Debian Traffic #13 For 30 Nov 2000

Editor: Zack Brown

By Steve Robbins  and  Zack Brown

Debian Home Page ( | Weekly News ( | Social Contract ( | Constitution ( | Policy Manual (
Developer's Reference ( | Documentation Project ( | debian-devel Archives (

Table Of Contents


Want to help write KC Debian? See the KC Authorship page (../author.html) the KC Debian homepage (index.html) , and the Thread Summary FAQ (../summaryfaq.html) . Send any questions to the KCDevel mailing list. (

Mailing List Stats For This Week

We looked at 426 posts in 1595K.

There were 187 different contributors. 79 posted more than once. 0 posted last week too.

The top posters of the week were:

1. Women In Debian

16 Nov 2000 - 19 Nov 2000 (35 posts) Archive Link: "women in debian"

Summary By Zack Brown

People: Carlos LaviolaAn Thi-Nguyen LeSean PerryDiana GalletlyS.J. BlackJan Martin MathiassenBranden RobinsonJulian StoevMagni Onsøien

Carlos Laviola asked, "how many female developers there are right now in debian, if any?" An Thi-Nguyen Le replied that she thought there were a few, although she herself was still in the process of becoming one. She added, "I can't see how this matters all that much though. We're all just dudes who happen to work on Debian." Sean Perry felt that Debian was the "epitomy of the all guys testosterone engineering groups" and mentioned "We fight, angle for superiority, and generally act like animals fighting for dominancy" as evidence. Diana Galletly said at one point, "I can't say I've noticed many differences in attitude between debian-devel and my normal social circle. But then, that's hardly surprising ;-)." She added, "Some women do want to be a part of Debian; I'm one of them (although I'm shamefully lacking in time at the moment, and Debian's not top of my priority list, I'm afraid)."

Elsewhere, Sean said, "Open Source is for many people about ego. "Look at what I did", or "I could write that better than they did". This is a very male tendency." He added, "my experience in the CS world is that women find hacking on minutiae because it is fun quite boring actually. Most of them are more of the "does it work, good enough". As I said, this is from my own college and work experience. Men in computers seem willing to devote hours to trivial matters or work hard and long while many bitch and complain. Few women in computers I have met cared."

S.J. Black said, "fellas don't have a copyright on ego. There are a lot of women who, for various reasons, trade in it too - but, like wizards (or sysadmins!), are subtle...and many have not learned the value of the "direct approach" in their daily pursuits." And added, "There are both guys and women who will sell someone else down the river for their own ends. It's not a specifically gender-based tendency, nor is it without its place. But it's not best-used in general information exchange, nor is it all that helpful in the school or the workplace on a daily basis." And concluded:

It'd be a very cool thing to see more women getting into technical anything...a good many don't for the simple reason that they've been raised to believe it's not ladylike, or that they ought to be thinking more along the lines of socially-related occupations, or artistic pursuits, or any number of stupid and destructive beliefs about females in non-traditional occupations/hobbies/ whatever.

The guys who want to see more of a balance between the genders in debian/ tech in general might want to let their sisters, girlfriends, moms, female acquaintances know that they *can* learn to do tech stuff, that their company in such pursuits is welcome. Encourage confidence in ability.

Show patience when they learn. Women might try to get each other turned onto Open Source projects, or even collaborate to get some new ones going. Whatever. It boils down to interest, and to a welcome environment in which to exercise that interest.

Later, S.J. asked, "I wonder if more women code in Java than in C/C++? Or whether VRML is the language of choice?" Jan Martin Mathiassen replied, "women code in java/c/c++, men code in assembler (and cry a lot). :)" S.J. replied:

See, and i can tell you from personal experience that most women are more inclined to boom obscenities at their non-compiling code than to cry. But then, even when we do venture into doing assembler, we instinctively know that any machine that understands and obeys such language is impervious to tears. 8)

We, however, are not. <g>

Elsewhere, someone mentioned that his "partner" used 'mh' for email and 'vi' for editing. Branden Robinson replied, "No wonder this guy couldn't get around to packaging perl 5.6. His sex life must be great, with them moaning ed scripts at each other and quoting RFC's during orgasms..." Diana replied, "Funnily enough, geek and girlgeek tend to leave geekiness behind when in bed ... at least, they do round here (damn, I'm not on #chiark, perhaps I ought to censor this email .... ;-))"

At one point, Julian Stoev said, "I am currently trying to show Debian to my wife. It works slowly by slowly. But I really can't imagine she is going to read 10 mail lists. This is a problem, because I got most of my experince browsing Internet and asking on mail lists. I am afraid I will have a problem to explain her, that this is the best way. She reads the book, but IMHO she will have to ask people when she does not understand something more complicated. Maybe this is something specific to males?"

Magni Onsøien replied, "I am currently trying to show Debian to my brother. It works slowly by slowly. But I really can't imagine he is going to read 10 mail lists. This is a problem, because I got most of my experince browsing Internet and asking on mail lists. I am afraid I will have a problem to explain him, that this is the best way. He reads the book, but IMHO he will have to ask people when he does not understand something more complicated. Maybe this is something specific to sisters?"

2. Debian Doc Site

19 Nov 2000 - 20 Nov 2000 (6 posts) Archive Link: "Call for articles, tips, & HOW-TOs"

Summary By Zack Brown

People: Randy EdwardsJordi Mallach

Randy Edwards announced:

A new web site has been started called "debianHELP" at

DebianHELP's name is self-explanatory and the site is a slash-like format designed to feature helpful articles/advice on running Debian GNU/Linux, and for people to trade tips/carry on conversations to solve problems. For a more verbose idea of the site, read <>.

The purpose of this message is to solicit articles and helpful advice for the web site.

These articles will hopefully be more likely to be read compared to some of the gems which come through the mailing lists and are tucked away in the list archives.

What kind of articles are needed? Everything! These don't have to be in-depth articles, but anything that shares your experience and would help a rookie Debian user. For example, here are some thoughts:

Can you think of any other ideas? Go for it!

It doesn't matter what your level of expertise is -- believe me, I'm sure there's knowledge you have that you can write up and share which would benefit others. If you feel the same way, drop by <> and post an article and share the wealth.

Jordi Mallach and others pointed out that this was much the same as the project. Some folks felt this was unnecessary duplication of effort, while others felt there was nothing wrong with having multiple Debian-related forums. But there was not much discussion at all.

3. Problems Uploading Packages

20 Nov 2000 - 24 Nov 2000 (10 posts) Archive Link: "Query: Is upload queue stalled?"

Summary By Zack Brown

People: Paul MartinMark BrownJames Troup

Paul Martin reported, "I made an upload to master on Friday, and I haven't had the usual "files moved to othermachine" email confirmations, and the files haven't appeared in incoming. Has something got stuck, or is it just me?" James Troup explained that the relevant install daemon hadn't been restarted when 'master' had been rebooted. Mark Brown and Petr Cech also added that developers should upload to 'auric' instead of 'master' anyway, since uploading to 'master' would just go from 'master' to 'samosa' to 'auric', giving the same result after more indirection.

4. A fix-pending Tag Does Not Close the Bug

21 Nov 2000 - 22 Nov 2000 (6 posts) Archive Link: "request for two new BTS tags"

Summary By Steve Robbins

People: Adam Di CarloManoj SrivastavaJulian GilbeyJosip Rodin

Adam Di Carlo suggested:

I would like two additional BTS tags:
bug is a documentation bug
bug has been fixed in the maintainer sources, but that isn't yet uploaded; this is very useful for packages like boot-floppies where there are many people working from CVS for the next version -- it lets them know not to both fixing a given bug.

There was a short discussion about the length of the tag name fixed-in-next-version. Using fixed was suggested and rejected, since there was already such a tag, for items that bugs that were fixed by e.g. an NMU. The term pending met with some approval.

There was some confusion over the difference between a tag and a severity. Manoj Srivastava, voiced his concerns:

The concern I have is this: Suppose someone reports a grave/important bug; and I fix it in my CVS sources; can I then just change the severity to fixed-in-next-version? (or pending, or whatever).

But the release management process is strongly tied to the severity levels grave/critical/important; and the pending severity should not be a loophole/end run around that.

In other wirds, one should not be able to downgrade high severity level bug reports just because we have it fixed on some random machine out in the known universe.

In fact, tags and severities are orthogonal, so adding a pending tag does not get around the release management. As Julian Gilbey said, a bug with a pending tag " retains its severity, but is noted that it is fixed in CVS or wherever. " Josip Rodin pointed Manoj at ( .

5. Original Source Must Be Properly Named

24 Nov 2000 - 26 Nov 2000 (8 posts) Archive Link: "dpkg-buldpackage thinks radiusd-cistron is debian-native"

Summary By Steve Robbins

People: Norbert VeberAdam HeathAdrian BunkSteve Robbins

Norbert Veber was confused, and asked, " How does dpkg-buildpackage determine the source version? I dont know why, but it keeps thinking that my package is debian native, and doesnt produce a package.orig.tar.gz or a diff, instead it just builds it as a debian native package.. "

Adam Heath suggested looking at the version string in debian/changelog, " If it doesn't contain a -, then it is a native package. " In Norbert's package, however, the version did have a dash.

Adrian Bunk came through with the answer, " Is there a ../radiusd-cistron_1.6.3.orig.tar.gz? If not that's the reason why dpkg-buildpackage think's it's a native package. "

It later emerged that Norbert had the original source in a file named ../radiusd-cistron-1.6.3.tar.gz. He thought this was sufficient because it had worked on another package that he debianized using dh_make. What escaped his notice, however, is that dh_make did the renaming of ../foo-x.y.z.tar.gz to ../foo_x.y.z.orig.tar.gz.

(ed. [Steve Robbins] The moral of the story is: read and re-read the packaging manual. I got caught by the same mistake recently! )







Sharon And Joy

Kernel Traffic is grateful to be developed on a computer donated by Professor Greg Benson and Professor Allan Cruse in the Department of Computer Science at the University of San Francisco. This is the same department that invented FlashMob Computing. Kernel Traffic is hosted by the generous folks at All pages on this site are copyright their original authors, and distributed under the terms of the GNU General Public License version 2.0.