From 529f2ff872e2dd47e6903895d936c9d34e607ae7 Mon Sep 17 00:00:00 2001 From: Nina Di Cara Date: Thu, 30 May 2024 13:26:50 +0100 Subject: [PATCH] Reorganise info between labels and index with admonition boxes --- site/index.md | 21 ++++----------------- site/labels.md | 45 ++++++++++++++++++++++++++++++++++++++++----- 2 files changed, 44 insertions(+), 22 deletions(-) diff --git a/site/index.md b/site/index.md index 30fa5d6..8515597 100644 --- a/site/index.md +++ b/site/index.md @@ -10,26 +10,13 @@ Data Hazards is a project to help us all identify the hazards of using data scie Data scientists are great at selling our work, for example communicating the gains in efficiency and accuracy, but we are less well-practiced in thinking about the ethical implications of our work. The ethical implications go beyond most ethics Institutional Review Boards, to questions about the wider societal impact of data science and algorithms work. -### Aims -We aim to __create resources__ to: -1. Create a community-driven open-source vocabulary of ethical hazards for data-intensive research and development, in the form of [Data Hazard labels](contents/data-hazards). -2. Make ethical and future-thinking more accessible to data scientists, computer scientists and applied mathematicians to apply to their own work. -3. Enable bringing together and respecting diverse and interdisciplinary viewpoints to this work, and produce resources to do this. +```{admonition} Project Aims +1. Create a __community-driven open-source vocabulary__ of ethical hazards for data-intensive research and development, in the form of [Data Hazard labels](contents/data-hazards). +2. Make ethical and future-thinking __more accessible__ to data scientists, computer scientists and applied mathematicians to apply to their own work. +3. Enable bringing together and respecting __diverse and interdisciplinary viewpoints__ to this work, and produce resources to do this. 4. Find out in what circumstances, and for who, these resources work best. - -```{admonition} Why are the Hazard Labels so scary-looking? -We know that the [Data Hazards labels](contents/data-hazards) are a bit frightening. Argh, there's a skull! -Please know that __we don't want these labels to scare anyone away from considering ethics or from doing data science__, and we will do everything that we can to make applying Data Hazards labels as welcoming and approachable as possible, but also have some good reasons for choosing these images. - -We chose this format because of the similarity to [COSHH hazard labels](https://www.hse.gov.uk/chemical-classification/labelling-packaging/hazard-symbols-hazard-pictograms.htm) - hazard labels for chemicals. -We made this choice because we want a similar response from people: -1. Attention-grabbing, asking people to stop and think, and take the safety precautions seriously, rather than as an optional extra. -2. We're asking people to "handle with care", not to stop doing the work. We still use chemicals, but we think about how it can be done safely and how to avoid emergencies. -3. They are familiar, especially to scientists, who (within universities) tend to have the least experience of applying ethics. - ``` - ## Contributors Our brilliant contributors are listed here, and you can [read more detail about our contributing process here](contribute). diff --git a/site/labels.md b/site/labels.md index 0da0038..9e24c1c 100644 --- a/site/labels.md +++ b/site/labels.md @@ -2,13 +2,26 @@ __Version 1.0__ -This page contains the Data Hazard labels themselves. -We welcome you to suggest changes, so please check our [contribution guidelines](contribute) if you would like to. +On this page you can see an overview of the current Data Hazards with short descriptions. Click on each for their full information. +We welcome you to suggest changes, so please check our [contribution guidelines](contribute) if you would like to or scroll down for current suggestions below. -Each hazard has: -- An image, title, and description to describe the Hazard. +Each individual Data Hazard page contains: +- A __title__, __description__ and __icon__ to describe the Hazard. - __Examples__ to clarify what the hazard covers. -- __Safety Precautions__ - suggestions of how Hazards could be mitigated. +- __Safety Precautions__ as suggestions of how Hazards could be mitigated. + + +:::{dropdown} Why are the Hazard Labels designed this way? +:color: success + +Please know that __we don't want these labels to scare anyone away from considering ethics or from doing data science__, and we will do everything that we can to make applying Data Hazards labels as welcoming and approachable as possible, but also have some good reasons for choosing these images. + +We chose this format because of the similarity to [COSHH hazard labels](https://www.hse.gov.uk/chemical-classification/labelling-packaging/hazard-symbols-hazard-pictograms.htm) - hazard labels for chemicals. +We made this choice because we want a similar response from people: +1. Attention-grabbing, asking people to stop and think, and take the safety precautions seriously, rather than as an optional extra. +2. We're asking people to __"handle with care"__, not to stop doing the work. We still use chemicals, but we think about how it can be done safely and how to avoid emergencies. +3. They are familiar, especially to scientists, who (within universities) tend to have the least experience of applying ethics. +::: @@ -21,6 +34,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with an exclamation mark in the middle. :link: /hazards/general-hazard :link-type: doc + +Data Science is being used in this output, and any negative outcome of using this work are not the fault of "the algorithm" or "the software". :::: ::::{grid-item-card} Automates Decision Making @@ -28,6 +43,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with two connected cogs that have arrows coming out of the top of them. :link: hazards/automates-decision-making :link-type: doc + +Automated decision making can be hazardous for a number of reasons, and these will be highly dependent on the field in which it is being applied. :::: ::::{grid-item-card} Danger Of Misuse @@ -35,6 +52,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with a hammer raised above a bent screw in the middle. :link: hazards/danger-of-misuse :link-type: doc + +There is a danger of misusing the algorithm, technology, or data collected as part of this work. :::: ::::{grid-item-card} Difficult to Understand @@ -42,6 +61,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with an image of a closed box and a question mark next to it. :link: hazards/difficult-to-understand :link-type: doc + +This may apply if the technology itself is hard to interpret (e.g. neural nets), or documentation is poor/unavailable. :::: ::::{grid-item-card} High Environmental Cost @@ -49,6 +70,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with an image of a globe on fire in the middle. :link: hazards/high-environmental-cost :link-type: doc + +Indicates methodologies that are energy-hungry, data-hungry, or require special hardware with rare materials. :::: ::::{grid-item-card} Lacks Community Involvement @@ -56,6 +79,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with figures in the middle who have a speech bubble above their heads with a big cross through it. :link: hazards/lacks-community-involvement :link-type: doc + +This applies when technology is being produced without input from the community it is supposed to serve. :::: ::::{grid-item-card} Lacks Informed Consent @@ -63,6 +88,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) containing a magnifying glass hovering over a cross on a piece of paper. :link: hazards/lacks-informed-consent :link-type: doc + +This hazard applies to datasets or algorithms that use data which has not been provided with the explicit consent of the data owner/creator. :::: ::::{grid-item-card} May Cause Direct Harm @@ -70,6 +97,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with a skull in the middle. :link: hazards/direct-harm :link-type: doc + +The application area of this technology means that it is capable of causing direct physical or psychological harm to someone even if used correctly. :::: ::::{grid-item-card} Ranks Or Classifies People @@ -77,6 +106,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with three people standing on a podium of first, second and third place. :link: hazards/ranks-classifies :link-type: doc + +Ranking and classifications of people are hazards in their own right and should be handled with care. :::: ::::{grid-item-card} Reinforces Existing Biases @@ -84,6 +115,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with a dark head and shoulders, who has a white triangle in their mind. They are looking out at a black circle, a black square and a larger white triangle just ahead of them. This indicates that the largest shape is the one that they think of. :link: hazards/reinforces-biases :link-type: doc + +Reinforces unfair treatment of individuals and groups. This may be due to for example input data, algorithm or software design choices, or society at large. :::: ::::{grid-item-card} Risk to Privacy @@ -91,6 +124,8 @@ Each hazard has: :img-alt: A red diamond shaped outline (like a warning sign) with a picture of a surveillance camera in the middle. :link: hazards/risk-to-privacy :link-type: doc + +This technology may risk the privacy of individuals whose data is processed by it. :::: :::::