Re: Why are commits on detached heads good, again?

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Feb 7, 2007, 15:32 UTC
Message-ID: <Pine.LNX.4.63.0702071629210.22628@wbgn013.biozentrum.uni-wuerzburg.de>
In-Reply-To: <20070207142642.23503.qmail@science.horizon.com>

Hi,

On Wed, 7 Feb 2007, linux@horizon.com wrote:

> Can someone explain why committing on top of a detached head is a killer 
> feature and thinking up a name sooner rather than later is a huge 
> inconvenience?

I agree that "because I could" is a weak excuse for this feature.

But actually, I think it is more convenient to be able to play around with an idea, properly committing simple steps, and finally, if, and only if, it works as expected, make a topic branch of it.

IOW:

checkout v2.6.20
<hackhackhack>
commit
<hackhackhack>
commit
<hackhack..
	Oh darn, it _cannot_ work after all, so let's forget about it.

No hard thinking of a proper name, no cleanup required.

Yeah, maybe not a killer feature, but maybe a butcher feature ;-)

Ciao, Dscho