10 ms·

Your battery status may be used to track you online

170 points by randomname2 10y ago

y0ghur7_xxx 10y ago
why is this even available to websites? navigator.getBattery().then(function(a) {console.log(a);}) https://developer.mozilla.org/en/docs/Web/API/Battery_Status_API https://developer.mozilla.org/en/docs/Web/API/Battery_Status...
- axlee 10y ago
  To be able to load a less resource intensive version (i.e. disable animations, reduce refresh rate, remove canvas elements etc...) of a website if the battery is low.
  - Globz 10y ago
    Yea ok...but who does that really ? lets face it, this is just some extra signals so they can track you more easily. I know it was never intended to be used as tracking information but the least they could do is be less precise about your battery percentage.
    - rtpg 10y ago
      This is less of an issue if websites prompted you. I think the iOS permission model (which is what the web is going towards) is a pretty good model. This lets sites who want to do "real things" with battery levels have access, without it being available "for free" (prompt is annoying).
    - TeMPOraL 10y ago
      This is ridiculous IMO. Instead of providing false (or "less precise") data let's not provide that data in the first place. My guess is this is just another part of the attempt to move desktop features to the web browsers so that people don't have to write actual applications. Personally, I don't like this trend.
      - braythwayt 10y ago
        As I noted elsewhere, it’s about more than “so that people don’t have to write actual applications” (although I grant you that you have identified a real thing). For the ideologically inclined, it’s also about disintermediating the relationship between users and developers. Web sites don’t need an app store.
  - Aardwolf 10y ago
    Then it would certainly be useful to be able to spoof this battery level reported by your browser: then you can always get the less resource intensive versions and surf longer :D
    - leni536 10y ago
      Be careful with that, you might see higher prices then. > “Some companies may be analysing the possibility of monetising the access to battery levels,” he writes. “When battery is running low, people might be prone to some – otherwise different – decisions. In such circumstances, users will agree to pay more for a service.”
      - dasboth 10y ago
        That just feels so sketchy. What a waste of time and knowledge just to chase more profits.
        jacquesm 10y ago
        > What a waste of time and knowledge just to chase more profits. That describes a scarily large portion of all web stuff.
        dasboth 10y ago
        True, but for some reason this one hit a nerve. At least in most cases digital marketing can (at least slightly) benefit the user, but in this case the benefit is 100% with the company. Uber were alleged to charge more for those with low batteries too. Ugh.
        mindslight 10y ago
        Marketing used to be beneficial to consumers by helping them become aware of new products and their features. In an era of instant information search and retrieval, this is no longer necessary. It's all zero-sum spam intended to overwhelm one's decision making capabilities coupled with surveillance to enable better price discrimination.
  - xorcist 10y ago
    If the battery already is too low, it's kind of late to switch to a less resource intensive web page. That should have been done from the beginning.
    - gakada 10y ago
      I agree. The correct way to discriminate between low- and high-resource requiring functionality is "on battery" vs "plugged in". The specific battery level is completely useless.
      - leni536 10y ago
        The API makes this information available too.
  - OMGWTF 10y ago
    Practically nobody cares about not draining your battery. All they care about is your attention and/or your wallet.
- braythwayt 10y ago
  Websites? No. Web applications served over HTTP, on the other hand... Mozilla have it as part of their mission to make technology open and accessible (insert DRM debate here). Native apps have been thrown behind the app store “walled garden.” So, they want to make it possible to create rich apps over HTTP. Thus, you find things like this available to web pages.
milliams 10y ago
Surely the answer here is for the browser to not give such fine-grained information. For example always give the charge level to the nearest 10 percent and the battery life to the nearest 10 seconds.
- dkx 10y ago
  Or just Green (high), Amber (medium) and Red (low). There's no need for more granularity than that.
  - josho 10y ago
    Or take the iOS approach. The phone is either in a low power mode or not. The benefit is that the user doesn't need an additional setting for each potential web app to set the threshold for a low power mode, it's all done once at the OS level.
  - kalleboo 10y ago
    I remember back in the dumbphone Flash Lite days a lot of phones only did that. Seemed to work fine.
- leni536 10y ago
  Or put it behind an explicit permission like every not necessarily conscious user input (video, audio, location).
  - majewsky 10y ago
    I'm skeptical. If you ask for permission to capture video, that's something that users will understand, and they'll be able to make an informed decision. But if you ask for permission to check the battery level, most people will just click through the request because they don't understand the privacy implications.
    - ape4 10y ago
      "Allow this website to know your battery level? (Can be used to identify you)" yes/no
amga_ 10y ago
Bullshit..
etatoby 10y ago
about:config, dom.battery.enabled = false Plus ad blocking and "social button" blocking, as a preventive measure for this kind of issue. The Web is the new Wild West.
- chris0x00 10y ago
  Thanks, came here to say the same thing. This was as surprsing to me as the first time a spammy website started vibrating my phone (dom.vibrator.enabled)
- yAnonymous 10y ago
  You are now one of 10 people with that flag disabled. Not sure that makes you less trackable.
  - denzil_correa 10y ago
    This has always been an interesting question. What is anonymity (less trace-ability) - hiding in the crowd or uniqueness?
    - softawre 10y ago
      What is the argument for uniqueness?
      - denzil_correa 10y ago
        Example - Satoshi Nakamoto of Bitcoin fame
        Phlarp 10y ago
        Satoshi is the exception, not the rule. Becoming famous and staying anonymous are diametrically opposed.
        mikeash 10y ago
        Not really anonymous. People have a pretty good idea of which bitcoins are his. He just never does anything with them, so that knowledge doesn't serve much of a purpose. It would be sort of like setting your browser's user-agent to a unique identifier that you never change, but then never actually using the browser for anything. It's not the unique ID that hides you, it's the fact that you never use it.
amelius 10y ago
Shouldn't things like this be illegal? I mean, you are not allowed to use TV cameras anywhere, unless strict rules are obeyed (closed circuit, etc). Why should we allow companies recording other features that can identify us?
- mdrzn 10y ago
  But this is more like saying "I'm walking in a shop, the shop shouldn't record what tipe of shirt I'm wearing". It's not something hidden if the phone is giving the info out. The only thing we could do is stop the phone from sending the information in the first place.
  - amelius 10y ago
    But your shirt is also radiating information :) (btw, I don't think this should be a technical debate)
erdevs 10y ago
Mobile device fingerprinting has been a big deal for a long time. Multiple startups have been created to do it and some have been acquired. And many adtech companies incorporate this functionality. I hope some day soon your mobile OS let's you choose to restrict the device information available to sites and apps. It really ought to be a user's choice whether to make the tradeoff between providing more info in exchange for supposedly better functionality or not.
aaronbrager 10y ago
Worth noting that Safari on iOS doesn't support this API.
- kalleboo 10y ago
  I guess sometimes I'm happy using "the IE6 of 2016"
zaroth 10y ago
This was a terrible idea from the start. In the best case scenario, what is supposed to happen? The sites I visit suddenly start being served as shells of their former selves in order to save the compute cost of rendering them? Web designers are supposed to figure out an entirely new design mode for the whole site to cater to visitors with low batteries? Now back to the real world, where many sites I visit peg my desktop CPU trying to serve so many ads and so much tracking to squeeze every possible cent of profitability from my visit. If there's a war against ad-blockers, did we think these sites would relent and say, oh, OK, I see you might be low on battery, I'll serve just the content you want this time?! No, this is a purely client-side concern, with plenty of purely client-side mitigation which can be put into place. That the protocol actually specified 14m degrees of granularity -- from what at the very least ought to have been a binary setting -- makes me wonder if the point wasn't user tracking all along. Missing from the article: Do all user agents actually provide this information in request headers? Is there a way to shut it off?
- DougWebb 10y ago
  As a web application developer, one idea that comes to mind is that, if I were writing an app that supports offline usage, I'd probably want to be aware of a low battery situation so that I can save my state and warn the user that the app needs to shut down to avoid data loss. I might also switch from online mode to offline mode, again letting the user know (and letting them override the choice, maybe.) I can't see any reason to make use of the battery status server-side. But, there's also no way to prevent the information from being sent, once you make it available to javascript code on the device. If the standard api didn't send it, then you'd just see ad-hoc apis gathering the data locally and sending it.
  - DashRattlesnake 10y ago
    However, I don't think the battery state needs to be as granular as it is to support those use-cases. A simple boolean hasLowBattery or a much coarser 0%-25%-50%-75%-100% enumeration seems like it would have been sufficient.
    - 10y ago
wtbob 10y ago
Not mine, since I have JavaScript turned off (how do you know I have JavaScript turned off? Don't worry, I'll let you know every time there's a browser exploit or privacy hole due to JavaScript — so at least thrice weekly).
brlewis 10y ago
Wow, I'm surprised how much info is available in desktop chrome. Events get fired when I unplug my laptop and plug it back in: https://www.audero.it/demo/battery-status-api-demo.html https://www.audero.it/demo/battery-status-api-demo.html I searched Chrome settings for "battery" and found nothing. I did find this article with instructions for turning it off in Firefox: https://www.hackread.com/smartphone-laptop-battery-invading-privacy/ https://www.hackread.com/smartphone-laptop-battery-invading-... I'm not opposed to the concept of a battery API, but as others have commented, it would be just as useful with a lot less granularity.
- battre 10y ago
  Chrome has fixed the granularity a long time ago https://chromium.googlesource.com/chromium/src/+/637776d1183dd286f040349341e1881023851ef9 https://chromium.googlesource.com/chromium/src/+/637776d1183... and I guess most other browsers have fixed it as well.
  - brlewis 10y ago
    They've closed the window but left the door open. See the link I posted. When the page knows "Battery will be discharged in 18365 seconds" it doesn't matter that the battery percentage is only two digits.
dolzenko 10y ago
Every time these new user tracking hacks come up (like the one based on Audio API etc.) I wonder why in the world this is not handled in a manner similar to Android where it prompts you to give permissions to the any specific API its going to use.
guelo 10y ago
Because web developers have this dream of replacing native apps, the industry has been busy removing piece by piece the sandbox that browser apps live in. They want to have all the power and features of native apps except one: app store review and approval. But if you have all the power of native without any moderation, plus the wild west of user-hostile ad-networks... actually i don't know where it leads but it's not good.
- brlewis 10y ago
  I have that dream but didn't ask for the sandbox to be removed. Having users approve use of location data is quite fine by me, and clearly superior to the way it works with native Android apps, where the user has to approve a bunch of permissions at install time, not knowing for sure which ones are actually necessary.
frgewut 10y ago
Such abuse of technology was the main target of the "EU cookie law", not the use of session cookies that all sites are warning about now.
aftbit 10y ago
How do I disable this in Chromium? I feel like I ask this question at least once a month, by the way. Last time, it was for WebRTC information leakage...
mkagenius 10y ago
Uber had done some data science on battery levels too: http://www.huffingtonpost.in/entry/uber-surge-pricing-battery-life_us_573f2057e4b0613b512a0130 http://www.huffingtonpost.in/entry/uber-surge-pricing-batter...
joshkpeterson 10y ago
Why does the battery status have to be so accurate that it can be used to identify? Your device should be able to fuzz it and just send an integer between 1 and 100 - that should be enough granularity for the intended use case. Update the damn spec.
tjpaudio 10y ago
These types of articles are silly in that they leave out the context that nearly every variable that is pseudo-unique to you is already being used and that this is a common practice known as fingerprinting that has been done since forever. Headline: "New variable x can be used to track you". Well, yea. Add that to the size your browser window is, what plugins you have installed, your IP address, and many more ways a website can use to surmise a relationship between two disjoint events.
Cpoll 10y ago
It's just another data point to add to a long list for fingerprinting. See: https://panopticlick.eff.org/ https://panopticlick.eff.org/
- barefootcoder 10y ago
  I was about to post the same link in another response, but then I saw you'd already posted it. It's quite scary, really. Here's my result: Your browser fingerprint appears to be unique among the 119,205 tested so far. Privacy is great...
gbl08ma 10y ago
As someone building a cross-platform battery monitor that depends on APIs like this being available, I think the best solution would be to treat battery status like the location information, where the user is asked on a case-by-case basis whether to provide the information. You might argue that this sort of stuff should only be available to native applications, that browsers should not attempt to replace native apps or even recreate an OS within themselves (something Chrome is often criticized for). The thing is, my battery monitor has native apps and yet users have been asking for web-based clients, be it browser extensions or plain web pages. And with some platforms loading a website or installing an extension has much less friction than installing a native app. Sure, extensions can have access to a lot of stuff webpages don't have. But I would rather use standardized APIs that work across all browsers and are well documented in more than one website, than have to rewrite my extension for each browser and not have it work on plain old pages. If the battery API was made to work like the location one I think everyone would be happy - users are informed and in control, developers can still offer features based on it, and it raises red flags when it needs to ("why is this advert asking about my battery status?").
cimnine 10y ago
This, and certain other capabilities, should be guarded like access to location data: Available upon user agreement only and just on secure connections (i.e. TLS).

170 points by randomname2 10y ago

why is this even available to websites? navigator.getBattery().then(function(a) {console.log(a);}) https://developer.mozilla.org/en/docs/Web/API/Battery_Status_API https://developer.mozilla.org/en/docs/Web/API/Battery_Status...

axlee 10y ago
To be able to load a less resource intensive version (i.e. disable animations, reduce refresh rate, remove canvas elements etc...) of a website if the battery is low.
- Globz 10y ago
  Yea ok...but who does that really ? lets face it, this is just some extra signals so they can track you more easily. I know it was never intended to be used as tracking information but the least they could do is be less precise about your battery percentage.
  - rtpg 10y ago
    This is less of an issue if websites prompted you. I think the iOS permission model (which is what the web is going towards) is a pretty good model. This lets sites who want to do "real things" with battery levels have access, without it being available "for free" (prompt is annoying).
  - TeMPOraL 10y ago
    This is ridiculous IMO. Instead of providing false (or "less precise") data let's not provide that data in the first place. My guess is this is just another part of the attempt to move desktop features to the web browsers so that people don't have to write actual applications. Personally, I don't like this trend.
    - braythwayt 10y ago
      As I noted elsewhere, it’s about more than “so that people don’t have to write actual applications” (although I grant you that you have identified a real thing). For the ideologically inclined, it’s also about disintermediating the relationship between users and developers. Web sites don’t need an app store.
- Aardwolf 10y ago
  Then it would certainly be useful to be able to spoof this battery level reported by your browser: then you can always get the less resource intensive versions and surf longer :D
  - leni536 10y ago
    Be careful with that, you might see higher prices then. > “Some companies may be analysing the possibility of monetising the access to battery levels,” he writes. “When battery is running low, people might be prone to some – otherwise different – decisions. In such circumstances, users will agree to pay more for a service.”
    - dasboth 10y ago
      That just feels so sketchy. What a waste of time and knowledge just to chase more profits.
      - jacquesm 10y ago
        > What a waste of time and knowledge just to chase more profits. That describes a scarily large portion of all web stuff.
        dasboth 10y ago
        True, but for some reason this one hit a nerve. At least in most cases digital marketing can (at least slightly) benefit the user, but in this case the benefit is 100% with the company. Uber were alleged to charge more for those with low batteries too. Ugh.
        mindslight 10y ago
        Marketing used to be beneficial to consumers by helping them become aware of new products and their features. In an era of instant information search and retrieval, this is no longer necessary. It's all zero-sum spam intended to overwhelm one's decision making capabilities coupled with surveillance to enable better price discrimination.
- xorcist 10y ago
  If the battery already is too low, it's kind of late to switch to a less resource intensive web page. That should have been done from the beginning.
  - gakada 10y ago
    I agree. The correct way to discriminate between low- and high-resource requiring functionality is "on battery" vs "plugged in". The specific battery level is completely useless.
    - leni536 10y ago
      The API makes this information available too.
- OMGWTF 10y ago
  Practically nobody cares about not draining your battery. All they care about is your attention and/or your wallet.
braythwayt 10y ago
Websites? No. Web applications served over HTTP, on the other hand... Mozilla have it as part of their mission to make technology open and accessible (insert DRM debate here). Native apps have been thrown behind the app store “walled garden.” So, they want to make it possible to create rich apps over HTTP. Thus, you find things like this available to web pages.

milliams 10y ago

Surely the answer here is for the browser to not give such fine-grained information. For example always give the charge level to the nearest 10 percent and the battery life to the nearest 10 seconds.

dkx 10y ago
Or just Green (high), Amber (medium) and Red (low). There's no need for more granularity than that.
- josho 10y ago
  Or take the iOS approach. The phone is either in a low power mode or not. The benefit is that the user doesn't need an additional setting for each potential web app to set the threshold for a low power mode, it's all done once at the OS level.
- kalleboo 10y ago
  I remember back in the dumbphone Flash Lite days a lot of phones only did that. Seemed to work fine.
leni536 10y ago
Or put it behind an explicit permission like every not necessarily conscious user input (video, audio, location).
- majewsky 10y ago
  I'm skeptical. If you ask for permission to capture video, that's something that users will understand, and they'll be able to make an informed decision. But if you ask for permission to check the battery level, most people will just click through the request because they don't understand the privacy implications.
  - ape4 10y ago
    "Allow this website to know your battery level? (Can be used to identify you)" yes/no

amga_ 10y ago

Bullshit..

etatoby 10y ago

about:config, dom.battery.enabled = false Plus ad blocking and "social button" blocking, as a preventive measure for this kind of issue. The Web is the new Wild West.

chris0x00 10y ago
Thanks, came here to say the same thing. This was as surprsing to me as the first time a spammy website started vibrating my phone (dom.vibrator.enabled)
yAnonymous 10y ago
You are now one of 10 people with that flag disabled. Not sure that makes you less trackable.
- denzil_correa 10y ago
  This has always been an interesting question. What is anonymity (less trace-ability) - hiding in the crowd or uniqueness?
  - softawre 10y ago
    What is the argument for uniqueness?
    - denzil_correa 10y ago
      Example - Satoshi Nakamoto of Bitcoin fame
      - Phlarp 10y ago
        Satoshi is the exception, not the rule. Becoming famous and staying anonymous are diametrically opposed.
      - mikeash 10y ago
        Not really anonymous. People have a pretty good idea of which bitcoins are his. He just never does anything with them, so that knowledge doesn't serve much of a purpose. It would be sort of like setting your browser's user-agent to a unique identifier that you never change, but then never actually using the browser for anything. It's not the unique ID that hides you, it's the fact that you never use it.

amelius 10y ago

Shouldn't things like this be illegal? I mean, you are not allowed to use TV cameras anywhere, unless strict rules are obeyed (closed circuit, etc). Why should we allow companies recording other features that can identify us?

mdrzn 10y ago
But this is more like saying "I'm walking in a shop, the shop shouldn't record what tipe of shirt I'm wearing". It's not something hidden if the phone is giving the info out. The only thing we could do is stop the phone from sending the information in the first place.
- amelius 10y ago
  But your shirt is also radiating information :) (btw, I don't think this should be a technical debate)

erdevs 10y ago

Mobile device fingerprinting has been a big deal for a long time. Multiple startups have been created to do it and some have been acquired. And many adtech companies incorporate this functionality. I hope some day soon your mobile OS let's you choose to restrict the device information available to sites and apps. It really ought to be a user's choice whether to make the tradeoff between providing more info in exchange for supposedly better functionality or not.

aaronbrager 10y ago

Worth noting that Safari on iOS doesn't support this API.

kalleboo 10y ago
I guess sometimes I'm happy using "the IE6 of 2016"

zaroth 10y ago

This was a terrible idea from the start. In the best case scenario, what is supposed to happen? The sites I visit suddenly start being served as shells of their former selves in order to save the compute cost of rendering them? Web designers are supposed to figure out an entirely new design mode for the whole site to cater to visitors with low batteries? Now back to the real world, where many sites I visit peg my desktop CPU trying to serve so many ads and so much tracking to squeeze every possible cent of profitability from my visit. If there's a war against ad-blockers, did we think these sites would relent and say, oh, OK, I see you might be low on battery, I'll serve just the content you want this time?! No, this is a purely client-side concern, with plenty of purely client-side mitigation which can be put into place. That the protocol actually specified 14m degrees of granularity -- from what at the very least ought to have been a binary setting -- makes me wonder if the point wasn't user tracking all along. Missing from the article: Do all user agents actually provide this information in request headers? Is there a way to shut it off?

DougWebb 10y ago
As a web application developer, one idea that comes to mind is that, if I were writing an app that supports offline usage, I'd probably want to be aware of a low battery situation so that I can save my state and warn the user that the app needs to shut down to avoid data loss. I might also switch from online mode to offline mode, again letting the user know (and letting them override the choice, maybe.) I can't see any reason to make use of the battery status server-side. But, there's also no way to prevent the information from being sent, once you make it available to javascript code on the device. If the standard api didn't send it, then you'd just see ad-hoc apis gathering the data locally and sending it.
- DashRattlesnake 10y ago
  However, I don't think the battery state needs to be as granular as it is to support those use-cases. A simple boolean hasLowBattery or a much coarser 0%-25%-50%-75%-100% enumeration seems like it would have been sufficient.
  - 10y ago

wtbob 10y ago

Not mine, since I have JavaScript turned off (how do you know I have JavaScript turned off? Don't worry, I'll let you know every time there's a browser exploit or privacy hole due to JavaScript — so at least thrice weekly).

brlewis 10y ago

Wow, I'm surprised how much info is available in desktop chrome. Events get fired when I unplug my laptop and plug it back in: https://www.audero.it/demo/battery-status-api-demo.html https://www.audero.it/demo/battery-status-api-demo.html I searched Chrome settings for "battery" and found nothing. I did find this article with instructions for turning it off in Firefox: https://www.hackread.com/smartphone-laptop-battery-invading-privacy/ https://www.hackread.com/smartphone-laptop-battery-invading-... I'm not opposed to the concept of a battery API, but as others have commented, it would be just as useful with a lot less granularity.

battre 10y ago
Chrome has fixed the granularity a long time ago https://chromium.googlesource.com/chromium/src/+/637776d1183dd286f040349341e1881023851ef9 https://chromium.googlesource.com/chromium/src/+/637776d1183... and I guess most other browsers have fixed it as well.
- brlewis 10y ago
  They've closed the window but left the door open. See the link I posted. When the page knows "Battery will be discharged in 18365 seconds" it doesn't matter that the battery percentage is only two digits.

dolzenko 10y ago

Every time these new user tracking hacks come up (like the one based on Audio API etc.) I wonder why in the world this is not handled in a manner similar to Android where it prompts you to give permissions to the any specific API its going to use.

guelo 10y ago

Because web developers have this dream of replacing native apps, the industry has been busy removing piece by piece the sandbox that browser apps live in. They want to have all the power and features of native apps except one: app store review and approval. But if you have all the power of native without any moderation, plus the wild west of user-hostile ad-networks... actually i don't know where it leads but it's not good.

brlewis 10y ago
I have that dream but didn't ask for the sandbox to be removed. Having users approve use of location data is quite fine by me, and clearly superior to the way it works with native Android apps, where the user has to approve a bunch of permissions at install time, not knowing for sure which ones are actually necessary.

frgewut 10y ago

Such abuse of technology was the main target of the "EU cookie law", not the use of session cookies that all sites are warning about now.

aftbit 10y ago

How do I disable this in Chromium? I feel like I ask this question at least once a month, by the way. Last time, it was for WebRTC information leakage...

mkagenius 10y ago

Uber had done some data science on battery levels too: http://www.huffingtonpost.in/entry/uber-surge-pricing-battery-life_us_573f2057e4b0613b512a0130 http://www.huffingtonpost.in/entry/uber-surge-pricing-batter...

joshkpeterson 10y ago

Why does the battery status have to be so accurate that it can be used to identify? Your device should be able to fuzz it and just send an integer between 1 and 100 - that should be enough granularity for the intended use case. Update the damn spec.

tjpaudio 10y ago

These types of articles are silly in that they leave out the context that nearly every variable that is pseudo-unique to you is already being used and that this is a common practice known as fingerprinting that has been done since forever. Headline: "New variable x can be used to track you". Well, yea. Add that to the size your browser window is, what plugins you have installed, your IP address, and many more ways a website can use to surmise a relationship between two disjoint events.

Cpoll 10y ago

It's just another data point to add to a long list for fingerprinting. See: https://panopticlick.eff.org/ https://panopticlick.eff.org/

barefootcoder 10y ago
I was about to post the same link in another response, but then I saw you'd already posted it. It's quite scary, really. Here's my result: Your browser fingerprint appears to be unique among the 119,205 tested so far. Privacy is great...

gbl08ma 10y ago

As someone building a cross-platform battery monitor that depends on APIs like this being available, I think the best solution would be to treat battery status like the location information, where the user is asked on a case-by-case basis whether to provide the information. You might argue that this sort of stuff should only be available to native applications, that browsers should not attempt to replace native apps or even recreate an OS within themselves (something Chrome is often criticized for). The thing is, my battery monitor has native apps and yet users have been asking for web-based clients, be it browser extensions or plain web pages. And with some platforms loading a website or installing an extension has much less friction than installing a native app. Sure, extensions can have access to a lot of stuff webpages don't have. But I would rather use standardized APIs that work across all browsers and are well documented in more than one website, than have to rewrite my extension for each browser and not have it work on plain old pages. If the battery API was made to work like the location one I think everyone would be happy - users are informed and in control, developers can still offer features based on it, and it raises red flags when it needs to ("why is this advert asking about my battery status?").

cimnine 10y ago

This, and certain other capabilities, should be guarded like access to location data: Available upon user agreement only and just on secure connections (i.e. TLS).