Let’s Own This! (“LOT”) is organizing a set of technologists, data professionals, and data scientists, to participate in a community that furthers the LOT mission.
At LOT, our overall objective is to help individuals and families become carbon positive, or in other words, remove more carbon from the environment than they create. This calculus requires a reduction in carbon release, and an increase in carbon capture.
On the carbon consumption side, people need better tools to help them make more informed carbon consumption decisions. Currently in the world, such measurements are ‘directional’ in nature, and not sufficiently precise, immediate, universal, or easy to use.
On the other side of the equation, LOT aims to create and manage projects allowing individuals and families to directly and measurably offset their carbon consumption. Currently, most carbon offset projects are limited to financial schemes, where individuals pay organizations that engage in carbon recovery. While such operations are laudable (and we hope to partner with them), we believe significant value can be created by offering hands-on experiences.
And Data Professionals?
Ultimately the LOT mission depends on reliable collection, management, analysis, and propagation of data. The “radical openness” LOT lives by holds water only when stakeholders believe the story. We need to collect and consolidate new kinds of data, from new sources, and analyze it in new ways, with openness that defies skepticism. It’s gonna be hard!
We’re looking for data scientists, big data engineers, and data architects willing to work together to design and build an infrastructure. AND, we need them to do it for free!
Wow, that ‘s an ask!
Why would you do this? It’s a fair question. I’m hoping a number of people can combine together on a particular data science project, spend 3-5 hours a week on it for a limited number of months, and then move on to something else, hopefully in support of the LOT mission, but maybe not. Here are the reasons I can think of that you might be willing to give the time and effort:
- You believe in the LOT mission, that individual climate cooling efforts can be scaled across millions of people, and that such improvements can add up to real global benefit.
- You believe you can learn from other data professionals in a project team and as part of a community of data scientists.
- Your work supporting the program could be translated into earned reputational carbon credits. This is a real possibility, just not established yet :-).
- You believe you can build a network of data scientists and other professionals who can help you professionally in the future.
- You can build your brand, reputation and portfolio by publishing code, blogs, articles and applications. You can do this in ways you probably can’t or won’t elsewhere, given our uncompromising radical openness. Other (including me!) will be there to help you along the path, provide recommendations in support of future employment, and introduce you to other data professionals!
The data science projects described below are intended to fit together and ultimately support a system that helps users make informed decisions about their carbon consumption and remediation. Combined with existing 3rd party technology and analysis from other sources, we hope to enable negative carbon footprints AT SCALE.
Below is a list of use cases we hope to deliver, and projects that need to be done to accomplish them, that will help fulfill the LOT mission (and help participants fulfill their own objectives).
Use Cases
- In a supermarket, restaurant, or clothing store, a consumer can consult a mobile application that helps them make the most appropriate carbon consumption decision.
- At home, a consumer can direct her utility provider to share monthly household electricity consumption data with LOT
- At home, a consumer can view a web application that collects detailed information about his carbon footprint, along with charts and graphs elucidating current patterns. The application should also deliver suggested changes in behaviour that could improve carbon management.
- On a vacation, during an evening, during the workday, or a weekend an individual, family, youth organization or meetup group can participation in a project that sequesters carbon, and then sees the offset presented in a carbon “account” in the apps described above.
- Given a variety of data inputs, LOT can determine where to best organize physical carbon offset programs, and with which priority.
- Given sentiment analysis, LOT can determine when is the best time to publicize its programs.
- Given market and measurement conditions, LOT can understand optimal timing for selling financial carbon credits.
Data Science Projects supporting the above use cases
- Image Recognition, screen scraping, and fuzzy logic supporting consumer product carbon footprint recognition:
- Most consumer products–especially food items, are sold in packaging that includes a GTIN (Global Trade Item Number). GTIN’s are also often represented by a barcode, which are scanned when consumers purchase the items at retail. The items themselves, their transportation, their packaging, their consumer use profile (like cooking, for example), and recycling, all contribute to a carbon footprint.
- Databases behind GTINs are often non-public, incomplete, and in any case, insufficient to score a carbon footprint.
- When attempting to assemble a consumer’s carbon footprint, it is likely a system will receive a list of GTINs in an API call, and assign carbon consumption to each item.
- This series of projects would include the following:
- Identify a category of consumer products for examination (breakfast cereals, soda pop 6 packs, leafy green produce, footwear, etc.)
- Identify the range of data sources and assumptions useful for assigning a carbon footprint
- Identify the optimal modeling approach and techniques for determining the results
- Publish a paper describing your results
- Sentiment analysis and NLP:
- Judge, quantify, communicate trends on:
- Particular categories of environmental sentiment
- Mindshare of certain environmental concerns (including seasonality)
- How users want to consume various environmental experiences
- Conduct some comparative analysis around the world
- Geospatial analysis:
- Identify optimal locations for forest operations and gathering/supply points based on:
- Availability of nearby population centers and their attitudes towards environment
- State and local policy
- Nearby forests, and their density, including national forests
- Available labor force, accommodation, etc.
- Regional acuteness of climate change effects, including incidents of forest fires, and lower air quality
- Publish a paper to that effect
- Locations of disposal sites
- Local slashpile practices
- API design
- Design and build POCs for APIs that can be implemented by Credit Card issuers, debit cards, loyalty cards (airlines, retailers, discount clubs, etc.), and utilities providers. Take into account:
- Categories of accounts or issuers
- Security measures
- Including required data elements
- Including desired data
- Shiny, Dash or Mobile App Development
- Design or build consumer app enabling the following:
- Manual input of data
- Appropriate user experience for mobile and for desktop
- Incorporates barcode reading
- Presents data originating from various sources
- Develop and publish R/Python packages or modules
- These could vary, grow, and expand over time, but could include:
- Common unit translations (molecular weights, carbon proportion in wood, kilowatts to pounds)
- Common functions
- Common datasets
- Common APIs
- Develop and publish R/Python packages or modules
- Build models LOT can use to understand when and how to sell carbon credits
- Account for regions, regulations, supply/demand, production and demand seasonality, and sentiment.
Want to join us? Zip me an email at serge.bushman@gmail.com
