[CrypticShift]

Previous (recent) discussion : https://news.ycombinator.com/item?id=36799548

Commendable breadth in your vision ! but might I suggest going more... atomic? As someone who leans towards outlining, your "tasks" type caught my attention. Ideally, I wish each block could behave like an object in a similar fashion. While transforming blocks into objects is possible, it somewhat defeats the purpose of an outliner's "bird's eye view."

I also doubt the feasibility of centralizing all your personal knowledge in one app; there's a corollary to this. I consistently find myself in need of a "merge feature." The larger the database becomes, the more crucial this feature is.

[pronkin]

We definitely plan to enable effortless transformation of any block into an object, for example, by adding a tag to it or simply by mentioning it in another object 'transclusion'.

We also have plans to develop a simpler local API and a local AI helper that will be able to assist you with import/export tasks in ways you might think of

[orthecreedence]

What do you mean by a merge feature?

[CrypticShift]

I mean that when I import a new CSV, I sometimes want to merge its rows with an existing Anytype database objects if key columns are identical.

For larger personal databases, I find that I often need to clean them up or enrich them with web APIs. Therefore, I will export the data, manipulate it elsewhere, and then reimport it. This requires the data to be "merged" with the existing records in order to preserve the existing IDs, links, and so on.

[orthecreedence]

Thanks for explaining, that's a totally valid use-case.