leaf-user

Re: [leaf-user] Future of Bering and Bering-uClibc?

[Timothy J. M. <tm...@mo...>]

lea...@li... wrote on 01/29/2004 08:00:09 AM:

 > > I have been using Bering quite successfully for some time now, but I'm
 > > at a spot where it would be good to evaluate a change to uClibc.  I
 > > haven't seen any discussion regarding development on Bering.  Of 
course,
 > > most of the development on uClibc has been to recompile existing Bering
 > > packages...  :)
 >
 > Either I miss you're irony, or you better reread the Changelog for
 > Bering-uClibc - none of the entries mentions other packages than 
theone from
 > the base image, and I can assure that recompiling packages has been a 
minor
 > effort compared to the changes for base image, addition of ipv6 and 
gaining
 > more space on the base image.

 From my (admittedly limited) research, it seemed that most of the 
messages regarding new items for Bering-uClibc were from people 
compiling (new) versions of software that already existed under 
traditional Bering.  Maybe I am mistaken:  I have not followed it that 
closely.

I asked a similar question about 4 months ago.  I was told to stick with 
Bering unless I needed the reduction in size that uClibc gave me.  
Seeing as I'm running on EPIA's with 128MB RAM and 32MB DOM, I really 
didn't.  The only thing that has prompted my question is I have seen no 
real development on Bering since 1.2, and Bering uClibc is readying its 
second or third release since then.

 > But maybe I understand better what you mean with "real future 
development" if
 > you give some more explanations.

I'm really not looking for anything specific.  I just want something 
that is going to keep up with, e.g., bugfixes in the underlying packages 
and kernel.  I'm not looking for any new features.

Tim Massey

Re: [leaf-user] Future of Bering and Bering-uClibc?

[Eric W. <le...@wo...>]

Hello Timothy, list

> lea...@li... wrote on 01/29/2004 08:00:09 AM:
> 
>  > > I have been using Bering quite successfully for some time now, but I'm
>  > > at a spot where it would be good to evaluate a change to uClibc.  I
>  > > haven't seen any discussion regarding development on Bering.  Of 
> course,
>  > > most of the development on uClibc has been to recompile existing Bering
>  > > packages...  :)
>  >
>  > Either I miss you're irony, or you better reread the Changelog for
>  > Bering-uClibc - none of the entries mentions other packages than 
> theone from
>  > the base image, and I can assure that recompiling packages has been a 
> minor
>  > effort compared to the changes for base image, addition of ipv6 and 
> gaining
>  > more space on the base image.
> 
>  From my (admittedly limited) research, it seemed that most of the 
> messages regarding new items for Bering-uClibc were from people 
> compiling (new) versions of software that already existed under 
> traditional Bering.  Maybe I am mistaken:  I have not followed it that 
> closely.
> 
> I asked a similar question about 4 months ago.  I was told to stick with 
> Bering unless I needed the reduction in size that uClibc gave me.  
> Seeing as I'm running on EPIA's with 128MB RAM and 32MB DOM, I really 
> didn't.  The only thing that has prompted my question is I have seen no 
> real development on Bering since 1.2, and Bering uClibc is readying its 
> second or third release since then.
I am working on a webinterface at the moment, due to professional ( not leaf 
related) changes) time is somewhat limited.  ( and the nice snow is distracting 
;) ) 
The inherent problem of Bering is the library, which isn't maintained anymore. 
on the other hand the "uniformity" of the "old lib" means that there are a 
whole bunch of packages available.
One of the reasons the uclibc group did release new versions is that if the 
uclibc library is updated, the packages mostly must be recompiled. What I mean 
with this is that those changes "deserve" the name  new release. 
The interesting is that the IMHO most important improvement the uclibc guys 
made, stays largely unnoted .namely the  use of  "automatic package making". 

 The necessity of recompiling with every new version of uclibc  will hopefully 
change as soon as uclibc will reach a version 1.0. 

My plans with bering are. Updateing to a new kernel version, thereby keeping as 
close to uclibc as possible.   
Improvement of the installation and maintainance issue. 
Working on a "change a setting only in one place" version. This will also be 
usuable with ucblic.
Recompiles of single packages were and are done, but didn't need a complete new 
"release". so they weren't that obvious.

Hope to have answered your question with that. 

Regards
Eric Wolzak
member of the bering Crew

 
> I'm really not looking for anything specific.  I just want something 
> that is going to keep up with, e.g., bugfixes in the underlying packages 
> and kernel.  I'm not looking for any new features.
> 
> Tim Massey
>