Yes, yes, but if there is no net gain then you revert to simply sending out the whole rpm. But when you have some 100 updates or more then you want to send this Fedora user the smallest possible patch you can. It should be a crazy challenge to come up with the most minute update, the modicum of all updates, that can be achieved. Sure, repositories will have to grow up and embrace a new system. But with the Web 2.0 coming out everything is slowly changing anyways. This must be done. Who wants to start? Who could start? Benjy, AWWTF <div>On 11/28/05, Brian Gerst <<a href="mailto:bgerst@didntduck.org">bgerst@didntduck.org</a>> wrote:<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> Benjy Grogan wrote: > Please, this is ridiculous. Of course it's possible. You make a > decision: i) will a few patches be small/easy to install ii) should i > generate a unified patch for this user iii) situation is so bad, just > send a whole new rpm... and there the repository has done some work ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Therein lies the problem: this requires more than a simple web server to implement. > and > come up with the best solution for you. ... and yes, there are many > more decisions.. such as cross-package decisions etc... but this is an > absolute must for the next gen linux distros. > > What's Redhat doing? What's Novell doing? > > Microsoft has their solution We don't need to repeat their mistakes. Most RPMs are small enough that there is no net gain using this method. -- Brian Gerst -- fedora-devel-list mailing list <a href="mailto:fedora-devel-list@redhat.com">fedora-devel-list@redhat.com</a> <a href="https://www.redhat.com/mailman/listinfo/fedora-devel-list"> https://www.redhat.com/mailman/listinfo/fedora-devel-list</a> </blockquote></div>