Re: Recommend these .gitconfig settings for git integrity.

[Paul Eggert]

On 02/03/2016 08:20 AM, Óscar Fuentes wrote:
> Instead of making up reasons and proving yourself right, the
> reasonable thing to do is to ask the git maintainers.

I'd rather not waste their time on this unless the concerns are 
warranted. We would ill-serve Emacs development by giving ourselves a 
reputation for cluelessness. Let's wait for real problems to turn up 
before ringing alarm bells elsewhere.

> It either works or doesn't.

It does both, and we're concerned about the part that doesn't work. A 
client can operate for a while with tainted objects, and do (say) a 
rebase that computes new checksums for the bad data, and then ship the 
tainted data off to the server with the newly computed checksums. The 
server can't detect the problem because the recomputed checksums match 
the tainted data.

> in the future, I would appreciate if you discuss a change like this 
> before rushing to commit it and cause irreversible consequences. 

The consequences are largely benign and entirely reversible.

I often install changes to Emacs, and obviously can't clear each change 
before installing it, as that would slow us all down for little benefit. 
So I use my best judgment to decide which changes to install in master, 
which in emacs-25, and which to discuss first. I sometimes make mistakes 
in that area, for which I apologize, but what can I say? The only way to 
prevent all mistaken changes is to not make any changes. This change 
seemed like a no-brainer at the time, but we've decided to take a more 
cautious approach for now. Such things happen, and life moves on.