Having made great strides towards putting my law library onto AT, I am wondering whether my way of keywording is as efficient as it could be.
I have discovered that for my purposes Group works best on a single keyword. So I have a field labelled ‘Subject’ - (A) - which is customised to multiple select in which I have added as I go the paricular topic. For example, Alienation.
Alongside, I have a field label Keywords - (B) - also customised to multiple select in which I have added as I go the particular keywords: For example: Assignment, Underlease, Sub-let, Consent, Unreasonable.
So far so good? But I am wondering whether there is an easy (novice) way to have a single field which would still limit Group to the first keyword but avoid situations, for example, where the keyword would be the same as the subject or where I’m unsure which field to use for a particular keyword.
The examples above are straightforward, A has unique terms. The quandry arises with, for example, Costs. As a subject, Costs is embracing but I’d like Calderbank to be a Group so Costs would have to go into B. which means that when I filter for Costs I’d have to filter in both A and B.
I hope you can follow what I’m saying.
Is what I’m doing the only way or would it be possible for all keywords to be in one field, yet still being able to Group on one keyword?
Sadly, I can’t. I get parts of it, but the bigger picture still eludes me. However, I’m pretty sure I understand it enough to answer this:
That won’t work. The way that Airtable’s grouping feature operates, it builds groups based on the entire contents of a given field. It’s not possible to group based on a subset of a field’s contents.
For example, say your multiple-select field has just two options—A and B—and you have a mix of choices between those across your records: some with only A chosen, some with only B, and some with both. If you were to group by that field, you would get three groups:
Again, grouping operates based on the entire contents of a field. Because some groups have both choices, that combination is considered to be a distinct group compared to those with only one. With this in mind, you can see how the number of possible groups goes up rapidly the more options you have in your field setup.
In the end, it sounds like you have a clash between data organization and data reporting. While Airtable is great at giving you lots of organization options, its reporting options don’t always meet all needs. Unfortunately, I’m more of a formulas-and-general-organization guy and have very little experience setting up effective reporting systems, so I’ll gladly step aside if anyone else has ideas on how to achieve what you want. As a heads-up, it will most likely mean using third-party tools for the reason stated above, though there may be some blocks that could help.
Before I posted I knew that Group picks up the entire contents of a field so it wasn’t really necessary to explain in detail as you have done why the answer to my question is no, what I’m doing is the only way. But thank you for the taking the time to do so. Also your detailed explanation is sure to be help to others.
Um, okay - I’m not following entirely either, but if we can understand that that the above statement encapsulates the deeper requirement, perhaps it will help to expand on this before chosing an implementation approach.
Ideally, keywords and terms should be managed in a single field.
Keywords can be used without constraint to create filters and views despite being managed in a single field.
The system must be able to create dynamic views group by targeted keywords.
Is it safe to say that the need for a view grouped by a specific keyword is the result of a transient (ad-hoc) process? (i.e., I’d like to see the data grouped at this moment by keyword “x”, whereas you might switch to keyword “y” later)
Assuming these are the general requirements, it suggests you need a way to identify a “prime keyword” and allow grouping to be based on that selection.
To achieve this imagine a formula (field) that returns a computed value if the prime keyword exists in the keywords collection for any given record. It returns a different computed value if the keyword is not present in the list. Now you’d have a field could be used to establish groups.
The drawback of this is that to change the prime keyword - i.e., the targeted keyword for grouping purposes - you’d have to adjust the formula; not a pretty solution. However, if this “novice” method works, you could build the prime keyword selection process into a Script Block.