More of the same??? On Tue, 2005-04-19 at 23:15 -0700, Mike Bird wrote: > On Tue, 2005-04-19 at 22:40, Rodd Clarkson wrote: > > until you can give the developers something concrete with which to > > work all the other comments about this aren't going to amount to squat. > > Rodd, > > Grub's DEVELOPERS have already abandoned Grub. Grub is legacy, > obsolete, bereft of life, resting in peace. Grub is only here because > some MAINTAINER at Redhat nailed Grub to the perch. This might actually be a specific example of why fedora should look elsewhere other than grub. However, since the one of the benefits of open source is that you don't need to rely on the original developer of a product for ongoing support, and given that RH take bug reports for grub and someone in RH seems to be actively developing grub, I would appear that grub isn't exactly the abandonware you claim it to be. > What is wrong with Grub is that Grub is not as reliable as Lilo. That > is why Lilo is needed for serious work. Software developers understand > that reliability is important. System administrators understand that > reliability is important. Users understand that reliability is > important. Business managers may not know how to change their > screensavers but even they understand that reliability is important. I appreciate reliability too and like many, many people using FC and RH, I find grub to be both reliable and pleasant to use. However, this doesn't change the facts that these are all generalities and that they don't help isolate what you claim is so wrong with grub. > Simple test cases are great but not always possible. Repeatable failure > modes are not the only kind of bugs and recoding is not the only kind of > solution. Sure, simple test cases aren't always possible. But since you are so adamant that grub is so shit, let me tell you two obvious (and specific examples) you could give about grub that the developers will listen too: 1. grub can't do <something specific>. In this case you explain that you want to be able to do <something specific that lilo can do> and that grub should be able to do this. 2. grub failed and you had to take the following steps to get grub working again. This may not highlight what went wrong, but it shows that grub failed and the action taken to solve the problem may well help in detecting why it failed. Both these example are specific, and are able to be assessed on their merits. Rodd > > The solution here is easy, and doesn't require any coding. Redhat needs > to put Lilo back in Fedora Core is all. > > --Mike Bird