Re: Breaking out of menu on "live disk", repairing grub

[Felix Miata]

On 2013-03-25 12:12 (GMT-0700) Jordan Uggla composed:

> > What I very well remember is it does everything I need it to do, and without
> > repeatedly warning me about deprecated cmdline arguments or stuffing dire
> > warnings about consequences of not installing to MBR in my face. And any

> Times change. Kernel arguments that were never really kernel arguments
> in the first place get deprecated (though it's still supported, just
> with a warning). Warnings about unreliable configurations are good
> (and the configuration was just as unreliable with grub legacy, just
> no warning).

Whether vga= was technically a kernel argument or not is orthogonal to the 
issue that on the cmdline is where it worked and works.

As I've been using it over a decade it has established itself as reliable 
enough for my and many others' needs. Thus the warnings are effectively rude, 
and rudeness is reason enough to avoid using it as long as a less rude or 
non-rude alternative that gets the job done exists, which in this case is true.

> > time I'm somehow faced with grub> instead of an expected menu at boot time,
> > I already know what commands do what in order to proceed, because they
> > weren't superceded during a massive code rewrite.

> Times change. Some of your knowledge of how GNU/Linux worked a decade
> ago no longer applies (even less of decade old knowledge of other IT
> sectors likely applies). I don't think it's a compelling argument to
> keep things the same when a new method, which is much simpler, can be
> devised instead.

You say simpler, but I don't see simpler. I seen an rpm roughly twice the 
size of the old. I see msdos where dos could have sufficed, and take offense 
that ms was unnecessarily prepended to a common string within a FOSS 
application. I see 0,1 has a different meaning than it previously had. I see 
the basic design made to inhibit concurrent support for both old and new in 
the same environment, forcing distros wishing to support both to change 
scripts, commands, docs and more to diverge from upstream.

That something newer might have advantages doesn't obviate the advantage of 
status quo that is spending 0 time to learn what's different, which often 
means unlearning something well ingrained and second nature. Keeping the old 
also means spending 0 time on replacing the old. On a system with one or few 
operating systems, changeover time can be nominal, but in an environment with 
many multiboot systems, changing provides elevated opportunity to encounter 
unforeseen and or unforeseeable conflicts or other time gobblers. What seems 
obviously simpler to you won't necessarily be seen the same by all.

Often times there are broader implications than those that are obvious. 
Decades ago I learned Lotus 1-2-3 for DOS, then migrated to Quattro Pro for 
DOS via its support of a 1-2-3 based menu system. QPro even today does what 
it did then exactly the same today, which is exactly what I need, including 
lack of need to adapt or unlearn just to keep on doing what I'm already 
doing. Modern spreadsheets may or may not be able to do as well or better, 
just like Grub2 may be able to do more or better, but the fact that they do 
their jobs without requiring _any_ investment in learning or unlearning is 
reason enough stick with the status quo. If and when someday something the 
old can't support becomes must have is soon enough to make the switch. By 
then maybe Grub2 will have stabilized and matured into a non-moving target 
with fewer docs that conflict according to distro source and/or release version.

Though the current Grub2 has a v2.0 moniker, I, like many, consider it v1.0 
caliber software, marginally out of beta. Reading here and elsewhere I'm 
familiar with various issues that frustrate users, both those who are new to 
multiboot and bootloaders and those familiar with them. Such writings 
reinforce my perception that maturity suited to my comfort has yet to be reached.

Automagic like grub-mkconfig isn't always what it's cracked up to be, 
particularly to those like myself who prefer working behind the pretty 
curtain and/or avoiding black-box processes. I'm dealing with enough alpha 
and beta software already, and choose to avoid this particular one, at least 
for the foreseeable future, possibly long enough that need here for any 
bootloader at all has expired.
--
"The wise are known for their understanding, and pleasant
words are persuasive." Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/