thangdnsf/consumsis

consumption analysis plugin for chrome browser

Hi there 👋

My name's Thang Nguyen (Davis)

• 🔭 I’m currently working as an IT & MLops manager and AI engineer at Torus Actions SAS, Toulouse, France
• 👯 I’m looking to collaborate on Deep learning, Machine learning, Data mining
• 📫 How to reach me: in/thangdnsf
• ⚡ Goal: Seeking a full-time position/Ph.D. program in Machine learning/ Deep learning and Data-mining in medical or healthcare.

consumsis's Issues

Electricity consumption wrongly reported.

There is an issue in the reporting of the electricity consumption. For instance, I get:

• Spent 32MB
• GHG : 0.25 gCO2e
• electricity : 25852 Wh

whereas the 0.25 gCO2e seems to match "(7.20e-11+1.52e-10)321e60.0351000", with the "0.035" coming from device.GHG for France, I expect the electricity to be:

"(7.20e-11+1.52e-10)321e6*1000" = 7.16 Wh

and not the extravagant 25852 Wh !

Issues in conversion factors

The conversion factor for walking is currently set as 750gCO2e/km:

const co2perwalkingkm = 0.75;

which is three times the factor by car !

Pessimistic estimations are closer to 0.01, but converting a walking distance to CO2e is really tricky and subject to endless debates on what should be taken into account. I would thus strongly advice to remove this metaphor.

The same remark applies for bicycling.

For electricity, 0.035 kgCO2e/kWh in France is extremely optimistic (it seems to come from the IEA) and this number neglects the importations and infrastructures. A more consensual value would be 0.056 as reported by the ADEME, or independently computed by electricy map. Others report 0.086 or even 0.112.