•   Distro Watch
•   Linux Today News Service
•   Linux and Open Source News, Reviews and Strategy from eWEEK.com
•   News for nerds, stuff that matters
•   The O'Reilly Network ONLamp Articles and Weblogs

Source: LinuxTracker.org

Category: OpenBSD Size: 245.79 MB Status: 1 seeders and 2 leechers Added: 2006-05-12 10:29:44

Source: LinuxTracker.org

Category: PC-BSD Size: 619.15 MB Status: no seeders and 4 leechers Added: 2006-05-12 06:24:49

Source: LinuxTracker.org

Category: dyne:bolic Size: 574.27 MB Status: 6 seeders and 4 leechers Added: 2006-05-12 04:37:00

Source: LinuxTracker.org

Category: QiLinux Size: 4.05 GB Status: 1 seeders and 1 leechers Added: 2006-05-12 04:31:38

Source: LinuxTracker.org

Category: BeleniX Size: 690.76 MB Status: 4 seeders and 5 leechers Added: 2006-05-12 03:00:12

Source: centos

Johnny Hughes has announced the release of the first ever CentOS live CD: "The CentOS Development team is pleased to announce the availability of the first CentOS 4 i386 Live CD. This CD is based on our CentOSPlus kernel and the CentOS 4.3 distribution. It can be used .

Source: rpath

Michael K. Johnson has announced the availability of an updated release of rPath Linux 1 for both i386 and x86_64 architectures: "Refreshed ISO images, release 1.0.2, have been made available for new installations of rPath Linux 1. These images include all updates through and including updates released on .

Source: Linux Today

I think we may have hurt Novell's feelings

Source: Linux Today

The software business will be less profitable in the future--much less profitable

Source: Linux Today

"Sun Micro's new chief, Jonathan Schwartz, has to innovate his way out of the company's agonizing five-year decline. There are few options left.

Source: Linux Today

I'd come to India with the sense that, like Brazil and other countries outside the West that are taking free software seriously, India is moving into a new phase in its use of free and open source software

Source: Linux Today

Draft GNU GPLv3 language would require embedded systems and devices incorporating GPLv3-licensed software to be user-modifiable, stated license author Richard Stallman in a recent speech .

Source: Linux Today

Miro International is continuing to lose developers in its attempt to maintain control of its 'open-source' Mambo CMS

Source: Linux Today

The French Senate bowed to intense lobbying pressure, approving a watered down digital copyright bill that differs significantly from the controversial version passed at the National Assembly last March

Source: Linux Today

Well, it appears to be 1992 all over again. Andy Tannenbaum is once again advocating microkernels, and Linus Torvalds is once again saying what an ignorant fool Andy is

Source: Linux Today

CEO Steve Ballmer said Linux has been outselling Microsoft in areas such as file services, e-mail security and e-science, and that demonstrates Microsoft needs to innovate

Source: Linux Today

The Unix descendent will soon be able to match Linux feature for feature, with GNOME desktop support expected this year, according to a core developer

Source: Linux Today

Most organizations aren't ready to migrate to a wireless, network-centric, thin hardware, server/client model, which makes the Nokia 770 Internet Tablet a bit ahead of its time

Source: Linux Today

Last week we used ping and tcptraceroute to pinpoint connectivity problems, and nmap to spy on users

Source: Linux Today

I haven't yet reached a final decision, but I have some choice words for anyone weighing the idea of starting up a new Internet coffee shop

Source: Linux Today

A lot of folks use Vim, but many exploit only a small percentage of the editor's features

Source: Linux Today

If you prefer to use a computer to manage your passwords, take a look at Password Gorilla

Source: Linux Today

The Nautilus program in GNOME is not only the default file manager, it creates and manages the desktop

Source: Linux Today

The need to balance business continuity with cost has created a new set of storage best practices that blend backup and recovery with leading-edge data replication to accommodate varying information availability requirements

Source: Linux Today

Last year we were first to publish our findings from the OCZ Rally series, and today we are doing the same for OCZ Technology's latest creation--the Mini-Kart Flash Drive

Source: Linux Today

Today's security advisories: mozilla-firefox (Debian GNU/Linux) and MySQL (Gentoo Linux).

Source: Linux Today

LinuxWorld Magazine gets a name change: "Sometimes I worry that I sound like a broken record repeating the phrases Open Source, open standards, and virtualization over and over again like an obsessive parrot "

Source: Linux Today

[You need to show] [h]ow to decide which OpenSource product is really cool and does what it says on the tin and which is vapourwareaspirational

Source: Linux Today

'Public Relations' is one of those funny phrases that has very little to do with what it really means

Source: Linux Today

Ok, I've let the release time between -rc's slide a bit too much again, but -rc4 is out there, and this is the time to hunker down for 2.6.17

Source: Linux Today

"Penguicon, in spite of the obvious Linux reference, is not a conference about computers only " Includes an interview with Google's Chris DiBona.

Source: Linux Today

Linux continues to make highly visible inroads into IT infrastructure, with IDC reporting 14 consecutive quarters of double-digit growth in Linux server shipments through the third quarter of last year

Source: eWEEK Linux

In this week's InfraSpectrum podcast: Peter Coffee examines a plug-compatible alternative to Microsoft Exchange, hosted on Linux and working with standards-based management and collaboration tools, is the proposition offered by PostPath, which emerged on May 9 from stealth mode after two years of development.

Source: eWEEK Linux

Miro continued to lose developers in its attempt to sell and maintain control of its "open-source" Mambo content management system. Now, Mambo's core development team leader Martin Brampton has called it quits. (Linux-Watch)

Source: Slashdot: Linux

simoniker writes "Talking to the Japanese press, Sony executive Izumi Kawanishi has illuminated some of his company's PlayStation 3 Linux plans, indicating that it will be possible for individual 'homebrew' coders to create playable content for PS3, something actively blocked for Sony's PSP handheld. He commented: "Other then game studios tied to official developer licenses, we'd like to see various individuals participate in content creation for the PS3"."

Source: Slashdot: Linux

jbeaupre writes "Groklaw has spawned a new feature that may prove useful well beyond the current SCO litigation. The UNIX Methods and Concepts Database began as a list of books and papers on Groklaw that became so large it demanded organization. The new system was announced and is currently a work in progress."

Source: Slashdot: Linux

An anonymous reader writes "Distrowatch reports - Following some eight months of testing, the openSUSE project has finally released the long-delayed and much-awaited SUSE Linux 10.1: 'After lot of work and several delays, we proudly announce the availability of SUSE Linux 10.1. As usual, we ship all the latest open source packages available at the time. We want to give special mention to Xgl for 3D acceleration on the desktop, NetworkManager for getting painless WiFi access everywhere, the completely open source AppArmor 2.0, and the full integration of Xen 3 in YaST.' OSDir has some great screenshots of the fresh SUSE Linux in the SUSE Linux 10.1 Screenshot Tour."

Source: ONLamp.com

This 30-day project explores the refactoring of a legacy system. The Everything Engine is an aging software
project that powers Perl Monks, Everything 2, and a few other websites.
It suffers from poor design and maintainiability. Learn what it’s like to look
over the shoulder of an experienced developer as he refactors, redesigns, and
updates the code.

Today’s task is finish porting the nodegroup tests. I finally made a decision about how to handle parent test methods.

Day 18: Can I Finish nodegroup?

This is the third day of porting the nodegroup tests. Can I
finish them? I hope so. I’ve certainly earned enough experience that porting
the simple tests goes quickly and debugging the complex tests is easier.
Unfortunately, I just hit the XML tests.

Where did I leave off? 285 tests run, 17 fail, plenty skip, and 65% of the
test is in the new form.

The first test to port is test_field_to_XML(). I think I can
inherit this one, as the only difference is when the $field
parameter is group.

With a quick port run, 293 tests run and 15 fail. One of those is due to the
change in SUPER() handling. After running the parent test method
first, 14 tests fail. Great!

While checking these tests, I noticed that the parent test for
destruct() failed because nodegroup’s implementation
does not return true. Fixed, so 13 tests fail.

xmlTag() is the next method to test. It looks like inheriting
will also work here (and revising the flow of control through the method itself
will make it clearer). After another quick port, 296 tests run and 13 fail. One
of those failures is another SUPER() API change. Fixing that makes
another test pass. Adding test inheritance runs 305 tests, throws some
warnings, and makes the total number of failing tests go up to 15.

Why? Oh, there’s a method call before the SUPER()
call. Mocking that should fix it… but that’s not the problem. The problem is
that a mock in the parent tries to handle multiple calls to the method with a
set_series() mock. Calling the mocked method more times than that
test anticipates causes a problem.

That’s easy to fix; change that method to set_always() and call
it just before each trip through xmlTag(). That makes 305 tests
run and only 12 fail.

With 80% of the test file ported, the next method is
applyXMLFix(). I can probably inherit the test for no fixby node
directly, so that might make code to delete. I can also probably inherit the
parent test with ease.

After the quick porting, 315 tests run and 16 fail. Where did I get four
more failures? One of them is from a SUPER() call. After fixing
that, 315 tests run and only 5 fail. Wow.

Those errors seem to come from a series of calls on the database object,
which I haven’t mocked yet. Is this an easy fix? 315 tests run and 3 fail. One
failure is from this method. It’s because the mocked node doesn’t have a
group attribute. Adding that makes 315 tests run and 2 pass.

Can I inherit the parent test? I think so. After clearing the errors in one
spot in the child method, 315 tests run and 2 pass. Why did 315 run before I
called the parent implementation? I misspelled the test name in the child
method. After fixing that, everything is good again.

The next method to test is clone(). I don’t think inheriting
this will work, but I’m willing to try.

I found a bug in the test, where one test would always succeed. I fixed
it.

After that and a few other porting cleanups, 320 tests run and 1 fail. The
failure is elsewhere. Just for laughs, will test inheritance work here? Not
easily. I’ll skip it for now. (It’ll be easier to make work when testing
against a database.)

With 91% of the tests ported, the next method is
restrict_type(). This method uses the really old pre-Test::MockObject
approach to mocking. Yuck. Wow, the code itself uses the non-OO database
accessors. Double-yuck. No wonder the test code sucks.

Why not clean them both at the same time?

After a few moments of work, 327 tests run and 1 fails. The code is also
cleaner in the method and somewhat cleaner in the test method.

The getNodeKeepKeys() test method ports almost instantly and
inheritance works flawlessly. 331 tests run and 1 fails.

The final method to port is for conflictsWith(). Fortunately,
that’s where the one remaining failure is. With the test ported and a
SUPER() case fixed, 333 tests run and one fails. Ugh, I’ll have
to count to figure out why.

Oh, the problem is that I made a typo and passed an extra first argument to
the method. Now 333 tests run, all pass, and this entire test file is in the
new form. Great!

Inheriting this test doesn’t work very well, as the parent test method
passes a newnode of its own, one lacking an attribute that
nodegroup checks. I could fix that by storing session
data in the Test::Class object
and using that to share information between tests, but that’s more work than I
want to do right now. I’m not sure it’s quite right either.

How do the rest of the tests work? Just fine. 2878 tests run, with only a
handful of expected skips. This is checkin #855, and the end of another
day.