Documentation: Increase example cache timeout to 1 hour #1412
Conversation
|
There are issues in commit 30f8262: |
hickford
force-pushed
the
nudge-cache-longer
branch
from
30f8262
to
c11f094
Compare
|
There are issues in commit c11f094: |
hickford
force-pushed
the
nudge-cache-longer
branch
from
c11f094
to
6c35d68
Compare
|
/submit |
|
Submitted as pull.1412.git.1667989181611.gitgitgadget@gmail.com To fetch this version into To fetch this version to local tag |
|
On the Git mailing list, Jeff King wrote (reply to this): On Wed, Nov 09, 2022 at 10:19:41AM +0000, M Hickford via GitGitGadget wrote:
> From: M Hickford <mirth.hickford@gmail.com>
>
> Previously, the example *decreased* the cache timeout compared to the
> default, nudging users to make cache less usable.
I don't mind at all changing this as your patch does. The existing
example was mostly just to illustrate the syntax. But...
> Instead, nudge users to make cache more usable. Currently many users
> choose store over cache for usability. See
> https://lore.kernel.org/git/Y2p4rhiOphuOM0VQ@coredump.intra.peff.net/
I don't see how my email argues for this. The only thing I mentioned
about credential-cache there is that it's not available on all
platforms.
But if you want my opinion on its usability, the main problem is not
that the cache timeout. It's that entering the credential at all is a
pain, either because it's a semi-automated environment that needs to
operate without user input, or because the credential itself is awkward
for the user to enter (like a long token). And that's what pushes people
to "store" over "cache".
> diff --git a/Documentation/git-credential-cache.txt b/Documentation/git-credential-cache.txt
> index 0216c18ef80..432e159d952 100644
> --- a/Documentation/git-credential-cache.txt
> +++ b/Documentation/git-credential-cache.txt
> @@ -69,10 +69,10 @@ $ git push http://example.com/repo.git
> ------------------------------------
>
> You can provide options via the credential.helper configuration
> -variable (this example drops the cache time to 5 minutes):
> +variable (this example increases the cache time to 1 hour):
>
> -------------------------------------------------------
> -$ git config credential.helper 'cache --timeout=300'
> +$ git config credential.helper 'cache --timeout=3600'
> -------------------------------------------------------
The patch itself is obviously correct.
-Peff |
Previously, the example *decreased* the cache timeout compared to the default, making it less user friendly. Instead, nudge users to make cache more usable. Many users choose store over cache. https://lore.kernel.org/git/CAGJzqskRYN49SeS8kSEN5-vbB_Jt1QvAV9QhS6zNuKh0u8wxPQ@mail.gmail.com/ The default timeout remains 15 minutes. A stronger nudge would be to increase that. Signed-off-by: M Hickford <mirth.hickford@gmail.com>
hickford
force-pushed
the
nudge-cache-longer
branch
from
6c35d68
to
e3e0437
Compare
|
/submit |
|
Submitted as pull.1412.v2.git.1668010273573.gitgitgadget@gmail.com To fetch this version into To fetch this version to local tag |
|
On the Git mailing list, M Hickford wrote (reply to this): On Wed, 9 Nov 2022 at 13:18, Jeff King <peff@peff.net> wrote:
>
> > Instead, nudge users to make cache more usable. Currently many users
> > choose store over cache for usability. See
> > https://lore.kernel.org/git/Y2p4rhiOphuOM0VQ@coredump.intra.peff.net/
>
> I don't see how my email argues for this. The only thing I mentioned
> about credential-cache there is that it's not available on all
> platforms.
I'll amend the commit message.
> But if you want my opinion on its usability, the main problem is not
> that the cache timeout. It's that entering the credential at all is a
> pain, either because it's a semi-automated environment that needs to
> operate without user input, or because the credential itself is awkward
> for the user to enter (like a long token). And that's what pushes people
> to "store" over "cache".
I agree. I have some ideas to help human users; I'll share in the other thread.
> The patch itself is obviously correct.
Thanks for the review |
|
On the Git mailing list, Taylor Blau wrote (reply to this): On Wed, Nov 09, 2022 at 04:09:14PM +0000, M Hickford wrote:
> > The patch itself is obviously correct.
>
> Thanks for the review
Thanks, both. I picked up the updated round.
Thanks,
Taylor |
|
User |
|
This branch is now known as |
|
This patch series was integrated into seen via git@24e4388. |
|
This patch series was integrated into seen via git@76850f9. |
|
This patch series was integrated into seen via git@0cd07d5. |
|
This patch series was integrated into next via git@afe54db. |
|
This patch series was integrated into seen via git@645f482. |
|
This patch series was integrated into seen via git@be2fc89. |
|
This patch series was integrated into seen via git@86cacd7. |
|
This patch series was integrated into seen via git@7718b96. |
|
This patch series was integrated into seen via git@78fcc69. |
|
This patch series was integrated into seen via git@96b4656. |
|
This patch series was integrated into seen via git@3f98d7a. |
|
This patch series was integrated into master via git@3f98d7a. |
|
This patch series was integrated into next via git@3f98d7a. |
|
Closed via 3f98d7a. |
Previously, the example decreased the cache timeout compared to the
default, nudging users to make cache less usable.
Instead, nudge users to make cache more usable. Currently many users
choose store over cache for usability. See
https://lore.kernel.org/git/Y2p4rhiOphuOM0VQ@coredump.intra.peff.net/
The default timeout remains 15 minutes. A stronger nudge would
be to increase that.
Signed-off-by: M Hickford mirth.hickford@gmail.com
CC: Jeff King peff@peff.net
cc: Taylor Blau me@ttaylorr.com