[Grml] removal of "buffer" command? alternatives?

[Michael Prokop]

* Csillag Tamas [Wed Dec 28, 2011 at 11:08:29AM +0100]:

> I am a bit sad glancing on that list.
> Some of the tools removed are big (ok, so one can accept that and it
> can be justified), but most of them are small so from the size point
> of view it does not help if you remove.

> I name just a few:
> afio

Will be further discussed [1]

> aircrack-ng

Not available from Debian anymore/yet

> awesome (i do not use this a friend does)

Was discussed but wasn't considered to be essential enough to be
included.

> build-essential (so from now on building inside grml will be not that easy)
> gcc, g++ also gone...

It's not considered as part of our install & rescue" mission to
compile software on the live system. If you want/need a
full-featured development environment we recommend remastering and
ship everything *you* need.

> comgt (180k)

Thanks, re-added already: http://git.io/rtTeQg

> dns2tcp (201k) (I am using this if I stay at a hotel with insane internet rates)

Will be further discussed [1]

> emacs23 (this is a beast I know, more than one friend use this)

Sorry, I like and use emacs on my own, but emacs23 is just too large
for inclusion and doesn't match our "install & rescue" mission
neither.

> fakeroot

Will be further discussed [1]

> firmware-qlogic (this means that it will not boot on server containing FC card?)

Will be further discussed [1]

> irssi

If you've network access you can just install it :)

> ipmitool

Will be further discussed [1]

> ldapvi

Will be further discussed [1]

> libnet-*-perl

They used to be dependencies of other packages that aren't available
any longer. Is there any specific package that's considered relevant
for install and rescue?

> mutt
> postfix (I was using this for educating mailing basics also good for
>          testing)

Both shouldn't be relevant for a live system (and both don't match
the install and rescue mission), and if you've network access you
can once again just install it :)

> pppoe (no one with adsl anymore?)

See Christian's mail

> radvd

Will be further discussed [1]

> runit (520k)

What's the benefit for the live system?

> rxvt-unicode (this was part or the release from the begining)

xterm is there and should be enough for install and rescue mission, IMO

> tcptraceroute

Will be further discussed [1]

> virtualbox
> tp_smapi

External kernel modules available via DKMS can't be shipped as
standalone binary Debian packages yet (due to limitations in DKMS).
Evgeni Golov is working on something to resolve that, but until this
is ready-to-go I'm afraid we won't provide external kernel packages.

> zsh-lovers is your package afaik

It wasn't considered relevant enough for the install and rescue mission.
Not sure what our people think about it?

> If a tool is used when you are online it makes sense to remove it (nsd,
> bind9) and one can reinstall easily.
> If you remove pppoe how can one connect facing an adsl connection?
> If an package mostly used offline is removed a functionality is lost.

See Christian's mail (short version: ppp still supported, pppoe is
old and deprecated).

> I make my /boot 1Gb to be able to put a grml there for recovery
> purposes.

Same for me :)

> I used to make my own (remaster) grml flavour and a bit of this or
> that. But now that most of the tools gone maybe I will just stick to
> the old one.

Remastering became easier than ever with this release:

  http://blog.grml.org/archives/364-Remastering-Grml-2011.12-will-be-as-easy-as-never-before.html

And there's no reason to not remaster it. We can't provide one Grml
version that makes everyone happy. (And the old model just doesn't
work for us any longer if the manpower stays the same, see below.)

> I used to recommend grml to friends who are not that experienced to
> make their own grml (and to tell the truth most of us are just lazy or
> lack the time).

Well, that's exactly our problem: we also lack time and we couldn't
keep the project up and running any longer if we wouldn't have
chosen the "reset button" for this release. With 3 flavours for 2
architectures we had way too many ISOs to support with the available
manpower.

> What kind of testing is needed to get (most/some of) the tools back
> on the cd?

Testing is very important, yes. But it's not just testing but also:

* taking care of failing builds (we provide daily builds at
  http://grml.org/daily/ and trigger ISO builds with each git
  commit)

* integrating packages into Grml tools (grml-quickconfig,
  grml-x,...), window manager, providing sane default configs,...

* taking care of bugs (see http://bts.grml.org/grml/) and
  user support ("why doesn't foo work?")

We highly appreciate and welcome any contributors helping us out.

Reminder: we're an open source project. Everyone can contribute and
improve areas where work needs to be done. But we just can't do all
the work for you if it's "just take but don't give back".

> I used to be around on IRC, but when some folk switched from
> english I was unable to follow the conversation so I dropped out.

Sorry for that. The channel language is english-only nowadays, so
please feel free to join the IRC channel again.

> I want grml to be *the* rescue/sysadmin cd again not just one of the
> bootable linux cds out there.
> Thanks.

Thanks for your feedback, we highly appreciate that.

[1] Thanks for the list. We will review and discuss your software
selection in further detail, promised.

regards,
-mika-
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: Digital signature
URL: <http://ml.grml.org/pipermail/grml/attachments/20111228/de1189f4/attachment.pgp>