Granting new feature requests, previously the domain of technology vendors, may change dramatically with the advent of AGI. Here's a simple example.
A formula that returns a comma separated list of all values across records in a view for a given field:
This is a good feature request. I made a similar request in 2018 for Split(). That was 2,160 days ago. Still waiting. I hope Stephen is more successful in getting Airtable’s attention.
What he’s asking for is tantamount to a Parquet data format, which is one of the fastest file types to read generally and much faster than either JSON or CSV. It’s like a pivot table - fields are listed down the data structure, and columns are represented as arrays for each field. Stephen only wants a single column’s values as an array, and Airtable should accommodate this.
The advantages of a Parquet data format are massive, and it’s no coincidence that Pandas data frames are fundamentally able to read and write this format. It puts the science in data science for these reasons:
Enough about Parquet. Here’s why it’s relevant to Stephen’s feature request.
Given a table like this, he wants a column transformed into an array.
"Name": ["Tesla", "CNN", "Airtable", "SalesForce", "IBM"],
"Stock Ticker": ["TSLA", "CNN", "ATB", "CRM", "IBM"],
"# Employees": [70757, 4000, 1000, 100000, 352000],
"Market Cap": [750000000000, 10500000000, 1050000000, 260000000000, 118000000000]
How’d I get this? AI.
By simply transforming the Airtable table into Markdown, I had the Parquet table in a few seconds.
I no longer want Split(). I want a feature to perform inline inferencing against a blistering fast LLM such as PaLM 2.
Let’s take the list of values we created for Stephen’s case and parse it into an actual array.
Certainly, there are many ways to extract these values, either through a script or formula if Airtable gets around to it. This article is not about any particular methodology for managing or working with data. Rather, it’s about the possible future of platform features.
This example demonstrates that vendors should think carefully about how to implement new features because it may be possible to end the feature request madness by simply adding one feature - ai().
Can it be that simple? Hardly. However, developers and platform vendors should consider how AGI will transform architectures in the future.
Hi, I am trying to figure out, in a very basic way, how AI will impact my use of Airtable outside of creating briefs (which as of right now is the only example I've been able to find of its usage). Maybe I'm posting this reply in the wrong spot (and I'll be copying and pasting this in other areas just in case), but how is (if it actually can) AI going to work with my automations, with my formulas and sync tables, with creating a record and linking the record, etc....the actual basics of Airtable. I am not a developer and maintain my companies Airtable setup with the limited knowledge I have and find it increasingly frustrating in how complicated it can be. I also find it so limiting for such basic functions with no easy workarounds. Can anyone shed some light on this for me?
Your question is quite valid in this thread. My post attempts to help others see that historically challenging features can be addressed by AGI and it makes sense to keep AI resources top of mind. This reflects on a new movement occurring in development circles known as AI-First development. Here's an article demonstrating another case involving data cleansing of names.
I'm currently working on an approach that uses AI to spot fraud in Airtable transactions. And I have another project that eliminates the need to use geo-encoding APIs to map locations. The possibilities are endless.
I agree with you - once you get past the common use cases that everyone has raved about, the applicability of AGI seems narrow. I believe AI will seep into many areas of data management, not just content development. And once it does, it won't be called AI, nor will we give it a second thought. That's when we'll know AGI is mainstream.
Thus far, Airtable hasn't shown much about the deeper and more productive use cases as AI pertains to data. They don't seem much interested in such uses, but they may over time. That leaves only the scripting environments to take advantage of AI possibilities.