[Exherbo-dev] GNU coreutils alternatives

Bernd Steinhauser exherbo at bernd-steinhauser.de
Thu Sep 17 20:35:04 UTC 2015


I actually had the same question.

On 17/09/15 21:39, Kylie McClain wrote:
> On 09/17/2015 03:18 PM, Ciaran McCreesh wrote:
>> What is the compelling reason that makes it worth spending any effort
>> on this?
> I have a few, actually.
>
> 1. It allows for more choice with respect to core parts of the system. I
> wonder how quickly someone's going to crucify me for saying "choice" is
> a reason. I believe it is reasonable to provide this option, it's not
> that much different from how we provide the choice of libav over ffmpeg,
> or libressl over openssl.
Well, this is not Gentoo. We don't like to introduce options just so there is an 
option.
> 2. Making the core of the system more flexible and interchangeable will
> allow for easier usage on systems with low memory, or any other system
> where usage of programs normally used on embedded systems is more
> desired than their heavier counterparts.
Terms like "heavier counterparts" are used quite often but there is no reliable 
definition on any term like that.
And if people try to have a go at it, it very often proves them wrong.
I'm just saying that you should be very careful with using terms like that.

Anyway. I'm totally against too much flexibility in the basic system. All this 
will lead to is a long list of bugs and unforseen failures and consequences.
It's different for libav/ffmpeg, as these don't belong to the basic system, so 
changing them will result in a more/less pleasing multimedia experience for 
people if it makes a change at all. And actually I think that we should remove 
this one as well, as I've seen quite a few people strongly advising against 
using libav (that was different some time ago).
For libressl/openssl it's different as well, as—to my knowledge—they are 
compatible for any day2day use.

I don't know if there are actually people building an embedded system based upon 
Exherbo. If there are any, please step up and clarify if these changes are 
necessary.
If space and memory consumption really matters to them, I suspect they will make 
much more severe changes to the basic system than just replace a few binaries.
>
> This is actually my most important reason; the idea of having a single
> package that can provide a lot of the base of the system without much
> issue,s and take up less space than more full-featured is a very
> appealing idea.
Does anybody really care about space on an Exherbo system (serious question)?
(We are talking about a few MBs here at most.)
> 3. In the same way that the multiarch change proved as a test to things
> like autotools, this can serve as a test to how alternative
> implementations are out there and work just fine. I'm going to use that
> phrase that everyone is probably tired of hearing me say now.
> Prevents monoculture.
And sounds like a pointless exercise. I'd leave that sort of thing to the Gentoo 
guys and focus on things that are more useful.

Don't get me wrong, I have nothing against using a different tool to work with. 
I myself love zsh and I'm using that one instead of the standard (bash).
But I would never even try to use it as a /bin/sh shell interpreter on Exherbo 
or even to try to use it for cave (which I guess would mean a lot of work).
If you like one tool more than another use that. If you want to have it as a 
standard in your interactive shell, it's almost trivial to setup to shell for 
that. But leave the system standard as-is.
In my opinion, we already have way too many alternatives modules.

BR,
Bernd



More information about the Exherbo-dev mailing list