:: Re: [DNG] Why Debian 8 Pinning is (…
Top Page
Delete this message
Reply to this message
Author: Simon Hobson
Date:  
To: dng
New-Topics: [DNG] Optional libraries
Subject: Re: [DNG] Why Debian 8 Pinning is (or isn't) pointless
Rick Moen <rick@???> wrote:

I think we're arguing in violent agreement - there is more than one way to approach the issue, more than one attitude to "risk", and what works for one person isn't necessarily what works for someone else. Isn't that a key tenet of the FOSS way - the freedom of choice ?


> Well, Works for Me.{™}


That's the point - I never suggested it didn't. But it doesn't work for me - as in that's not a route I'm happy taking in order to deal with it.

> Can you, with crystal ball, 100% guarantee that I won't deal with my
> system to effectively apply local policy using regular open source
> practices?


And again, you seem to be reading into it things I haven't said. I'm sure you'll carry on adding more tape as required and achieve the result you want in a way you are happy with. As I said, that's not a route I'm happy taking. I may have to if the combination of factors (including my own skill set) conspire to make it the least bad way to achieve something, but until then I'll be avoiding it. That doesn't mean I have a problem with you doing that - as you say, it works for you.



> You probably wouldn't even like removing libsystemd0 entirely and
> replacing it with an 'equivs' recipe, which could also be done if one
> really, really, really were concerned.
>
> But, for those interested in that technique, see: 'How To Satisfy
> Debian Dependencies Without Installing The Stupid Package' on
> http://shallowsky.com/blog/linux/install/blocking-deb-dependencies.html
>
> (You're welcome!)


Thank you for that link, I was aware of the equivs feature, but it's not something I've used.
But, for testing it can be simulated by just over-riding dependencies, as in dpkg --force-depends -r libsystemd0
And hey presto, the package causing me problems would fail to start - needless to say, when it calls a function in that library the call fails.

Now, if the devs/maintainers used a "if it's there call it, if it isn't then don't" approach then I could see that working. I assume that is a possible way of calling library functions given the number of programs with optional dependencies/features. When I raised it with the particular package maintainers, someone off-list sent me an email saying (of the response I got) "if that's the support department, I'd hate to see the complaints department" !


> Duct tape, actually. It's like the Force. It has a light side and a
> dark side, and it holds the universe together.
>
> Duct tape is cool.


Over here, the common term is gaffer tape. Duct tape is a common term as well, and of course the commercial product playing on the homophone (if I go the right word) with Duck Tape