So much of technology is solving the problem of inter-operating between disparate systems e.g. cpu architectures, operating systems, language runtimes.
Humans provide the greatest interop layer unknowingly. We seamlessly work between different technologies without missing a beat. We interoperate between our desktops and mobile devices, weaving together broken legacy systems at work to operationalize a task, and speak different languages (and computer languages). We manage an enormous amount of complexity without thinking about it.
Links to this note
-
AI Is the next Great Interop Layer
I had previously observed that humans are the great interop layer—we are the glue that fits together disparate processes and tools into usable systems. After using large language models, I’m becoming convinced that they can offload a large amount of the interop cost that currently falls to us. In a nut shell, AI can ‘do what I mean not what I say’ pretty darn well.
-
Tools for Thought Are Necessarily Monolithic
I read Unbundling Tools for Thought and I find myself agreeing and disagreeing. They are correct that people over invest in tools for networked thought and can get stuck in an unproductive cycle of rebuilding the system over and over again. They are incorrect that the solution is to unbundle into multiple tools and interop between them—point solutions are good at what they do but don’t provide as much building material needed to make it a home.
-
The Mundaneum Was a Paper Internet
In 1895, Paul Otlet and Henri La Fontaine built a paper internet with 18MM index cards affectionately referred to as the Mundaneum. They sought to catalog the world’s information (much like Google) but quickly ran into the physical limitations of such a thing. As more information was added, they realized the 15,000 drawers needed to hold all those index cards was never going to be enough.
-
On Mobile Safari, text inputs can not be
autofocusby design. Apple expects the user to initiate the input every time. -
Natural Language User Interface
One of the super powers of large language models is that it can “do what I mean” instead of “do what I say”. This ability to interpret prompts can drastically lower the barriers to accessing and interoperating between systems. For example, writing “Send a slack message to the Ops channel with a list of customers from HubSpot that signed up in the last week” would generate actions that query the HubSpot Contacts API, parse and extract the results, and make another API request to Slack to post to the
#opschannel. -
Most elements of our digital lives (e.g. apps) only exist within themselves and seldom work together. Because workflows are more useful than point solutions that leaves users with the burden of getting things to work together, humans are the interop layer.
-
Completing a tasks, documenting the problem and what you did to solve it serves as a kind of automation for the future. Anyone can read it and pattern match to the situation before carrying out the sames set of tasks to fix it again. This also ends up becoming research or a spec for how to automate the task programmatically (a higher order form of automation).
-
The Internet Is a Disjointed Memex
The memex device imagined a lattice of information that grows and can be built on top of incrementally. The internet and hypertext are that—nearly all the world’s information is now captured in the format of the web. However, it’s disjointed which makes it largely inaccessible.
-
Zapier NLA Is Bad at Generating API Parameters
I tried out Zapier Natural Language Actions API and found that it’s not particularly good for the one thing I needed it to be good at glueing together other APIs with natural language. API endpoints that are simple and straightforward are easy for large language models to generate the right inputs but more complicated (and poorly designed) like HubSpot are unusable.