Re: A suggestion

[Anibal Avelar]

> It is not just people new to Debian/Linux who get that
>   impression.  I think testing is perfectly fine for a desktop system.

Totally I agree. Testing is fine for a personal Desktop machine.


>  > If Debian wants to issue stronger warnings for those using Testing on
>  > the desktop, maybe mentionning that it is also a development tool and
>  > it is not inteded for normal use would do the job…
>
> Well, I guess we could, but then I think we would also be
>   unnecessarily alarmist.

I agree too. I have been using Debian Testing how my desktop system
for a long time (years) without any real problems (in my job and
house).

I think Testing is enough for a Linux user with intermediate level (or less).

Regards.

-- 
Anibal Avelar (FixXxeR) http://fixxxer.cc
GPG: 83B64656 - C143 4AD8 B017 53FA B742 D6AA CEEA F9F3 83B6 4656

Re: A suggestion

[Manoj Srivastava]

On Thu, 3 Apr 2008 10:29:30 +0900, Charles Plessy <[EMAIL PROTECTED]> said: 

> Le Thu, Apr 03, 2008 at 01:44:56AM +1300, Chris Bannister a écrit :
>> 
>> Unfortunately, many people new to Debian/Linux get the impression
>> that testing is perfectly suitable for a desktop system.

> Maybe because they have read Debian's website?

It is not just people new to Debian/Linux who get that
 impression.  I think testing is perfectly fine for a desktop system.

> If Debian wants to issue stronger warnings for those using Testing on
> the desktop, maybe mentionning that it is also a development tool and
> it is not inteded for normal use would do the job…

Well, I guess we could, but then I think we would also be
 unnecessarily alarmist.

manoj
-- 
Is it clean in other dimensions?
Manoj Srivastava <[EMAIL PROTECTED]>   
1024D/BF24424C print 4966 F272 D093 B493 410B  924B 21BA DABB BF24 424C

Re: A suggestion

[Charles Plessy]

Le Thu, Apr 03, 2008 at 01:44:56AM +1300, Chris Bannister a écrit :
> 
> Unfortunately, many people new to Debian/Linux get the impression that
> testing is perfectly suitable for a desktop system.

Maybe because they have read Debian's website?

 "things should not break as badly as in unstable or experimental
 distributions"
http://www.debian.org/releases/testing/index.en.html

 "The unstable distribution is where active development of Debian
 occurs. Generally, this distribution is run by developers"
http://www.debian.org/releases/index.en.html

There is nothing in these pages that warns against the use of Testing on
a desktop system, except of course that Debian "primarily recommends
using" Etch.

There are three alternative options for escaping Etch's bugs (for
instance white-on-white selections in Evince):

 - Backporting some packages by hand, but this requires skill.
 - Using backports.org, but it is unofficial.
 - Using Testing or Sid.

If Debian wants to issue stronger warnings for those using Testing on
the desktop, maybe mentionning that it is also a development tool and it
is not inteded for normal use would do the job…

Have a nice day,

-- 
Charles Plessy
http://charles.plessy.org
Wakō, Saitama, Japan


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: triggers wishlist

[Joe Smith]

"Michael Biebl" <[EMAIL PROTECTED]> wrote in message 
news:[EMAIL PROTECTED]

I guess Joss is right here. triggers tell you *if* something has changed

>(in a subdirectory), but not *what*.

Remember, that we have to call
gtk-update-icon-cache /usr/share/icons/$subdir
for the directory that has changed.

Michael


How expensive is running gtk-update-ican-cache? I mean, it would be easy 
enough to have a script run
that runs "gtk-update-icon-cache /usr/share/icons/$subdir" for each of the 
subdirectories. Then the /usr/share/icons
trigger would just run that script. 




--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#474050: RFP: jaxer -- headless iceweasel

[OuoU]

Package: wnpp
Severity: wishlist

* Package name: jaxer
  Version : 0.9.6
  Upstream Author : Aptana
* URL or Web page : http://www.aptana.com/jaxer/download
* License : GPL
  Description : iceweasel headless

A headless web browser is important for running JS applications on 
a server and for testing applications.

>From the web page:

Modern web sites and applications use Ajax to create engaging user
experiences: the HTML and CSS are set in motion using JavaScript
running in the browser and calling back the server. To achieve this,
the server needs to prepare the web page appropriately, and to know
what to do when the JavaScript calls it. But the server knows nothing
about the HTML and CSS DOM, nor how to handle JavaScript data, and you
can't code it in JavaScript...

...Jaxer changes all that. Jaxer is the world's first true Ajax
server. HTML, JavaScript, and CSS are native to Jaxer, as are
XMLHttpRequests, JSON, DOM scripting, etc. And as a server it offers
access to databases, files, and networking, as well as logging,
process management, scalability, security, integration APIs, and
extensibility.

-- 
+33 1 76 60 72 81  Loic Dachary mailto:[EMAIL PROTECTED]
http://dachary.org/loic/gpg.txt sip:[EMAIL PROTECTED]
Latitude: 48.86962325498033 Longitude: 2.3623046278953552



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#474016: ITP: desktop-data-model -- a library for Mugshot and Online-Desktop

[Julien Lavergne]

On Wed, 02 Apr 2008 15:27:23 -0500
William Pitcock <[EMAIL PROTECTED]> wrote:

> Hi,
> 
> On Wed, 2008-04-02 at 15:15 -0500, Ron Johnson wrote:
> > That doesn't appear to be a valid address,  It redirects to
> > http://www.mugshot.com/ which seems to be just a bunch of links to
> > scam and valid commercial web sites.
> 
> Mugshot has always been at mugshot.org; not mugshot.com. So the ITP
> creator should take note of this. http://developer.mugshot.org/ is a
> mediawiki instance :)

Yes, sorry about that, I made a stupid mistake.
Thanks William for the correction :)


Regards,
Julien Lavergne <[EMAIL PROTECTED]>


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: A suggestion

[Hendrik Sattler]

Am Mittwoch 02 April 2008 schrieb Chris Bannister:
> On Tue, Apr 01, 2008 at 03:16:48PM +0200, Andrea Bolognani wrote:
> > On Tue, 1 Apr 2008 00:38:36 -0300
> >
> > Joel Franco <[EMAIL PROTECTED]> wrote:
> > > I think that the testing or above releases may be suitable for the
> > > desktop of a EXPERT linux user. I need a stable release because my
> > > desktop must simply to work; it is not a so critical system like a
> > > server but this is critical to my business go on.
> > >
> > >From my experience, testing is perfectly suitable for a
> > > desktop/development
> >
> > system. I have used it exclusively on both my desktop and my laptop for a
> > couple of years now, and I have experienced only a couple of minor issues
> > so far.
> >
> > The main difference between a desktop and a server is that usually on a
> > server you can't afford any downtime, while on a desktop system you can
> > screw up and reinstall every once in a while if needed...
>
> Unfortunately, many people new to Debian/Linux get the impression that
> testing is perfectly suitable for a desktop system.

It really depends on the person(s) using that desktop. But your wording 
is "black or white" anyway (see: perfectly).

HS


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#474036: ITP: xnetcardconfig -- tool to configure your network cards using an easy wizard

[Roberto C . Sánchez]

On Wed, Apr 02, 2008 at 11:05:46PM +0200, Philippe Coval wrote:
> Package: wnpp
> Severity: wishlist
> Owner: Philippe Coval <[EMAIL PROTECTED]>
> 
> 
> * Package name: xnetcardconfig
>   Version : 0.2.1
>   Upstream Author : Benedikt Meurer <[EMAIL PROTECTED]>
> * URL : http://www.example.org/

Where is the package really hosted upstream?

> * License : GPL
>   Programming Lang: Ruby
>   Description : tool to configure your network cards using a wizard GUI
> 
> It's just a ruby script to be used along sudo.
> 
> Note that I am about to upload the package to debian mentors 
> and start looking for a sponsor.
> 
If it us "just a ruby script", is it sensible to use an entire package
for it?  What about trying to get it included in another package that
has similar functionality already?

Regards,

-Roberto

-- 
Roberto C. Sánchez
http://people.connexer.com/~roberto
http://www.connexer.com


signature.asc
Description: Digital signature

Re: Adding lzma to dpkg's Pre-Depends

[Hendrik Sattler]

Am Mittwoch 02 April 2008 schrieb Mike Bird:
> On Wed April 2 2008 01:52:39 Steve Langasek wrote:
> > On Wed, Apr 02, 2008 at 10:24:43AM +0200, Lucas Nussbaum wrote:
> > > The results are a bit misleading, because they compare the absolute
> > > gain.
> >
> > It's the absolute size savings that counts - that's what's going to have
> > the biggest impact on mirror size, mirroring time, CD set size, and
> > download times for users.
>
> Lucas made a good point.  Better to save 20% on ten 10MB packages than
> saving 10% on one 100MB package.
>
> Although we can probably have both, modulo decompression RAM issues.

Well, oo.org probably needs the same amount of RAM to actually run, so it's 
not an issue, there. That not the case for the libgl1-mesa-dri but probably 
for programs that use OpenGL . So the consideration should always include 
what effect this actually has for the users of that package.
No question that for Oo.org this is pure gain.

HS


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#474036: ITP: xnetcardconfig -- tool to configure your network cards using an easy wizard

[Philippe Coval]

Package: wnpp
Severity: wishlist
Owner: Philippe Coval <[EMAIL PROTECTED]>


* Package name: xnetcardconfig
  Version : 0.2.1
  Upstream Author : Benedikt Meurer <[EMAIL PROTECTED]>
* URL : http://www.example.org/
* License : GPL
  Programming Lang: Ruby
  Description : tool to configure your network cards using a wizard GUI

It's just a ruby script to be used along sudo.

Note that I am about to upload the package to debian mentors 
and start looking for a sponsor.

Regards

--
http://rzr.online.fr/q/ruby



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#474034: ITP: gpxe -- PXE network bootloader for x86 PCs

[Jan Lübbe]

Package: wnpp
Severity: wishlist
Owner: "Jan Lübbe" <[EMAIL PROTECTED]>


* Package name: gpxe
  Version : 0.9.3
  Upstream Author : Michael Brown <[EMAIL PROTECTED]> and the
Etherboot Project
* URL : http://www.etherboot.org
* License : GPL
  Programming Lang: C
  Description : PXE network bootloader for x86 PCs

gPXE provides a direct replacement for proprietary PXE ROMs, with many
extra features such as DNS, HTTP, iSCSI.
It can be stored in a number of places, including BIOS Flash, EPROMs,
floppy, CD, HD, or other bootable media.

gPXE evolved from Etherboot and is maintained by the Etherboot project.
It will supersede Etherboot eventually.

-- System Information:
Debian Release: lenny/sid
  APT prefers unstable
  APT policy: (900, 'unstable'), (300, 'experimental')
Architecture: i386 (i686)

Kernel: Linux 2.6.24-x60s (SMP w/2 CPU cores; PREEMPT)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#474016: ITP: desktop-data-model -- a library for Mugshot and Online-Desktop

[William Pitcock]

Hi,

On Wed, 2008-04-02 at 15:15 -0500, Ron Johnson wrote:
> That doesn't appear to be a valid address,  It redirects to
> http://www.mugshot.com/ which seems to be just a bunch of links to
> scam and valid commercial web sites.

Mugshot has always been at mugshot.org; not mugshot.com. So the ITP
creator should take note of this. http://developer.mugshot.org/ is a
mediawiki instance :)

William


signature.asc
Description: This is a digitally signed message part

Re: Bug#474016: ITP: desktop-data-model -- a library for Mugshot and Online-Desktop

[Ron Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 04/02/08 13:04, Julien Lavergne wrote:
> Package: wnpp
> Severity: wishlist
> X-Debbugs-CC: debian-devel@lists.debian.org
> 
> --- Please fill out the fields below. ---
> 
>Package name: desktop-data-model
> Version: 1.2.2
> Upstream Author: Mugshot Developers
> URL: http://developer.mugshot.com

That doesn't appear to be a valid address,  It redirects to
http://www.mugshot.com/ which seems to be just a bunch of links to
scam and valid commercial web sites.

- --
Ron Johnson, Jr.
Jefferson LA  USA

We want... a Shrubbery!!
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFH8+lSS9HxQb37XmcRAny3AKDvGEySnVmR9WSIjMe4RfNZKfI6CQCeJ+jm
a38qF+yxB1QdKgeup/aiKB4=
=GyVZ
-END PGP SIGNATURE-


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

dm-crypt and boot process

[John Goerzen]

Hi,

I've been very happy that the Etch installer supports dm-crypt out of the 
box.  This is a wonderfully nice feature.

Here's my gripe: it gets in the way of unattended boots.  Let's say that you 
have /home as a separate encrypted filesystem on a given machine.  You want 
the machine to be able to boot even if you aren't there -- say because the 
power goes out or something.  But you have a passphrase for /home.

You could set it up with a timeout in crypttab, but here's the rub...  when 
you do that, and the timeout expires, the boot process halts.  You have to 
sit at the console and give the root password, then /etc/init.d/cryptdisks 
start, then proceed.

In a case like this, it seems desirable to have the boot process not be 
interrupted.  If the machine boots without /home, I could at least ssh into 
it as root and fix that problem.

As far as I can tell, there is no way in the installer to indicate this 
preference, and no way in fstab to specify that a failure to find the crypt 
device for a given filesystem should just be ignored, leaving that 
filesystem unmounted.

So I haven't submitted a bug anywhere because I don't know where to do so, or 
if perhaps new code needs to be written to accommodate this scenario.  Does 
anyone know?

Thanks,

-- John


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: file path length in .deb package.

[Mark Brown]

On Wed, Apr 02, 2008 at 08:43:47PM +0200, Mathieu Malaterre wrote:

> Could you *please* give me the bug number,

If you can't find a relevant bug you should report a new one (anyone can
do so).  See:

http://www.debian.org/Bugs/Reporting

for instructions on how to do this manually or there is a tool called
'reportbug' in the reportbug package which will automate the process.

-- 
"You grabbed my hand and we fell into it, like a daydream - or a fever."


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: file path length in .deb package.

[Mathieu Malaterre]

On Wed, Apr 2, 2008 at 6:35 PM, Bart Martens <[EMAIL PROTECTED]> wrote:
>
>
>  On Wed, 2008-04-02 at 17:46 +0200, Mathieu Malaterre wrote:
>  > Hi there,
>  >
>  >   After reading the documentation of 'man deb' I cannot see any
>  > reference to any arbitrary length for filename in a deb package. For
>  > instance I have filepath (including sub directories) that reach ~100
>  > characters. the tar format completely support that, but I am having
>  > some weird issue on my debian etch amd64 box. I have created a package
>  > where all files with long filepaths simply don't get installed (while
>  > being accessible using ar xvf + tar xvfz data.tgz).
>  >
>  >   Let me know if this is a known issue.
>
>  Please consult the "Debian bug tracking system".

wow ! Thanks, may I ask what you did during the 1s you spare not
copy/pasting the bug number. You're loosing everyone time:
- mine (obviously)
- yours in answering such a lame response
- anyone else that will find my post after a google search

Anyway I did search on google for a *while* before posting, and could
not find anything. I did repeat the search again for the last 15min:

  long filename dpkg site:bugs.debian.org

  long apt site:bugs.debian.org

  long dpkg site:bugs.debian.org


It returned only one bug, but that was solved in 2002:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=128388


Could you *please* give me the bug number,

Thanks
-- 
Mathieu


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: triggers wishlist

[Ian Jackson]

Joey Hess writes ("triggers wishlist"):
> Things I want to see use triggers, in approximate priority order:

One thing I forgot to say yesterday:

If you do the triggerisation by modifying some dh_foobar-called script
called update-foobar, then you do not need to introduce a dependency
on the triggers-supporting dpkg.

Do as I do in my patch for initramfs-tools (#447611), and you can have
update-foobar do its stuff if the triggers are not supported.  That
means you can upload your change now and not wait for the triggers
dpkg to make it into sid or lenny.

Of course you should test your change with the triggers dpkg or you'll
find that your scripts break when the new dpkg makes it into lenny.

(When reading the initramfs-tools patch, you can ditch or ignore the
test for DPKG_RUNNING_VERSION as you like, as that's only dealing with
some buggy early triggers support in an unreleased Ubuntu version.
Also please disregard the comments in the bug report which are not
really of any interest unless you happen to like getting distracted by
a discussion of the purpose of `x' in  test "x$foo" = x  and the like.)

Ian.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#474016: ITP: desktop-data-model -- a library for Mugshot and Online-Desktop

[Julien Lavergne]

Package: wnpp
Severity: wishlist
X-Debbugs-CC: debian-devel@lists.debian.org

--- Please fill out the fields below. ---

   Package name: desktop-data-model
Version: 1.2.2
Upstream Author: Mugshot Developers
URL: http://developer.mugshot.com
License: LGPL 2+
Description: Desktop-Data-Model (ddm) is a common component for all 
Online-Desktop programs (Mugshot, BigBoard and Online-desktop).

It was initially included in Mugshot itself. This split in a separate package 
allow to release others components of Online-Desktop without the Mugshot client.

All work is available on the SVN of the Online-Desktop maintainers team 
(http://svn.debian.org/wsvn/online-desktop/trunk/desktop-data-model/?rev=0&sc=0)

Tarball is available on 
http://ftp.gnome.org/pub/gnome/sources/desktop-data-model/1.2/



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

file path length in .deb package.

[Mathieu Malaterre]

Hi there,

  After reading the documentation of 'man deb' I cannot see any
reference to any arbitrary length for filename in a deb package. For
instance I have filepath (including sub directories) that reach ~100
characters. the tar format completely support that, but I am having
some weird issue on my debian etch amd64 box. I have created a package
where all files with long filepaths simply don't get installed (while
being accessible using ar xvf + tar xvfz data.tgz).

  Let me know if this is a known issue.

Regards,
-- 
Mathieu


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#473998: ITP: debian-faq -- The Debian FAQ

[Cyril Brulebois]

On 02/04/2008, Joost van Baal wrote:
>  In this package you will find the Debian GNU/Linux FAQ, which gives
>  frequently asked questions (with their answers!) about the Debian
>  distribution Debian GNU/Linux and others) and about the Debian project.
   ^ mismatched parenthesis?

>  You'll find out that some answers assume some knowledge of Unix-like
   ^^^ I'm not sure this is welcome...

>  operating systems.  We'll try to assume as little prior knowledge as
   ^^ ... but that certainly is not.

>  The document is supplied in HTML, PDF, PostScript and plain ascii.
   ^ ASCII?

Cheers,

-- 
Cyril Brulebois


pgpGrENNN5l9x.pgp
Description: PGP signature

Bug#473998: ITP: debian-faq -- The Debian FAQ

[Joost van Baal]

Package: wnpp
Severity: wishlist
Owner: Joost van Baal <[EMAIL PROTECTED]>

* Package name: debian-faq
  Version : 3.1.6
  Upstream Author : Ray Dassen, Javier Fernandez-Sanguino, Susan G. Kleinmann,
   Josip Rodin, Sven Rudolph, Chuck Stickelman, Santiago Vila e.a.
* URL : http://www.debian.org/doc/user-manuals#faq
* License : MIT-ish
  Programming Lang: DebianDoc SGML
  Description : The Debian FAQ

 In this package you will find the Debian GNU/Linux FAQ, which gives
 frequently asked questions (with their answers!) about the Debian
 distribution Debian GNU/Linux and others) and about the Debian project.
 You'll find out that some answers assume some knowledge of Unix-like
 operating systems.  We'll try to assume as little prior knowledge as
 possible: answers to general beginners questions will be kept simple.
 .
 This document is available at http://www.debian.org/doc/manuals/debian-faq/
 as well as from the Debian ftp server at ftp://ftp.debian.org/debian/doc/FAQ
 and mirrors thereof.
 .
 The document is supplied in HTML, PDF, PostScript and plain ascii.
 .
 If you're new to Debian, and like to read documentation from your local
 system, without using the network, install this package.


Hi,

Up to 3.1.5, the Debian GNU/Linux FAQ was shipped with the doc-debian
package.  After some discussion on debian-doc list (
http://lists.debian.org/debian-doc/2008/03/threads.html#00065 ), work
started on splitting the Debian GNU/Linux FAQ from doc-debian, and
shipping it in a package of its own.  The debian-faq source package will
create a debian-faq binary package, holding the FAQ in English, as well
as debian-faq-de, debian-faq-fr, debian-faq-it, and debian-faq-zh-cn,
holding translations previously not shipped in packages.

Current maintainer of the FAQ is Javier Fernández-Sanguino Peña.

Bye,

Joost



signature.asc
Description: Digital signature

Re: triggers wishlist

[Norbert Preining]

Sorry for answering s late, but I am on VAC for another 2 weeks.

On Di, 01 Apr 2008, Ian Jackson wrote:
> update-initramfs is important, yes.  It's very tedious and gets run
> several times in some upgrade runs.  I did the adaptation for this in

The Debian TeX would need - for what ever you are discussion ;-) - also
something. We have 3 things which takes loads of time:
mktexlsr
updmap-sys
fmtutil-sys
calls (they differ between the packages). THere are several bugs
(Installing texlive calls updmap again and again and again) and yes,
that is a problem. So we would like to have something similar, if
possible.

Maybe someone else (Florent, fln) has already replied, I couldn't check.

Best wishes

Norbert

---
Dr. Norbert Preining <[EMAIL PROTECTED]>Vienna University of Technology
Debian Developer <[EMAIL PROTECTED]> Debian TeX Group
gpg DSA: 0x09C5B094  fp: 14DF 2E6C 0307 BE6D AD76  A9C0 D2BF 4AA3 09C5 B094
---
`She hit me on the head with the rock again.'
`I think I can confirm that that was my daughter.'
`Sweet kid.'
`You have to get to know her,' said Arthur.
`She eases up does she?'
but you get a better sense of when to
duck.'
 --- Ford and Arthur on Random.
 --- Douglas Adams, The Hitchhikers Guide to the Galaxy


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Adding lzma to dpkg's Pre-Depends

[Mike Bird]

On Wed April 2 2008 01:52:39 Steve Langasek wrote:
> On Wed, Apr 02, 2008 at 10:24:43AM +0200, Lucas Nussbaum wrote:
> > The results are a bit misleading, because they compare the absolute
> > gain.
>
> It's the absolute size savings that counts - that's what's going to have
> the biggest impact on mirror size, mirroring time, CD set size, and
> download times for users.

Lucas made a good point.  Better to save 20% on ten 10MB packages than
saving 10% on one 100MB package.

Although we can probably have both, modulo decompression RAM issues.

--Mike Bird


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#473984: ITP: synce-kpm -- KDE device manager for Windows Mobile devices

[Jonny Lamb]

Package: wnpp
Severity: wishlist
Owner: Jonny Lamb <[EMAIL PROTECTED]>
X-Debbugs-CC: debian-devel@lists.debian.org

* Package name: synce-kpm
  Version : 0.11
  Upstream Author : Guido Diepen <[EMAIL PROTECTED]>
* URL : http://synce.org/
* License : GPL
  Description : KDE device manager for Windows Mobile devices

 SynCE-KPM is a KDE device manager for Windows Mobile devices running with
 SynCE. It can:
 .
  * Show tray icon popups with (dis)connection events
  * Show some general information about the PDA
  * Manage Software:
- List of all installed programs
- Install CAB files to the device
- Uninstall program
  * Manage Partnerships: (Requires running sync-engine)
- List partnerships
- Remove partnership
- Add partnership
  * Unlock both password protected devices

-- 
Jonny Lamb, UK   [EMAIL PROTECTED]
http://jonnylamb.com GPG: 0x2E039402


signature.asc
Description: Digital signature

Re: A suggestion

[Chris Bannister]

On Tue, Apr 01, 2008 at 03:16:48PM +0200, Andrea Bolognani wrote:
> On Tue, 1 Apr 2008 00:38:36 -0300
> Joel Franco <[EMAIL PROTECTED]> wrote:
> 
> > I think that the testing or above releases may be suitable for the
> > desktop of a EXPERT linux user. I need a stable release because my
> > desktop must simply to work; it is not a so critical system like a
> > server but this is critical to my business go on.
> 
> >From my experience, testing is perfectly suitable for a desktop/development
> system. I have used it exclusively on both my desktop and my laptop for a
> couple of years now, and I have experienced only a couple of minor issues
> so far.
> 
> The main difference between a desktop and a server is that usually on a
> server you can't afford any downtime, while on a desktop system you can
> screw up and reinstall every once in a while if needed...

Unfortunately, many people new to Debian/Linux get the impression that
testing is perfectly suitable for a desktop system.

-- 
Chris.
==


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#206187: very CheapPrice Bacheelor, MasteerMBA, and Doctoraate dip1omas ira s0fb7

[Dona Escobar]

   
We can assist with Diplomas from prestigious universities based on your present 
knowledge and LifeExperience
 
No required examination, tests, classes, books, or interviews 
 
Bacheelor, MasteerMBA, and Doctoraate diplomas available in the field of your 
choice - 
that's right, you can become a Doctor and receive all the benefits that comes 
with it!   
  
No one is turned down 
Confidentiality assured   
CALL US 24 HOURS A DAY, 7 DAYS A WEEK   
 
For US: 1-309-407-3749  
Outside US: +1-309-407-3749  

"Just leave your NAME & PHONE NO. (with CountryCode)" in the voicemail  

Please call today!!   
  
 
 
 
   



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#107862: very CheapPrice Bacheelor, MasteerMBA, and Doctoraate dip1omas teifuod c163c

[Ethel K. Beach]

We can assist with Diplomas from prestigious universities based on your present 
knowledge and LifeExperience 

No required examination, tests, classes, books, or interviews  

Bacheelor, MasteerMBA, and Doctoraate diplomas available in the field of your 
choice -   
that's right, you can become a Doctor and receive all the benefits that comes 
with it! 
  
No one is turned down  
Confidentiality assured   
CALL US 24 HOURS A DAY, 7 DAYS A WEEK  
   
For US: 1-309-407-3749
Outside US: +1-309-407-3749

"Just leave your NAME & PHONE NO. (with CountryCode)" in the voicemail  
   
Please call today!!
 
   
 
  
  



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: changing the default syslog daemon for lenny?

[Holger Levsen]

Hi,

(Please don't cc: me, I'm subscribed.)

On Wednesday 02 April 2008 01:20, Michael Biebl wrote:
> Holger Levsen schrieb:
> > Did you file those bugs? Is there a usertag for them?
>
> I tried to collect the information at http://wiki.debian.org/Rsyslog and
> started filing the important bugs [1]. 

Only one bug is open in the usertag list, but most from the wiki are not filed 
yet.

> The real showstopper bug, which 
> made rsyslog uninstallable, in the psad package, has been fixed already.
> For the remaining ones (changing sysklogd | system-log-daemon -> rsyslog
> | system-log-daemon), I wanted to wait until rsyslog is actually
> accepted as default syslog daemon

Makes sense, but also makes this goal a bit look like its stalled.

> > I'm still willing to help... :) (But as its obvious now, I have lost the
> > status on this..)
> If the wiki page is missing information, please let me know.

Looks good to me, thanks.

> >>> - Document the change by preparing a patch for the release notes
> >>> - Get the priorities fixed by a ftpmaster (assistant)
> >> Should, I file a lenny release goal first and wait for it's approval, or
> >> can I take this thread as consensus that I can pursue changing the
> >> default system-log-daemon to rsyslog?
> > Or was it already too late for this?
> Dunno. We should bring this topic up on debian-release I think to get
> more input there. If you want to kick of the discussion, please go ahead.

So, hi, debian-release, what is your opinion on this? Do you think we still 
have the time to do this change for Lenny or is it too late already?


regards,
Holger

> [1]
> http://bugs.debian.org/cgi-bin/pkgreport.cgi?tag=goal-rsyslog;[EMAIL 
> PROTECTED]
>ebian.org


pgpiNad7zzJ40.pgp
Description: PGP signature

Re: triggers wishlist

[Josselin Mouette]

On mer, 2008-04-02 at 05:48 +0200, Michael Biebl wrote:
> Josselin Mouette wrote:
> > On mar, 2008-04-01 at 23:27 +0200, Mike Hommey wrote:
> >> On Tue, Apr 01, 2008 at 11:22:58PM +0200, Josselin Mouette wrote:
> >>> Actually, for icons, I’d say that a package shipping a theme should
> >>> install a trigger for /usr/share/icons/$theme.
> >> Wouldn't that be code duplicated in all theme packages ?
> > 
> > Only hicolor-icon-theme and gnome-icon-theme. Other packages can keep
> > the current code.
> 
> What do you mean by "Other packages can keep the current code" ?

Other *theme* packages. For application packages that ship icons in
hicolor or gnome, we will be able to entirely drop dh_icons.

-- 
 .''`.
: :' :  We are debian.org. Lower your prices, surrender your code.
`. `'   We will add your hardware and software distinctiveness to
  `-our own. Resistance is futile.


signature.asc
Description: Ceci est une partie de message	numériquement signée

Re: Adding lzma to dpkg's Pre-Depends

[Lucas Nussbaum]

On 02/04/08 at 01:52 -0700, Steve Langasek wrote:
> On Wed, Apr 02, 2008 at 10:24:43AM +0200, Lucas Nussbaum wrote:
> > > So of course besides OOo on there we also find the kernel packages.  We
> > > wouldn't have to use lzma for the kernels though, if that would raise the
> > > minimum memory requirements for servers, or lzma could be selectively
> > > enabled on a per-flavor or per-arch basis as appropriate.
> 
> > The results are a bit misleading, because they compare the absolute
> > gain.
> 
> No, why would that be misleading?  You don't want the overhead of lzma
> compression for a 10-fold reduction in the size of a package that's already
> in the bottom 5% of packages by absolute size.

The overhead (time and memory to compress/uncompress) is likely to be a
function of the data size. I'm just saying that it's not that simple.
For example, in the case of OOo, you save 27% of the size, but multiply
by 4 the time to compress.

If instead, you use lzma for ttf-arabeyes, smbclient,
gnome-system-monitor, tomboy, gdb, and language-pack-gnome-en-base, you
save the same size (10.5M in that case, 10.2M in openoffice's case), but
you only compress/decompress 43M of uncompressed data in this case,
while you touch 112M in the OO case. So it's likely that the overhead is
more important in the OO case.

I'm just saying that, if we don't use lzma for all packages, it's not
straightforward to decide for which packages to use it. It's probably a
function of absolute saving, saving ratio, popcon, etc.
-- 
| Lucas Nussbaum
| [EMAIL PROTECTED]   http://www.lucas-nussbaum.net/ |
| jabber: [EMAIL PROTECTED] GPG: 1024D/023B3F4F |


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Adding lzma to dpkg's Pre-Depends

[Steve Langasek]

On Wed, Apr 02, 2008 at 10:24:43AM +0200, Lucas Nussbaum wrote:
> > So of course besides OOo on there we also find the kernel packages.  We
> > wouldn't have to use lzma for the kernels though, if that would raise the
> > minimum memory requirements for servers, or lzma could be selectively
> > enabled on a per-flavor or per-arch basis as appropriate.

> The results are a bit misleading, because they compare the absolute
> gain.

No, why would that be misleading?  You don't want the overhead of lzma
compression for a 10-fold reduction in the size of a package that's already
in the bottom 5% of packages by absolute size.

It's the absolute size savings that counts - that's what's going to have the
biggest impact on mirror size, mirroring time, CD set size, and download
times for users.

-- 
Steve Langasek   Give me a lever long enough and a Free OS
Debian Developer   to set it on, and I can move the world.
Ubuntu Developerhttp://www.debian.org/
[EMAIL PROTECTED] [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Adding lzma to dpkg's Pre-Depends

[Lucas Nussbaum]

On 01/04/08 at 23:49 -0700, Steve Langasek wrote:
> On Tue, Apr 01, 2008 at 01:42:32PM +0200, Lionel Elie Mamane wrote:
> > On the more general issue of lzma-compresses packages, I find a 34MB
> > RAM requirement quite hefty for general purpose use; that is, unless
> > we restrict lzma compression to packages that wouldn't make sense on
> > hardware with so little RAM anyway (such as e.g. OpenOffice.org, but
> > nothing in base, nothing one would install on a pure router /
> > lightweight server such as iptables, kernels, FTP/HTTP daemons, ...).
> 
> lzma compression has to be explicitly enabled on a per-package basis.
> Ideally, it's only going to be enabled on those packages where it makes the
> most sense to do so - i.e., precisely those packages that get the greatest
> absolute size savings by using the different compression method.
> 
> OOo is a great example of a package that benefits.  Here is the analysis
> that was done based on Ubuntu 7.10 to evaluate the benefits of lzma,
> including a list of the top ten binary packages by size savings:
> 
>   https://wiki.ubuntu.com/dpkg-lzma
> 
> So of course besides OOo on there we also find the kernel packages.  We
> wouldn't have to use lzma for the kernels though, if that would raise the
> minimum memory requirements for servers, or lzma could be selectively
> enabled on a per-flavor or per-arch basis as appropriate.

The results are a bit misleading, because they compare the absolute
gain. Sorting by relative gain, I get:
libgl1-mesa-dri_7.0.1-1ubuntu3_i386.deb 78% saved
docbook-xml_4.5-4_all.deb 75% saved
ttf-arabeyes_1.1-9_all.deb 68% saved
foomatic-db-hpijs_20070813-0ubuntu1_all.deb 66% saved
libvte9_0.16.9-0ubuntu3_i386.deb 65% saved
ttf-gentium_1.02-2ubuntu2_all.deb 64% saved
smbclient_3.0.26a-1ubuntu2_i386.deb 64% saved
etc.
-- 
| Lucas Nussbaum
| [EMAIL PROTECTED]   http://www.lucas-nussbaum.net/ |
| jabber: [EMAIL PROTECTED] GPG: 1024D/023B3F4F |


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#107862: All the latest brands available here

[Roan ortega]

7 hobbies of highly effective people

You know you are cool when your boss asks you about that thing on your watch 
that tells time. http://www.retueiag.com/



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]