indeyets

Has strong TOR (Onion Routing) feeling

yes. less stable. more hallucinations

Doesn’t that defeat the purpose? You want to control it from outside of the sandbox, not to give agent escape hatch from sandbox

You mean late 1990’s? :)

They’re victims of agenda

Nats is getting there, but not yet.

Redis is still much more powerful: lists, sorted sets and bazillion of other data structures

It’s usually a question of “who are your users?”

As long as you consider developers of derived software to be your users — permissive makes most sense.

But if you consider end-users of software it’s definitely copyleft.

So, the browsers have to provide some means for choosing the desired translation engine (add-on API maybe?) and this is a standard API which all of the providers should implement.

right?

Is it the UI for penthouse lib? Settings look very similar :)

> Usually new data is generated regularly

This part was not obvious. In a lot of cases geodata is mostly stable and reads/searches dominate over appends. And that’s why we keep this in DB (usually postgis, yes).

So DuckDB is optimised for very different use case and it is not always obvious when it’s mentioned

But LLMs are not reliable enough, so you can not actually expect “specificity”

Well, jj is actually very nice in this regard, because:

1. it works on top of git — you keep using all the same infrastructure (GitHub, etc.) 2. in your local repositories you still have access to git-tools, as jj maintains git and its own states in sync.

After all of these months with jj I still find myself using GitUp when I need to review a long chain of commits or do some quick repo-archeology. And, from time to time, I even use Idea's merge tools because their "magic wand" saves a lot of time.

I mentioned this in the first part of series