Store has a scam app problem
https://www.theverge.com/2021/4/21/22385859/apple-app-store-scams-fraud-review-enforce
ment-top-grossing-kosta-eleftheriou
Process Spawning definition
https://en.wikipedia.org/wiki/Spawn_(computing)
H.R.5017  Open App Markets Act
https://www.congress.gov/bill/117th-congress/house-bill/5017/text#:~:text=SEC.%204.%20PRO
TECTING%20THE%20SECURITY%20AND%20PRIVACY%20OF%20USERS.
iOS App Store  Rule banning third party browser engines
https://developer.apple.com/app-store/review/guidelines/#:~:text=2.5.6%20Apps%20that%20b
rowse%20the%20web%20must%20use%20the%20appropriate%20WebKit%20framework%20a
nd%20WebKit%20Javascript.
Safari is the slowest major engine to fix security issues
https://googleprojectzero.blogspot.com/2022/02/a-walk-through-project-zero-metrics.html
Page 58 of 59
Competition in the Mobile Ecosystem  Browsers & Web Apps Response to NTIA
Version 1.0
8. Open Web Advocacy
Open Web Advocacy is a loose group of software engineers from all over the world, who work
for many different companies who have come together to fight for the future of the open web
by providing regulators, legislators and policy makers the intricate technical details that they
need to understand the major anti-competitive issues in our industry and potential ways to
solve them.
It should be noted that all the authors and reviewers of this document are software engineers
and not economists, lawyers or regulatory experts. The aim is to explain the current situation,
outline the specific problems, how this affects consumers and suggest potential regulatory
remedies.
This is a grassroots effort by software engineers as individuals and not on behalf of their
employers or any of the browser vendors.
We are available to regulators, legislators and policy makers for presentations/Q&A and we can
provide expert technical analysis on topics in this area.
For those who would like to help or join us in fighting for a free and open future for the web,
please contact us at:
Email
contactus@open-web-advocacy.org
Twitter
OpenWebAdvocacy
Web
https://open-web-advocacy.org
Page 59 of 59