You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
The pattern for Reduce Transmitted Data ignores the fact that network devices are always "on" and their energy needs are independent of actual traffic. So reducing traffic has no consequential effect on emissions, unless hardware is added/removed from the network.
I got triggered by this article on the Green Software Foundation site: measuring serverless emissions. The article points to the "Reduce Transmitted Data" pattern and assumes that cutting traffic will reduce emissions. While this may be true from an attributional point of view (your 'share' of the energy used on the network) it is not true for the consequential point of view (sending some more traffic or not has no effect on actual energy use and therefore emissions).
The article states that: "When comparing different measurements with varying picture sizes, we observe that the “Reduce transmitted data” pattern can achieve approximately a 4% reduction in carbon emissions".
I consider this statement of 4% reduction highly questionable, since it can only be realized if actual equipment is turned off.
Describe the solution you'd like
The pattern for Reduce Transmitted Data should at least point people towards the consequential vs attributional emission models, which are very relevant in this topic.
Is your feature request related to a problem? Please describe.
The pattern for Reduce Transmitted Data ignores the fact that network devices are always "on" and their energy needs are independent of actual traffic. So reducing traffic has no consequential effect on emissions, unless hardware is added/removed from the network.
I got triggered by this article on the Green Software Foundation site: measuring serverless emissions. The article points to the "Reduce Transmitted Data" pattern and assumes that cutting traffic will reduce emissions. While this may be true from an attributional point of view (your 'share' of the energy used on the network) it is not true for the consequential point of view (sending some more traffic or not has no effect on actual energy use and therefore emissions).
The article states that: "When comparing different measurements with varying picture sizes, we observe that the “Reduce transmitted data” pattern can achieve approximately a 4% reduction in carbon emissions".
I consider this statement of 4% reduction highly questionable, since it can only be realized if actual equipment is turned off.
Describe the solution you'd like
The pattern for Reduce Transmitted Data should at least point people towards the consequential vs attributional emission models, which are very relevant in this topic.
Describe alternatives you've considered
None.
Additional context
See for example this presentation about network energy use & emissions, presented at green.io Paris:
https://drive.google.com/file/d/1VehKnslmdvdWyN_8tRHWo-8rKocfZGuc/view
Or a very elaborate article on this topic in general: https://hubblo.org/blog/attributional-vs-consequential/
The text was updated successfully, but these errors were encountered: