/usr/share/man/man8/MAKEDEV.8

[Alex Zbyslaw]

Matthew D. Fuller wrote:

>On Thu, Nov 01, 2007 at 01:43:41PM +0100 I heard the voice of
>Bernd Walter, and lo! it spake thus:
>
>>Show me the positives that outweights the negatives and I'm on your
>>side.
>>
>
>Why do you think we're on different sides to begin with?  I've not
>advocated removing catman capability, or denied that different
>situations have different needs, or that those needs may include
>catman, or for that matter said anything at all applicable to tiny
>environments or appliance building. 
>
As the author of the original statement creating this little furore, I 
don't think I said anything of the kind either! 

It's kind of depressing that a throwaway comment at the end of an email 
gets turned into a straw man argument that even I could have knocked 
down, but no-one has bothered to comment on the, to my mind, clear logic 
error in man/catman which allows the wrong manual page to get displayed 
in the first place.

When the cat-ed man page concept was invented (when even the slowest 
machine cited here would have looked like science fiction) the notion 
that an upgrade process could install a manual page older than a 
pre-existing cat-ed version would have been unthinkable.  With FreeBSD 
as it is now, it clearly isn't unthinkable since two experienced 
upgraders already had their own personal steps in the upgrade process to 
avoid the issue.

At the very least, it would seem like adding that step to 
/usr/src/UPDATING and the handbook would be in order, but fixes to man 
and catman could make this issue happening a near impossibility.  (A 
scheme using something like md5 sums could make it even more improbable, 
but hardly seems worth the effort).

In addition, considering an *option* to simply not have cat-ed manual 
pages (for people with machines fast enough to just not care, or who 
have machines where you just don't read man pages often enough to care) 
does not seem out of order.  An option.  Not behaving like Microsoft.  
Not discriminating against anyone.

On Thu, Nov 01, 2007 at 01:43:41PM +0100 Bernd Walter said:

>My point was against retirement, which was mentioned by Alex.

No, it was not.  I said:

>Of course, with modern systems where nroff-ing a man page takes 
>negligible time and system resources, it could also be argued that 
>cat-ed man pages should be a thing of the past
>

"with modern systems where nroff-ing a man page takes negligible time and system resources".  All your arguments have been about systems which do not, by any stretch, meet that very specific criterion, and, for which, I therefore advocated nothing at all.  If you had wanted a clarification, you only had to ask.

jonathan michaels wrote:

>sorry for my noise, i am not complaining rather asking for a bit of
>thinkings and for some tolerance fro people who still use "old"
>machines ..

I'm sorry that you've seen intolerance, but I assure you that none was intended.  I think you are reading far more in to what was said, than was actually said.

--Alex