In #327 @dan-blanchard says:

I did not write the code by hand, but I was deeply involved in designing, reviewing, and iterating on every aspect of it. [ . . . ] But the evidence here is clear: 7.0 is an independent work, not a derivative of the LGPL-licensed codebase. The MIT license applies to it legitimately.

The current LICENSE file containing the text of the MIT license is prefixed by "Copyright (c) 2024 Dan Blanchard"

Possibly I'm misunderstanding what was done here, but if the code was essentially rewritten by Claude, and leaving aside the issue of whether relicensing away from LGPL was legitimate, then I don't understand the theory under which it is being associated with @dan-blanchard's copyright and licensed under the MIT license. While the MIT license is a very permissive license, it does have some conditions that have force because there is a human author. Is the idea that the activity of "designing, reviewing, and iterating on" but not "writ[ing] the code by hand" was enough for @dan-blanchard to have copyright ownership? I'm not saying this is implausible but it might be itself breaking new ground in open source and I think it's important to have more clarity on this issue, given the controversy surrounding the relicensing.

If there are some specific literal parts of chardet that were not authored by Claude but were authored by @dan-blanchard maybe those should be identified somehow. Otherwise, perhaps it would be appropriate to remove the MIT license altogether.