Many of the municipal websites use CivicPortal and other similar sites to host their content. CivicPortal for certain can be accessed directly by writing a small amount of code to download the data directly. I have done this to analyze town decision making over the last decade.
The AI is predominantly in the user interface layer where people ask about local questions and issues and then LocalMinutes interrogates the data or has downloaded.
In many ways, it is a good use of AI as it makes it easier for non-software engineers to get meaning from town government records and increases transparency.
Anne — thank you, this is exactly the kind of informed pushback that makes the reporting stronger, and you clearly know this terrain firsthand.
You're right that "scrapes" undersells it. Many Vermont towns host their minutes on platforms like CivicPlus, and that structured data can be pulled directly rather than crudely scraped — and as you say, the heavier AI lift sits at the interface layer, letting people who don't write code get answers out of records that were always public but rarely accessible. I've tightened that language in the piece to be more precise about the mechanism, and I appreciate you flagging it.
I'd gently draw one distinction, though. The article credits exactly what you're describing — it calls this a real transparency gain and a good use of AI, and on that we agree. Its question is narrower: when those same records are used to infer what communities *need*, the method inherits whatever the record leaves out — the people and towns that never make it into the minutes in the first place. That concern is about the leap from "what was discussed" to "what is needed," and it holds regardless of how skillfully the data is gathered.
But your point on mechanism is well taken. Thank you for reading closely and for the correction.
Thanks for correcting me on the portal name; I was typing from memory.
I think the term "scraping" can have a pejorative connotation and, depending on the
source, can violate the terms of service. LocalMinutes has been a helpful resource to track down meeting minutes that would otherwise require a lot of time and scrolling.
What is needed is higher-order analysis, which your article comments will be addressed using committees and other methods.
As someone who attends many town-related meetings (planning, DRB, selectboard), I observe that traffic is one of the big issues continually mentioned. I know this without AI, but I have logged many hours in local meetings. There are other benefits of attending these meetings—getting to know my neighbors and local officials—that cannot be obtained via AI. Different methods, different benefits.
To anyone reading, if your town is included in LocalMinutes, I encourage you to try it out.
I am not sure this article is entirely accurate.
Many of the municipal websites use CivicPortal and other similar sites to host their content. CivicPortal for certain can be accessed directly by writing a small amount of code to download the data directly. I have done this to analyze town decision making over the last decade.
The AI is predominantly in the user interface layer where people ask about local questions and issues and then LocalMinutes interrogates the data or has downloaded.
In many ways, it is a good use of AI as it makes it easier for non-software engineers to get meaning from town government records and increases transparency.
Anne — thank you, this is exactly the kind of informed pushback that makes the reporting stronger, and you clearly know this terrain firsthand.
You're right that "scrapes" undersells it. Many Vermont towns host their minutes on platforms like CivicPlus, and that structured data can be pulled directly rather than crudely scraped — and as you say, the heavier AI lift sits at the interface layer, letting people who don't write code get answers out of records that were always public but rarely accessible. I've tightened that language in the piece to be more precise about the mechanism, and I appreciate you flagging it.
I'd gently draw one distinction, though. The article credits exactly what you're describing — it calls this a real transparency gain and a good use of AI, and on that we agree. Its question is narrower: when those same records are used to infer what communities *need*, the method inherits whatever the record leaves out — the people and towns that never make it into the minutes in the first place. That concern is about the leap from "what was discussed" to "what is needed," and it holds regardless of how skillfully the data is gathered.
But your point on mechanism is well taken. Thank you for reading closely and for the correction.
Tom Davis - Publisher
Thanks for correcting me on the portal name; I was typing from memory.
I think the term "scraping" can have a pejorative connotation and, depending on the
source, can violate the terms of service. LocalMinutes has been a helpful resource to track down meeting minutes that would otherwise require a lot of time and scrolling.
What is needed is higher-order analysis, which your article comments will be addressed using committees and other methods.
As someone who attends many town-related meetings (planning, DRB, selectboard), I observe that traffic is one of the big issues continually mentioned. I know this without AI, but I have logged many hours in local meetings. There are other benefits of attending these meetings—getting to know my neighbors and local officials—that cannot be obtained via AI. Different methods, different benefits.
To anyone reading, if your town is included in LocalMinutes, I encourage you to try it out.