“one-size-fits-all” web experience does not work in a world with widely varying device capabilities. Web apps that delight users on high end devices can be unusable on low end devices, particularly in emerging markets. To support a diverse web ecosystem that is inclusive of users on low end devices, apps need to be able to tailor their content to the device constraints. This allows users to consume content suited to their device, and ensures they have a good experience and fewer frustrations.
Developers are interested in the “device-class” for the following known use-cases:
- Serve a “light” version of the site or specific components, for low end devices. Egs:
- Google’s "search lite" is a 10KB search results page used in EM.
- Serve a light version of video player in Facebook
- Serve lightweight tile images in Google Maps
- Normalize Metrics: analytics need to be able to normalize their metrics against the device-class. For instance, a 100ms long tasks on a Pixel is a more severe issue vs. on a low end device.
Device memory is an interesting signal in this context. Low memory devices devices (under 512MB, 512MB - 1GB) are widely used in emerging markets. Chrome telemetry indicates large number of OOM (out-of-memory) crashes on foreground tabs on these devices. In this case, serving a lite version not only improves the user experience, it is necessary for the site to be usable at all (as opposed to crashing due to memory constraint).
We propose a header and web exposed API to surface device capability for memory (RAM). The mechanism should be extensible to other device capabilities such as CPU i.e. number of cores, clock speed etc. A header will enable the server to deliver appropriate content, eg. a “lite” version of the site. The JS API will enable clients to make appropriate decisions eg. using more storage vs. making additional requests, requesting appropriate resources from the server etc.
ASIDE: the JS API for CPU Cores is already available via hardwareConcurrency API
Proposed Header for memory: device-ram
device-ram : <value>
where <value> is an approximation the amount of ram in GiB (floating point number). The <value> is calculated by using the actual device memory in MiB then rounding it to the smallest nearest number where only the 2 most signicant bits can be set and the rest are zeros (i.e of the binary form ((1|11)0*)). Then diving that number by 1024.0 to get the value in GiB.
The following table illustrates some examples:
| Actual in MiB | Rounded in MiB | Reported in GiB |
|---|---|---|
| 500 | 512 | 0.5 |
| 512 | 512 | 0.5 |
| 768 | 768 | 0.75 |
| 1000 | 1024 | 1 |
| 1500 | 1536 | 1.5 |
| 1792 | 1536 | 1.5 |
| 1793 | 2048 | 2 |
A conclusive list of all possible values between 0.25Gib and 128GiB is as follow:
0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 12, 16, 24, 32, 48, 64, 96, 128
HTTP caching doesn't deal well with mixed value headers, therefore separate headers are recommended. Also, rounding down to power of two enables caching and mitigates fingerprinting.
Currently Client Hints cannot be used to enable providing network performance data on the first request, however this is being actively addressed with this proposal.
The header is sent after an explicit per-origin opt-in via Client Hints mechanism. The following new hint will be added: Accept-CH: device-ram
For background, Client Hints provides a set of HTTP request header fields, known as Client Hints, to deliver content that is optimized for the device. In that sense using Client Hints is a great fit for this proposal.
Client Hints recently addressed a significant limitation in spec: opt-in can now be persisted across browser restarts using max-age in header.
We propose adding the following API to navigator: navigator.deviceRam
which returns number of GiB of ram (floating point number) rounded down to the nearest power of two (same as the header).
While exposing a composite “Device Class” would be useful for developers, it has a number of challenges: it’s hard to specify in a future-proof manner as it is constantly changing, it requires significant and ongoing work (constantly update algorithm OR classify known devices), it is difficult to reach agreement amongst vendors, and come up with something that works for all web sites etc. While this is something worth considering down the road, we think we can get most of the benefit by exposing a couple of specific signals device memory and device CPU cores. In particular device memory is a reasonable proxy for device class.
See Security section in Client Hints. Requiring per-origin opt-in with Accept-CH restricts when the header is advertised. Restricting to a ceiling value (rounded down to the nearest power of two), as opposed to exact value, reduces fingerprinting risk. This can be further restricted to returning values only for low memory cases i.e. less than 1GB.
https://github.com/facebook/device-year-class https://github.com/igrigorik/http-client-hints https://developers.google.com/web/updates/2015/09/automating-resource-selection-with-client-hints WICG/netinfo#46 https://developer.mozilla.org/en-US/docs/Web/API/NavigatorConcurrentHardware