On Tue, 2005-04-19 at 13:24, Chris Adams wrote: > If you can cite _specific instances_ of GRUB's alleged unreliability > that have not been addresses in the current version, please open > Bugzilla cases. Otherwise, your claims are baseless. Chris, In software development there's difference between a long-term lack of reliability and a repeatable failure mode. Both are bugs. If <insert politician's name here> stands up and says that nobody has prove that he lied today, that doesn't make him trustworthy. Grub has not been reliable. It has just been modified to support MD, except that nobody seems to know exactly how and there's no documentation and the developer won't tell us. Some people on this list expected it to write to the MBR by default, some expected it to not change the MBR without explicit permission. Why are we forced to playing guessing games with Grub instead of using a reliable alternative? If Redhat wants to keep Grub as the default, that's fine by me. But why discard the reliable alternative for those who need it? The cost to Redhat of keeping Lilo is negligable CD space, negligable overhead in Grubby, negligable overhead in Anaconda UI, negligable maintenance. It's all about developer ego. Redhat VP Michael Tiemann said that he "believes customers should be directly involved in designing and creating products from the earliest stages." That's what we're trying to do here. Tiemann said that in the past Redhat has "insulted some of our best supporters" and that things were going to change. Perhaps he forgot to tell Redhat developers. --Mike