Greg A. Woods wrote:

> Huh?  Why do you think so?  At worst the scripting feature can be used
> as a learning tool so that even a non-expert can quickly get a starting
> base upon which to learn how to do a non-standard install.

As I said: Mostly useless. :) Good for teaching and that's about it.

> Not, "sysinst", specifically, should not.

Yes, "sysinst", specifically, should. Just to be equally zealous.

> You're making the assumption that there's only one installer -- perhaps
> because you haven't fully embraced the software tools philosophy
> inherent in all unix-like systems.

Where, sir, did I make that assumption?

[stray point ommited]

>>Who says I wanna do the whole thing by hand? I just wanna twist enough 
>>of the knob so that it can get around a problem and then let it do it's 
>>thing.
> 
> Who says so indeed -- that's just the way of the world.  If you want to
> go against the grain (or jump out of the groove, to use the metaphor I
> was using before), then you have to be prepared to do what it takes.  :-)

Sounds to me like sysinst is the thing going against the grain here, 
though I could be wrong...

> Mavericks don't get to be mavericks by resting on their laurels.

No.... hey, what does that have to do with the topic? :) Wasn't the 
topic "sysinst needs a knob to to allow over ride of 1 specific sanity 
check that is stricter then the kernels version"?

>>Why should I manually do everything when there is this elegant 
>>program that will do it for me, except for while it's barfing on this 
>>problem that it won't let me help it with because it doesn't have an 
>>expert knob??
> 
> 
> Why indeed?   Why don't you write your own version with your own rules
> and philosophy and just be happy?  Perhaps your version will become more
> commonly used by others if you publish it.  :-)

Because it should have an expert knob that can be twisted by default.

> (FYI I have made my own significant changes to sysinst, though I
> certainly didn't turn off any sanity checks -- in fact I seem to
> remember adding even more.)

Okay, so you've created a "protect myself from me" mode. That's spiffy, 
package it in and include a knob to turn it on.

>>I would beg to differ on that. Yes, its important, but so are a lot of 
>>other things. What is ultimately harmed if we have an expert knob that 
>>allows us to ignore the fact that the sysadmin who obviously should have 
>>as much rights to the system as the kernel (remember, the sysadmin is 
>>trusting the kernel to handle the second to second tasks in his 
>>computer) feels like setting up something that _looks_ bizarre, but 
>>_might_ work given circumstances outside our programs range of 
>>comprehension?

[zealot holier then thou comments deleted, since they didn't answer the 
question]

> "sysinst" really does not need an infinite number of monkeys adding an
> infinite number of knobs to it, no matter how expert those monkeys
> really are.

No, it just needs a knob that allows an infinite number of monkies to be 
able to say "Yes, i'm an idiot, let me be one."

> If you want so much rope, well then twist it up for yourself.

Why? There's plenty of ropes already sitting around. Heck, seems you 
needed to twist up some extra to tie your hands a little extra firm in 
your personal sysinst derivative.

> Indeed I am.  These kinds of "non-expert" tools obviously have their
> place.  "sysinst" is a tool that has pre-determined expertise built into
> it.  If some other self-proclaimed expert has different ideas about how
> it should work, or who wants to break the rules other experts have
> implemented just because he or she can, then that other person can write
> their own new tool that does exactly what they want and leave the
> "sysinst" as it is with the common rules it needs.

Like sanity checks beyond those of the kernel? Doesn't sound to me like 
that is actually needed. Though I could be deaf. :)

>> That's a problem.
> 
> 
> What really irks me the most is that those of you who are arguing for
> "the expert knows best so shut up and do it" kinds of knobs are
> currently here in this thread basing your argument on a problem that
> _nobody_ has yet been able to justify with anything other than very
> poor and _very_ lame excuses.  Nobody has managed to even suggest a
> valid reason for generating overlapping paritions with "sysinst" and yet
> some of you cry foul just because you think your perceived freedoms are
> being taken away from you.  Well they're not -- you're free to modify
> the source or use other tools or do it by hand.  Get over it!

And you sir have not been able to justify the lack of such knobs. You 
have yet to properly justify your reasoning with anything but equally 
poor and lame excuses, if you have even bothered to do that. I have seen 
several valid reasons posted for overlapping partitions, but you sir 
seem unable to read them. Personally, I haven't had occasion to use 
them, but i've had other issues where it would have been nice to have a 
sysinst knob for it.

> If we were talking about other policy kinds of things instead of
> low-level sanity checks, such as whether "sysinst" should default to
> offering a separate /usr partition or not, then maybe your desire for a
> new knob to twist would hold some merit.  However this idea that sysinst
> should have some knob to turn off its low-level partitioning sanity
> checks is just so completely bogus and outright baseless as to be
> laughable!

And your lack of being able to understand why it does is equally laughable.

>>I know what sysinst does, I could do it on 
>>my  own, but why should I have to??? It's a perfectly good tool at doing 
>>what it does, save for it does occasionally need an expert knob to help 
>>it around a problem, or something bizarre that I happen to be doing but 
>>would like it's assistance with so I can go to lunch while it does things.
> 
> 
> THERE IS NO PROBLEM HERE!  Look at what you're asking for!  You're
> asking for some way to turn off sanity checks that _almost_ everyone,
> expert and non-expert alike, agrees are fundamental to the creation of a
> sane, safe, maintainable, and secure configuration.

Er, I could be wrong here, but in the last couple of hours i've seen 
only you defending these over zealous checks. Granted, these checks are 
needed, but should not be forced policy. I've not seen anyone asking for 
them to be turned off completely, just that they be given the option to 
say "yes, i'm doing something stupid".

[more zealot ranting deleted]

> You should feel lucky that you don't have to hexedit your overlapping
> partitions into the binary label yourself, but you also shouldn't
> complain if you had to.  After all you have the source and if you're
> expert enough then you can comment out any and all sanity checks that
> you wish -- and why don't you delete all those pesky cycle-wasting error
> checks while you're at it since you know how your hardware and your
> software is working and you'll know when something fails without having
> all those annoying hand-holding reminders!

I find your lack of ability to interact reasonably without jumping to 
invalid conclusions about what has been asked for in plain english, and 
then make erroneous expansions of peoples intent quite... childish.

Why should I feel lucky? Fact is I don't have to. Tools have been 
written. I'm thankful to those who wrote the tools, and wrote them 
correctly with enough of the right kinds of knobs to do everything that 
I need to.

Just because I can hex edit a disk (i've done it, though not for a disk 
label - had to break into a novell server I forgot the pass to) doesn't 
mean I know enough to program... I personally avoid programming, don't 
like it, and as such don't have much ability to program in knobs that 
programmers didn't include.

And just because an over sensitive disk sanity check is annoying and 
should be over ridable, doesn't mean some other sanity checks that are 
properly tuned should be removed. I haven't even read anyone asking for 
them to be - have you? Quote and reference.

--Gerald