Generic Sensor API

1. Introduction

Increasingly, sensor data is used in application development to enable new use cases such as geolocation, counting steps or head-tracking. This is especially true on mobile devices where new sensors are added regularly.

Exposing sensor data to the Web has so far been both slow-paced and ad-hoc. Few sensors are already exposed to the Web. When they are, it is often in ways that limit their possible use cases (for example by exposing abstractions that are too high-level and which don’t perform well enough). APIs also vary greatly from one sensor to the next which increases the cognitive burden of Web application developers and slows development.

The goal of the Generic Sensor API is to promote consistency across sensor APIs, enable advanced use cases thanks to performant low-level APIs, and increase the pace at which new sensors can be exposed to the Web by simplifying the specification and implementation processes.

This lacks an informative section with examples for developers. Should contain different use of the API, including using it in conjunction with requestAnimationFrame.

2. Scope

This section is non-normative.

The scope of this specification is currently limited to specifying primitives which enable expose data from local sensors.

Exposing remote sensors or sensors found on personal area networks (e.g. Bluetooth) is out of scope. As work in these areas mature, it is possible that common, lower-level primitives be found, in which case this specification will be updated accordingly. This should have little to no effects on implementations, however.

This specification also does not currently expose a sensor discovery API. This is because the limited number of sensors currently available to user agents does not warrant such an API. Using feature detection, such as described in §3 A note on Feature Detection of Hardware Features, is good enough for now. A subsequent version of this specification might specify such an API, and the current API has been designed with this in mind.

3. A note on Feature Detection of Hardware Features

This section is non-normative.

Feature detection is an established Web development best practice. Resources on the topic are plentiful on and offline and the purpose of this section is not to discuss it further, but rather to put it in the context of detecting hardware-dependent features.

Consider the below feature detection examples:

if (typeof Gyroscope === "function") {
    // run in circles...
}

if ("ProximitySensor" in window) {
    // watch out!
}

if (window.AmbientLightSensor) {
    // go dark...
}

// etc.

All of these tell you something about the presence and possible characteristics of an API. They do not tell you anything, however, about whether that API is actually connected to a real hardware sensor, whether that sensor works, if its still connected, or even whether the user is going to allow you to access it. Note you can check the latter using the Permissions API [PERMISSIONS].

In an ideal world, information about the underlying status would be available upfront. The problem with this is twofold. First, getting this information out of the hardware is costly, in both performance and battery time, and would sit in the critical path. Secondly, the status of the underlying hardware can evolve over time. The user can revoke permission, the connection to the sensor be severed, the operating system may decide to limit sensor usage below a certain battery threshold, etc.

Therefore, an effective strategy is to combine feature detection, which checks whether an API for the sought-after sensor actually exists, and defensive programming which includes:

1. checking for error thrown when instantiating a Sensor object,

2. listening to errors emitted by it,

3. handling all of the above graciously so that the user’s experience is enhanced by the possible usage of a sensor, not degraded by its absence.

try { // No need to feature detect thanks to try..catch block.
    var sensor = new GeolocationSensor();
    sensor.start();
    sensor.onerror = error => gracefullyDegrade(error);
    sensor.onreading = _ => updatePosition(sensor.latitude, sensor.longitude);
} catch(error) {
    gracefullyDegrade(error);
}

4. Security and privacy considerations

Sensor readings are sensitive data and could become a subject of various attacks from malicious Web pages. Before discussing the mitigation strategies we briefly enumerate the main types of the sensor's privacy and security threats. The [MOBILESENSORS] categorizes main threats into location tracking, eavesdropping, keystroke monitoring, device fingerprinting, and user identification. This categorization is a good fit for this specification.

The risk of successful attack can increase when sensors are used with each other, in combination with other functionality, or when used over time, specifically with the risk of correlation of data and user identification through fingerprinting. Web application developers using these JavaScript APIs should consider how this information might be correlated with other information and the privacy risks that might be created. The potential risks of collection of such data over a longer period of time should also be considered.

Variations in sensor readings as well as event firing rates offer the possibility of fingerprinting to identify users. User agents may reduce the risk by limiting event rates available to web application developers.

Note: do we really want this mitigation strategy?

Minimizing the accuracy of a sensor’s readout generally decreases the risk of fingerprinting. User agents should not provide unnecessarily verbose readouts of sensors data. Each sensor type should be assessed individually.

If the same JavaScript code using the API can be used simultaneously in different window contexts on the same device it may be possible for that code to correlate the user across those two contexts, creating unanticipated tracking mechanisms.

User agents should consider providing the user an indication of when the sensor is used and allowing the user to disable it. Additionally, user agents may consider allowing the user to verify past and current sensor use patterns.

Web application developers that use sensors should perform a privacy impact assessment of their application taking all aspects of their application into consideration.

Ability to detect a full working set of sensors on a device can form an identifier and could be used for fingerprinting.

A combination of selected sensors can potentially be used to form an out of band communication channel between devices.

Sensors can potentially be used in cross-device linking and tracking of a user.

4.1. Types of privacy and security threats

This section is non-normative.

4.1.1. Location Tracking

Under this type of threat, the attacks use sensor readings to locate the device without using GPS or any other location sensors. For example, accelerometer data can be used to infer the location of smartphones by using statistical models to obtain estimated trajectory, then map matching algorithms can be used to obtain predicted location points (within a 200-m radius)[MOBILESENSORS].

4.1.2. Eavesdropping

Recovering speech from gyroscope readings is an example of eavesdropping attack. See [GYROSPEECHRECOGNITION].

4.1.3. Keystroke Monitoring

Many user inputs can be inferred from sensor readings, this includes a wide range of attacks on user PINs, passwords, and lock patterns (and even touch actions such as click, scroll, and zoom) using motion sensors. These attacks normally train a machine learning algorithm to discover such information about the users. See [STEALINGPINSVIASENSORS].

4.1.4. Device Fingerprinting

Sensors can provide information that can uniquely identify the device using those sensors. Every concrete sensor model has minor manufacturing imperfections and differences that will be unique for this model. These manufacturing variations and imperfections can be used to fingerprint the device [MOBILESENSORS].

4.1.5. User Identifying

Sensor readings can be used to identify the user, for example via inferring individual walking patterns from smartphone or wearable device motion sensors' data.

4.2. Mitigation Strategies

This section is non-normative.

This section gives a high-level presentation of some of the mitigation strategies specified in the normative sections of this specification.

4.2.1. Secure Context

Sensor readings are explicitly flagged by the Secure Contexts specification [POWERFUL-FEATURES] as a high-value target for network attackers. Thus all interfaces defined by this specification or extension specifications are only available within a secure context.

4.2.2. Top-Level Browsing Context

To avoid the privacy risk of sharing sensor readings with contexts unfamiliar to the user, sensor readings are only available in the top-level browsing context.

Note: [FEATURE-POLICY] will allow securely relaxing those restrictions once it matures.

4.2.3. Losing Focus

When the top-level browsing context loses focus, or when a nested browsing context of a different origin gains focus (for example when the user carries out an in-game purchase using a third party payment service from within an iframe) the top-level browsing contexts suddenly becomes in a position to carry out a skimming attack against the browsing context that has gained focus.

To mitigate this threat, readings of sensors running in a top-level browsing contexts are not delivered in such cases. A security check is run before sensor readings are delivered to ensure that.

4.2.4. Visibility State

Sensor readings are only available in browsing contexts that are visible to the user, that is, whose visibility state is "visible". A security check is run before sensor readings are delivered to ensure that.

certain use cases require sensors to have background access. Using a more complex PermissionDescriptor. (e.g. with a boolean allowBackgroundUsage = false; dictionary member), might be the solution to relax this restriction.

4.2.5. Permissions API

Access to sensor readings are controlled by the Permissions API [PERMISSIONS]. User agents use a number of criteria to grant access to the readings. Note that access can be granted without prompting the user.

4.3. Mitigation strategies applied on a case by case basis

Each sensor type will need to be assessed individually, taking into account the use cases it enables and its particular threat profile. While some of the below mitigation strategies are effective for certain sensors, they might also hinder or altogether prevent certain use cases.

Note: These mitigation strategies can be applied constantly or temporarily, for example when the user is carrying out specific actions, when other APIs which are known to amplify the level of the threat are in use, etc.

4.3.1. Limit maximum sampling frequency

User agents may mitigate certain threats by limiting the maximum sampling frequency. What upper limit to choose depends on the sensor type, the kind of threats the user agent is trying to protect against, the expected resources of the attacker, etc.

Limiting the maximum sampling frequency prevents use cases which rely on low latency or high data density.

4.3.2. Stop the sensor altogether

This is obviously a last-resort solution, but it can be extremely effective if it’s temporal, for example to prevent password skimming attempts when the user is entering credentials on a different origin ([rfc6454]) or in a different application.

4.3.3. Limit number of delivered readings

An alternative to limiting the maximum sampling frequency is to limit the number of sensor readings delivered to Web application developer, regardless of what frequency the sensor is polled at. This allows use cases which have low latency requirement to increase sampling frequency without increasing the amount of data provided.

Discarding intermediary readings prevents certain use cases, such as those relying on certain kinds of filters.

4.3.4. Reduce accuracy

Reducing the accuracy of sensor readings or sensor reading timestamps might also help mitigate certain threats, thus user agents should not provide unnecessarily verbose readouts of sensors data.

However, certain use cases require highly accurate readings, especially when operations carried out on the readings, or time deltas calculated from the timestamps, increase inaccuracies exponentially.

Note: while adding random bias to sensor readings has similar effects, it shouldn’t be used in practice as it is easy to filter out the added noise.

4.3.5. Keep the user informed about API use

User agents may choose to keep the user informed about current and past use of the API.

Note: this does not imply keeping a log of the actual sensor readings which would have issues of its own.

5. Concepts

5.1. Sensors

The term device sensor refers to a device’s underlying physical sensor instance.

The device sensors measure different physical quantities and provide corresponding raw sensor readings which are a source of information about the user and their environment.

Each reading is composed of the values of the different physical quantities measured by the sensor at time t_n.

Known, predictable discrepancies between raw sensor readings and the corresponding physical quantities being measured are corrected through calibration.

Known but unpredictable discrepancies need to be addressed dynamically through a process called sensor fusion.

Calibrated raw sensor readings are referred to as sensor readings, whether or not they have undergone sensor fusion.

The term platform sensor refers to platform interfaces, with which the user agent ineracts to obtain sensor readings for a single sensor type originated from one or several device sensors.

Platform sensor can be defined by the underlying platform (e.g. in a native sensors framework) or by the user agent, if it has a direct access to device sensor.

In simple case a platform sensor corresponds to a single device sensor, but if the provided sensor readings are product of sensor fusion performed in software, the platform sensor corresponds to a set of device sensors involved in the sensor fusion process.

Note: platform sensors created through sensor fusion are sometimes called virtual or synthetic sensors. However, the specification doesn’t make any practical distinction between them.

5.2. Sensor Types

Different sensor types measure different physical quantities such as temperature, air pressure, heart-rate, or luminosity.

For the purpose of this specification we distinguish between high-level and low-level sensor types.

Sensor types which are characterized by their implementation are referred to as low-level sensors. For example a Gyroscope is a low-level sensor type.

Sensors named after their readings, regardless of the implementation, are said to be high-level sensors. For instance, geolocation sensors provide information about the user’s location, but the precise means by which this data is obtained is purposefully left opaque (it could come from a GPS chip, network cell triangulation, wifi networks, etc. or any combination of the above) and depends on various, implementation-specific heuristics. High-level sensors are generally the fruits of applying algorithms to low-level sensors—for example, a pedometer can be built using only the output of a gyroscope—or of sensor fusion.

That said, the distinction between high-level and low-level sensor types is somewhat arbitrary and the line between the two is often blurred. For instance, a barometer, which measures air pressure, would be considered low-level for most common purposes, even though it is the product of the sensor fusion of resistive piezo-electric pressure and temperature sensors. Exposing the sensors that compose it would serve no practical purpose; who cares about the temperature of a piezo-electric sensor? A pressure-altimeter would probably fall in the same category, while a nondescript altimeter—which could get its data from either a barometer or a GPS signal—would clearly be categorized as a high-level sensor type.

Because the distinction is somewhat blurry, extensions to this specification (see §9 Extensibility) are encouraged to provide domain-specific definitions of high-level and low-level sensors for the given sensor types they are targeting.

Sensor readings from different sensor types can be combined together through a process called sensor fusion. This process provides higher-level or more accurate data (often at the cost of increased latency). For example, the readings of a three-axis magnetometer needs to be combined with the readings of an accelerometer to provide a correct bearing.

Smart sensors and sensor hubs have built-in compute resources which allow them to carry out calibration and sensor fusion at the hardware level, freeing up CPU resources and lowering battery consumption in the process.

But sensor fusion can also be carried out in software. This is particularly useful when performance requirements can only be met by relying on application-specific data. For example, head tracking for virtual or augmented reality applications requires extremely low latency to avoid causing motion sickness. That low-latency is best provided by using the raw output of a gyroscope and waiting for quick rotational movements of the head to compensate for drift.

5.3. Default sensor

The Generic Sensor API is designed to make the most common use cases straightforward while still enabling more complex use cases.

Most of devices deployed today do not carry more than one device sensor providing sensor readings of the same type. The use cases which require a set of similar device sensors are rare and generally limited to specific sensor types, such as multiple accelerometers in 2-in-1 laptops.

The API therefore makes it easy to interact with the device’s default (and often unique) sensor for each type simply by instantiating the corresponding Sensor subclass.

Indeed, without specific information identifying a particular sensor of a given type, the default sensor is chosen.

let sensor = new Accelerometer({ frequency: 30 });

sensor.onreading = () => { ... }
sensor.start();

Note: extension to this specification may choose not to define a default sensor when doing so wouldn’t make sense. For example, it does not make sense to explicitly define a default sensor for geolocation sensor type as the implementation of its interface can use multiple backends.

In cases where multiple device sensors corresponding to the same type may coexist on the same device, specification extension will have to define ways to uniquely identify each one.

var sensor = new DirectTirePressureSensor({ position: "rear", side: "left" });
sensor.onreading = _ => console.log(sensor.pressure);
sensor.start();

5.4. Sampling Frequency

For the purpose of this specification, sampling frequency for a platform sensor is defined as a frequency at which the user agent obtains sensor readings from the underlying platform.

5.5. Reading change threshold

A platform sensor reports readings to the user agent considering the reading change threshold.

The reading change threshold refers to a value which indicates whether or not the changes in the device sensor's measurements were significant enough to update the corresponding sensor readings.

The threshold value depends on the surrounding software and hardware environment constraints, e.g., software power consumption optimizations or the underlying device sensor's accuracy.

6. Model

A diagram would really help here.

6.1. Sensor Type

A sensor type has an associated interface whose inherited interfaces contains Sensor.

A sensor type has a set of associated sensors.

If a sensor type has more than one platform sensor, it must have a set of associated identifying parameters to select the right platform sensor to associate to each new Sensor objects.

A sensor type may have a default sensor.

A sensor type has an associated PermissionName.

Note: multiple sensor types may share the same PermissionName.

A sensor type has a permission revocation algorithm.

6.2. Sensor

A platform sensor has an associated set of activated sensor objects. This set is initially empty.

The current browsing context's platform sensor has an associated latest reading map which holds the latest available sensor readings.

Note: User agents can share sensor readings map between different contexts only if the origins of these contexts' active documents are same origin-domain.

Any time the user agent obtains a new sensor reading for a platform sensor from the underlying platform, it invokes update latest reading with the platform sensor and the sensor reading as arguments.

The latest reading map contains an entry whose key is "timestamp" and whose value is a high resolution timestamp of the time at which the latest reading was obtained expressed in milliseconds that passed since the time origin. latest reading["timestamp"] is initially set to null, unless the latest reading map caches a previous reading.

The other entries of the latest reading map hold the values of the different quantities measured by the platform sensor. The keys of these entries must match the attribute identifier defined by the sensor type's associated interface. The return value of the attribute getter is easily obtained by invoking get value from latest reading with the object implementing the sensor type's associated interface and the attribute identifier as arguments.

The value of all latest reading entries is initially set to null.

A platform sensor has an associated current sampling frequency which is initially null.

For a nonempty set of activated sensor objects the current sampling frequency is equal to optimal sampling frequency, which is estimated by the UA taking into account desired sampling frequencies of activated Sensors and sampling frequency bounds defined by the underlying platform.

Note: For example, the UA may estimate optimal sampling frequency as a Least Common Denominator (LCD) for a set of desired sampling frequencies capped by sampling frequency bounds defined by the underlying platform.

7. API

7.1. The Sensor Interface

[SecureContext, Exposed=Window]
interface Sensor : EventTarget {
  readonly attribute boolean activated;
  readonly attribute boolean hasReading;
  readonly attribute DOMHighResTimeStamp? timestamp;
  void start();
  void stop();
  attribute EventHandler onreading;
  attribute EventHandler onactivate;
  attribute EventHandler onerror;
};

dictionary SensorOptions {
  double? frequency;
};

A Sensor object has an associated platform sensor.

The task source for the tasks mentioned in this specification is the sensor task source.

let acl = new Accelerometer({frequency: 30});
let max_magnitude = 0;
acl.addEventListener('activate', () => console.log('Ready to measure.'));
acl.addEventListener('error', error => console.log('Error: ' + error.name));
acl.addEventListener('reading', () => {
    let magnitude = Math.hypot(acl.x, acl.y, acl.z);
    if (magnitude > max_magnitude) {
        max_magnitude = magnitude;
        console.log(`Max magnitude: ${max_magnitude} m/s2`);
    }
});
acl.start();

7.1.1. Sensor lifecycle

When a Sensor object is eligible for garbage collection, the user agent must invoke deactivate a sensor object with this object as argument.

7.1.2. Sensor internal slots

Instances of Sensor are created with the internal slots described in the following table:

7.1.3. Sensor.activated

7.1.4. Sensor.hasReading

7.1.5. Sensor.timestamp

The getter of the timestamp attribute returns the result of invoking get value from latest reading with this and "timestamp" as arguments.

7.1.6. Sensor.start()

7.1.7. Sensor.stop()

7.1.8. Sensor.onreading

onreading is an EventHandler which is called to notify that new reading is available.

7.1.9. Sensor.onactivate

onactivate is an EventHandler which is called when this.[[state]] transitions from "activating" to "activated".

7.1.10. Sensor.onerror

onerror is an EventHandler which is called whenever an exception cannot be handled synchronously.

7.1.11. Event handlers

The following are the event handlers (and their corresponding event handler event types) that must be supported as attributes by the objects implementing the Sensor interface:

7.2. The SensorErrorEvent Interface

[Constructor(DOMString type, SensorErrorEventInit errorEventInitDict),
 SecureContext, Exposed=Window]
interface SensorErrorEvent : Event {
  readonly attribute Error error;
};

dictionary SensorErrorEventInit : EventInit {
  required Error error;
};

7.2.1. SensorErrorEvent.error

Gets the Error object passed to SensorErrorEventInit.

8. Abstract Operations

8.1. Construct sensor object

8.2. Connect to sensor

8.3. Activate a sensor object

8.4. Deactivate a sensor object

8.5. Revoke sensor permission

8.6. Set sensor settings

8.7. Update latest reading

8.8. Security check

Note: user agents are encouraged to stop sensor sampling when security check would return "insecure" in order to reduce resource consumption, notably battery usage.

8.9. Find the reporting frequency of a sensor object

8.10. Report latest reading updated

8.11. Notify new reading

8.12. Notify activated state

8.13. Notify error

8.14. Get value from latest reading

8.15. Request sensor access

9. Extensibility

This section is non-normative.

Its purpose is to describe how this specification can be extended to specify APIs for different sensor types.

Extension specifications are encouraged to focus on a single sensor type, exposing both high and low level as appropriate.

9.1. Security and Privacy

Extension specifications are expected to:

• conform with the generic mitigation strategies,

• consider mitigation strategies applied on a case by case basis,

• be evaluated against the Self-Review Questionnaire on Security and Privacy [SECURITY-PRIVACY-QUESTIONNAIRE],

• and in particular, be evaluated against the same-origin policy violations that can arise if sensors expose a new communication channel not governed by the same-origin policy.

9.2. Naming

Sensor interfaces for low-level sensors should be named after their associated platform sensor. So for example, the interface associated with a gyroscope should be simply named Gyroscope. Sensor interfaces for high-level sensors should be named by combining the physical quantity the platform sensor measures with the "Sensor" suffix. For example, a platform sensor measuring the distance at which an object is from it may see its associated interface called ProximitySensor.

Attributes of the Sensor subclass that hold sensor readings values should be named after the full name of these values. For example, the Thermometer interface should hold the sensor reading's value in a temperature attribute (and not a value or temp attribute). A good starting point for naming are the Quantities, Units, Dimensions and Data Types Ontologies [QUDT].

9.3. Unit

Extension specification must specify the unit of sensor readings.

As per the Technical Architecture Group’s (TAG) API Design Principles [API-DESIGN-PRINCIPLES], all time measurement should be in milliseconds. All other units should be specified using, in order of preference, and with the exception of temperature (for which Celsius should be favored over Kelvin), the International System of Units (SI), SI derived units, and Non-SI units accepted for use with the SI, as described in the SI Brochure [SI].

9.4. Exposing High-Level vs. Low-Level Sensors

So far, specifications exposing sensors to the Web platform have focused on high-level sensors APIs. [GEOLOCATION-API] [ORIENTATION-EVENT]

This was a reasonable approach for a number of reasons. Indeed, high-level sensors:

• convey developer intent clearly,

• do not require intimate knowledge of how the underlying hardware sensors functions,

• are easy to use,

• may enable the User Agent to make significant performance and battery life improvements,

• help avoid certain privacy and security issues by decreasing the amount and type of information exposed.

However, an increasing number of use cases such as virtual and augmented reality require low-level access to sensors, most notably for performance reasons.

Providing low-level access enables Web application developers to leverage domain-specific constraints and design more performant systems.

Following the precepts of the Extensible Web Manifesto [EXTENNNNSIBLE], extension specifications should focus primarily on exposing low-level sensor APIs, but should also expose high-level APIs when they are clear benefits in doing so.

9.5. When is Enabling Multiple Sensors of the Same Type Not the Right Choice?

TODO: provide guidance on when to:

• allow multiple sensors of the same type to be instantiated,

• create different interfaces that inherit from Sensor,

• add constructor parameters to tweak sensors settings (e.g. setting required accuracy).

9.6. Definition Requirements

The following definitions must be specified for each sensor type in extension specifications:

• An interface whose inherited interfaces contains Sensor. This interface must be constructible. Its [Constructor] must take, as argument, an optional dictionary whose inherited dictionaries contains SensorOptions. Its attributes which expose sensor readings are read only and their getters must return the result of invoking get value from latest reading with this and attribute identifier as arguments.

• A PermissionName.

An extension specification may specify the following definitions for each sensor types:

• A dictionary whose inherited dictionaries contains SensorOptions.

• A default sensor. Generally, devices are equipped with a single platform sensor of each type, so defining a default sensor should be straightforward. For sensor types where multiple sensors are common, extension specifications may choose not to define a default sensor, especially when doing so would not make sense.

• A set of identifying parameters. TODO: replace that by an abstract operation.

9.7. Extending the Permission API

An implementation of the Sensor interface for each sensor type must protect its reading by associated PermissionName or PermissionDescriptor. A Low-level sensor may use its interface name as a PermissionName, for instance, "gyroscope" or "accelerometer". Fusion sensors must request permission to access each of the sensors that are used as a source of fusion.

Even though, it might be difficult to reconstruct low-level sensor readings from fused data, some of the original information might be inferred. For example, it is easy to deduce user’s orientation in space if absolute or geomagnetic orientation sensors are used, therefore, these sensors must request permission to use magnetometer as it provides information about orientation of device in relation to Earth’s magnetic field. In contrast, relative orientation sensor does not expose such information, thus, it does not need to request permission to use magnetometer.

Permission descriptors can also be used to set maximum allowed limits for accuracy or sampling frequency. An example for a possible extension of the Permission API for accelerometer sensor is given below.

dictionary AccelerometerPermissionDescriptor : PermissionDescriptor {
    boolean highAccuracy = false;
    boolean highFrequency = false;
};

9.8. Example WebIDL

Here’s example WebIDL for a possible extension of this specification for proximity sensors.

[Constructor(optional ProximitySensorOptions proximitySensorOptions),
 SecureContext, Exposed=Window]
interface ProximitySensor : Sensor {
    readonly attribute unrestricted double distance;
};

dictionary ProximitySensorOptions : SensorOptions {
    double? min = -Infinity;
    double? max = Infinity;
    ProximitySensorPosition? position;
    ProximitySensorDirection? direction;
};

enum ProximitySensorPosition {
    "top-left",
    "top",
    "top-right",
    "middle-left",
    "middle",
    "middle-right",
    "bottom-left",
    "bottom",
    "bottom-right"
};

enum ProximitySensorDirection {
    "front",
    "rear",
    "left",
    "right",
    "top",
    "bottom"
};

10. Acknowledgements

First and foremost, I would like to thank Anssi Kostiainen for his continuous and dedicated support and input throughout the development of this specification, as well as Mikhail Pozdnyakov, Alexander Shalamov, Rijubrata Bhaumik, and Kenneth Rohde Christiansen for their invaluable implementation feedback, suggestions, and research that have helped inform the specification work.

Special thanks to Rick Waldron for driving the discussion around a generic sensor API design for the Web, sketching the original API on which this is based, providing implementation feedback from his work on Johnny-Five, and continuous input during the development of this specification.

Special thanks to Boris Smus, Tim Volodine, and Rich Tibbett for their initial work on exposing sensors to the web with consistency.

Thanks to Anne van Kesteren for his tireless help both in person and through IRC.

Thanks to Domenic Denicola and Jake Archibald for their help.

Thanks also to Frederick Hirsch and Dominique Hazaël-Massieux (via the HTML5Apps project) for both their administrative help and technical input.

Thanks to Tab Atkins for making Bikeshed and taking the time to explain its subtleties.

Thanks to Lukasz Olejnik and Maryam Mehrnezhad for their contributions around privacy and security.

The following people have greatly contributed to this specification through extensive discussions on GitHub: Anssi Kostiainen, Boris Smus, chaals, Claes Nilsson, Dave Raggett, David Mark Clements, Domenic Denicola, Dominique Hazaël-Massieux (via the HTML5Apps project), Francesco Iovine, Frederick Hirsch, gmandyam, Jafar Husain, Johannes Hund, Kris Kowal, Lukasz Olejnik, Marcos Caceres, Marijn Kruisselbrink, Mark Foltz, Mats Wichmann, Matthew Podwysocki, pablochacin, Remy Sharp, Rich Tibbett, Rick Waldron, Rijubrata Bhaumik, robman, Sean T. McBeth, smaug----, Tab Atkins Jr., Virginie Galindo, zenparsing, and Zoltan Kis.

We’d also like to thank Anssi Kostiainen, Dominique Hazaël-Massieux, Erik Wilde, and Michael[tm] Smith for their editorial input.

Conformance

Document conventions

Conformance requirements are expressed with a combination of descriptive assertions and RFC 2119 terminology. The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in the normative parts of this document are to be interpreted as described in RFC 2119. However, for readability, these words do not appear in all uppercase letters in this specification.

All of the text of this specification is normative except sections explicitly marked as non-normative, examples, and notes. [RFC2119]

Examples in this specification are introduced with the words "for example" or are set apart from the normative text with class="example", like this:

Because this document doesn’t itself define APIs for specific sensor types—that is the role of extensions to this specification—all examples are inevitably (wishful) fabrications. Although all of the sensors used a examples would be great candidates for building atop the Generic Sensor API, their inclusion in this document does not imply that the relevant Working Groups are planning to do so.

Informative notes begin with the word "Note" and are set apart from the normative text with class="note", like this:

Note, this is an informative note.

Conformant Algorithms

Requirements phrased in the imperative as part of algorithms (such as "strip any leading space characters" or "return false") are to be interpreted with the meaning of the key word ("must", "should", "may", etc) used in introducing the algorithm.

Conformance requirements phrased as algorithms or specific steps can be implemented in any manner, so long as the end result is equivalent. In particular, the algorithms defined in this specification are intended to be easy to understand and are not intended to be performant. Implementers are encouraged to optimize.

Conformance Classes

A conformant user agent must implement all the requirements listed in this specification that are applicable to user agents.