- we learn to use a tool by watching other humans use the tool as well as trying to use it ourselves. this is why I love pair programming with other developers. I learn far more by watching the way they interact with their machine and structure their work and think through problems in a whole-life way. see the apprentice/journeyman trades system in history, etc.
- digital spaces make it difficult to learn from others by watching. we have our own private glowing rectangles, isolated in our own rooms. even youtube videos of people working are highly edited / curated, not truly useful for learning how to use the tool.
- this is similar to ethnographic research. sitting with users while they live their lives and the tool usage takes place in context. there are some good works on this sort of thing in anthropology / sociology.
- basically, every product team that actually does their job well inherently does this. they enter into the WORLD of their user and place their software in context. unfortunately, this excellent research is trapped inside of companies in their internal wikis, their private video archives. for example, the Muse team did extend user research before building the first version of Muse. if only this research could be shared!
- We started experimenting with "workflow walkthroughs" during Tools for Thought Rocks sessions, where someone spends 5-7 minutes showing off how they execute a specific workflow, such as "taking notes in a meeting" or whatever. It's not feature focused or tool focused, it's the workflow, which usually consists of one or more tools used in tandem. It was enlightening to see what people actually did! But the default was to slide into describing "features" of the tool - which is not that helpful. "Let me watch while you do it."
I would love to somehow see this idea of a shared repository of workflows come to light! I think we could all stand to learn from it, whether you're a tool user, a toolsmith, or a researcher.
Even though you mentioned it wasn't what you were suggesting, there's absolutelly also something to be said for shareable workflows for users between tools (portability and discovery)!
What you say is so true, sharing between apps would also be very useful ( but a very large technical challenge). Let’s hope we get there one day. Tnx for writing.
In other industries, such as Built infrastructures domains, the sector struggles with interoperability for many years. In order to solve it, they've spent more than 25 years to develop what is called "Industry Foundation Classes" (IFC, https://www.buildingsmart.org/standards/bsi-standards/industry-foundation-classes/), which is an open Data Schema to transport structured data from one system to another. This is part of a bigger set of standards and services framed by standardization body called buildingSMART International https://www.buildingsmart.org/standards/bsi-standards/.
Related to this discussion, even assuming this is a distant domain topic, there is the bSI service called "UCM, Use Case Management tool" https://ucm.buildingsmart.org/use-case-management. The main three stages are:
- Use Cases: define propose and the scope for the information delivery.
- Process definition: is made through at least one of the following: Process map / Interaction map / Transaction map
- Information delivery exchange requirements in non-technical format.
Now let's try to build this for the Knowledge domain. We exchange information from many subjects (we could call it a multi-domain ecosystem). Some systems rely on text-based information (proprietary or open), non-structured or partially structured; others rely on proprietary structured data stored in database format.
We already have those Use Cases. We would need to define an standard way to define them.
Regarding the Process Definition, we have also many workflows, but we lack a standardized way to represent them. (i.e. ISO 29481-1:2010)
And finally, we would need to exchange our information based on specific Information requirement that we as common end-user can query using natural language but there is some system parsing it using a machine-readable structured language enabling to connect both sides of that information trade.
I love this concept! Several threads here:
- we learn to use a tool by watching other humans use the tool as well as trying to use it ourselves. this is why I love pair programming with other developers. I learn far more by watching the way they interact with their machine and structure their work and think through problems in a whole-life way. see the apprentice/journeyman trades system in history, etc.
- digital spaces make it difficult to learn from others by watching. we have our own private glowing rectangles, isolated in our own rooms. even youtube videos of people working are highly edited / curated, not truly useful for learning how to use the tool.
- this is similar to ethnographic research. sitting with users while they live their lives and the tool usage takes place in context. there are some good works on this sort of thing in anthropology / sociology.
- basically, every product team that actually does their job well inherently does this. they enter into the WORLD of their user and place their software in context. unfortunately, this excellent research is trapped inside of companies in their internal wikis, their private video archives. for example, the Muse team did extend user research before building the first version of Muse. if only this research could be shared!
- We started experimenting with "workflow walkthroughs" during Tools for Thought Rocks sessions, where someone spends 5-7 minutes showing off how they execute a specific workflow, such as "taking notes in a meeting" or whatever. It's not feature focused or tool focused, it's the workflow, which usually consists of one or more tools used in tandem. It was enlightening to see what people actually did! But the default was to slide into describing "features" of the tool - which is not that helpful. "Let me watch while you do it."
I would love to somehow see this idea of a shared repository of workflows come to light! I think we could all stand to learn from it, whether you're a tool user, a toolsmith, or a researcher.
I love this idea of shareable workflows.
Even though you mentioned it wasn't what you were suggesting, there's absolutelly also something to be said for shareable workflows for users between tools (portability and discovery)!
What you say is so true, sharing between apps would also be very useful ( but a very large technical challenge). Let’s hope we get there one day. Tnx for writing.
In other industries, such as Built infrastructures domains, the sector struggles with interoperability for many years. In order to solve it, they've spent more than 25 years to develop what is called "Industry Foundation Classes" (IFC, https://www.buildingsmart.org/standards/bsi-standards/industry-foundation-classes/), which is an open Data Schema to transport structured data from one system to another. This is part of a bigger set of standards and services framed by standardization body called buildingSMART International https://www.buildingsmart.org/standards/bsi-standards/.
Related to this discussion, even assuming this is a distant domain topic, there is the bSI service called "UCM, Use Case Management tool" https://ucm.buildingsmart.org/use-case-management. The main three stages are:
- Use Cases: define propose and the scope for the information delivery.
- Process definition: is made through at least one of the following: Process map / Interaction map / Transaction map
- Information delivery exchange requirements in non-technical format.
Now let's try to build this for the Knowledge domain. We exchange information from many subjects (we could call it a multi-domain ecosystem). Some systems rely on text-based information (proprietary or open), non-structured or partially structured; others rely on proprietary structured data stored in database format.
We already have those Use Cases. We would need to define an standard way to define them.
Regarding the Process Definition, we have also many workflows, but we lack a standardized way to represent them. (i.e. ISO 29481-1:2010)
And finally, we would need to exchange our information based on specific Information requirement that we as common end-user can query using natural language but there is some system parsing it using a machine-readable structured language enabling to connect both sides of that information trade.
This is explorative effort.