On Mon, 19 Sep 2005, Jeffrey Altman wrote:

> We know why the headers are not shipped.   The Sun krb5 implementation
> does not match the MIT implementation it is derived from.  They made
> changes to the internal data structures.  They are not comfortable
> enough with the changes to ensure that they won't change from release
> to release.   Since Sun does not change public ABIs from release to
> release they cannot open the APIs.
>
> Sun is committed to opening the krb5 implementation once there is an
> agreement on a standard krb5 api that can be safely implemented in
> multi-threaded environments with pluggable crypto implementations.

I did not mean to point an accusatory finger at Sun there.  Perhaps
I should have said, "for reasons unknown to me".  I applaud Sun for
not shipping the headers in the case you've outlined; I value the
ABI stability that Solaris provides.  Thanks for explaining why
those decisions were made.

--
Coy Hile
[EMAIL PROTECTED]
_______________________________________________
OpenAFS-info mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-info

Reply via email to