Questions about the Jasper page

[Randy McMurchy]

Ken Moffat wrote these words on 02/24/09 16:54 CST:
> On Tue, Feb 24, 2009 at 12:08:17PM -0600, Randy McMurchy wrote:
>> Why are the dependencies of Jasper "recommended" instead of "optional"?
>>
>  To me, they look like a good idea.

This is a technical book. Could you provide some technical details
why you think they should be recommended? The Jasper package was
added to the book for the library.

I would prefer if we put the dependencies as optional if we cannot
find some better argument than "To me, they look like a good idea"

What typically we do is say "Optional (required to build the XYZ
program). This let's users know they need the dependency if they
wish to have the XYZ program built. I cannot see us recommending
what programs the users have or don't have on their system when
those programs have no other affect on the book.


>  The viewer seems marginally useful.  I can understand keeping
> things as optional when they merely increase functionality of a
> program.  In this case, you don't have the program if you don't
> provide the dependency.

See above. And this is how we do it in most of the BLFS packages
where a dependency is soley used to build a sub-part of the package
that nothing else in the book depends on.


>> rmlscsi: [bogomips 1003.24] [GNU ld version 2.16.1] [gcc (GCC) 4.0.3]
>> [GNU C Library stable release version 2.3.6] [Linux 2.6.14.3 i686]
>> 11:55:00 up 17 days, 4:18, 1 user, load average: 0.07, 0.02, 0.0
> 
>  ¡ I hope this isn't the box where you are doing the BLFS-6.4
> upgrades !

What would it matter? x86 is x86. But for the record, no.

-- 
Randy

rmlscsi: [bogomips 1003.24] [GNU ld version 2.16.1] [gcc (GCC) 4.0.3]
[GNU C Library stable release version 2.3.6] [Linux 2.6.14.3 i686]
18:26:00 up 17 days, 10:49, 1 user, load average: 0.43, 0.43, 0.24