| DistroWatch Weekly
|DistroWatch Weekly, Issue 98, 2 May 2005
Welcome to this year's 18th issue of DistroWatch Weekly! If you are losing patience waiting for the ISO images of the new Mandriva Linux 2005, why not perform a network installation instead? It is easy and we'll show you how. Will will also address the issue of the never-ending stream of new distributions vying for our attention, and point out some serious problems with the latest release of the GCC compiler. In the biggest DistroWatch Weekly ever, we have a user-contributed review of the increasingly popular SLAX live CD, and present FetchYahoo in Robert Storey's "Tips, tricks & hints" section. Happy reading!
Install Mandriva Linux 2005 today
Although there is a rumour saying that the ISO images of Mandriva Linux 2005 Limited Edition should hit the public download mirrors early in May, if you have a broadband connection, there is no reason to wait. As a matter of fact, you can install Mandriva Linux 2005 today, directly from one of the FTP or HTTP servers. Depending on the speed of your connection, activity on the download server, and the number of packages you need, the system installation will take around 2 - 4 hours. If you start it just before you retire for the night, it will more than likely complete by the time you wake up in the morning.
The FTP/HTTP installation is pretty straightforward and intuitive. First, you need to download the bootable installation CD from one of the Mandriva mirrors, e.g.: this one: boot.iso. This is a 14.2MB file that includes the installer, together with most network kernel modules. While you are on the download page, choose one of the mirrors near you and write down the exact patch to the installation directory - you will need that information later. Now, burn the ISO to a CD using your preferred CD burning software, then place it into your CD-ROM tray and reboot your computer.
Follow the instruction on the screen, which shouldn't be too difficult for most users. The only tricky place is where you need to type in the FTP or HTTP server containing the Mandriva Linux distribution. To continue with our belnet.be example, this is what you'd need to type in (leave the login and passwords fields empty):
• FTP Server: ftp.belnet.be
• Mandrakelinux directory: packages/mandrakelinux/devel/2005/i586
The above entries will of course vary, depending on your chosen mirror. The rest of the installation process differs very little from standard installation from ISO images, except that it will take longer.
Despite its status as a "transitional release", we found Mandriva Linux 2005 an impressive distribution. Although in terms of packages included it is not as up-to-date as the recent SUSE or Ubuntu releases, it is clear that Mandriva has benefitted from the long public testing period and the experience of its developers. Additionally, Dr Zhu, our regular contributor in Beijing, reports that Mandriva 2005 has implemented substantial improvements in rendering of Asian fonts, which are now remarkably crisp and good-looking. Overall, Mandriva Linux 2005 is a fast and powerful distribution, certainly worth considering as your primary Linux desktop.
Mandriva Linux 2005 - an impressive release, despite the distribution's "transitional" status
(full image size: 210kB)
* * * * *
Are there too many Linux distributions? Linux.org has posted an interesting analysis of the current situation on the distribution front and explains why the growing number of Linux distributions might not be a good thing for the adoption of Linux on a larger scale: "But there is a downside to it that really needs to be addressed. That is that there are now too many distributions and the ability to identify exactly what Linux is and to 'sell' it may become more difficult when it should be getting easier. I think developers need to ask themselves whether it might be more advantageous for the community as a whole for them to volunteer on existing, popular distributions rather than to create their own."
This is certainly a valid point. At DistroWatch, we currently list over 350 active distributions, with over 100 more on the waiting list. The sad truth is that the majority of these projects will undoubtedly die after the initial enthusiasm for the project is over. As such, most distributions are now developed for the sole benefit of the developers themselves - to learn and to get some Linux experience. Judging by the easy manner with which some of these project get abandoned, we, the users, matter very little.
What do you think? Should we urge the developers of new distributions to join an existing project, rather than develop a yet-another-distribution? Or do you believe that new distributions can bring new ideas to the market and therefore should be encouraged? Please discuss below.
* * * * *
The recent release of GCC 4.0.0 was anticipated with much enthusiasm. The new version was expected to be a major improvement over the 3.x series, especially in terms of support for the increasingly fashonable 64-bit processors. As always, distributions are slow to adopt the new compiler release until it has been through its paces and, possibly, until a new point release fixes some of the more obvious problems.
Unfortunately, it seems that GCC 4.0.0 has not been well-tested before it was declared "stable". That's according to this post by the Ark Linux development team: "GCC 4.0.0 should have received a lot more testing before being released as 'stable'. It is currently known to mis-compile the KDE desktop as well as parts of the Linux kernel (this is already fixed in CVS). Also, while it is a major step forward in terms of optimization infrastructure, it doesn't bring a lot of things that are helpful to anyone but developers of GCC itself - in our tests [with a slightly pre-4.0 CVS snapshot], the speed was not much better than with earlier GCCs, but the binaries got about 15% bigger." More information about the tests by the Ark Linux developers can be read here.
|Featured distribution of the week: SLAX
Review of SLAX 5.0.4 by Ralph Patterson
Live distributions are like living creatures. They need to be fed and nurtured to survive and be useful. Without utility and evolution, operating systems and software generally die an early Darwinian death - remember all the Win9Xs. SLAX, KNOPPIX, and others capitalize on the ideology of being as small as possible without compromising functionality. This "without compromise" thought in many respects is of an advantage as the general movement of live distributions seek to put as robust features as possible into their packaging, making them some of the most useful of operating systems. SLAX is one of these uncompromising examples of a live distribution of Linux.
Using Slackware as its base, SLAX has a minimum of requirements; 30MB to boot, 64MB to run X window and 128MB to run X window with KDE, and a 486 processor or better. As a live distribution it can be launched using an IDE CDROM or a USB flash disk. Despite its size, it has all the open source software that the average PC user will ever need. It is an excellent alternative to a Windows based system given its portability and ease of use. It can literally be used on just about any machine with sufficient RAM. It is fast to load, easy and fun to use. SLAX will work from a command line or GUI interface. It is simply too cool of an operating system to be ignored.
As with most operating systems, revisions are inevitable. SLAX is no different with respect to having a healthy changelog with one exception - the speed at which fixes are generated and the distribution is updated. Not having dug into the guts of the differences between the distribution bases (be it Debian, Slackware, etc), the ease of this is not understood by me (and I suspect neither by others that haven't taken the time to evaluate the subdermal structures of the organs behind the various species of penguin they are using). Nevertheless, Tomáš Matějíček and an ambitious forum crew (Endymion, dustin_wielenga32, specter, pekster, and others), which Tomaš gives credit to, are quick to turn out updates and corrections. An excellent example is the 5.0.2 revision which didn't recognize the P/S2 mouse, Tomaš and the forum crew dug in and cranked out 5.0.3 less than a day later. So fast in fact, I left one evening thinking my mouse had gone the way of the PCs hard drive months before, only to arrive the next morning with an update posted on DistroWatch. Not to be outdone, 5.0.4 quickly followed with an installer, remount and sync options. SLAX has various flavors within itself including a "KillBill" version, all keeping within the frame work of fitting on a 200MB mini disk.
I've been using SLAX on and off now for about a year, lately on a PIII-450 which suffered a hard drive failure approximately 8 months ago and was about to be chit-canned by our IT department. Most in the lab thought I was crazy when I argued to keep the machine in light of the fact IT was refusing to order (or give ... they have a ton of surplus hard drives) a new drive for the aging (?) box. I wanted to keep the machine mostly because I'd scavenged a burner and extra memory mere days before its hard drive demise. Everyone, IT included, thought I was mad to want to keep a machine which appeared unbootable, let alone usable. They were amazed when I popped a copy of SLAX 5.0.1 in and began working on my graduate thesis proposal right were I left off in MS Word using KWrite. Given that I like to customize my desktop and have made configuration changes, IT thought they had me with a flaw in the live CD design; that is, until I saved my data and configuration changes to my thumb drive (configsave /dev/sda0 from a terminal window) ... makes you wonder if an MCSE is really worth having. IT, needing to get the last word in, wondered off stating they do not offer support for linux ... who needs 'em anyway when you've got such an active users' forum?!
If you haven't tried SLAX, it's time to give it a test drive. I've found it insurmountable in my arsenal of Linux distributions. If you find it useful, please pay homage, at the very least, by sharing it with someone else. On our modest college campus here in western North Carolina (USA) I'm finding enough interest to the point I may even be able to start up a small LUG before I graduate. It will definitely be one distribution available for viewing at our next Software Freedom Day. The SLAX forum can be found here. Check out the forum, the time for downloading SLAX is relatively fast given its small size, I think it is worthwhile and that you'll like what you'll find.
|Released Last Week
A new version of ParallelKnoppix has been released. What's new? "New version based on Knoppix 3.8.1. Setup has been streamlined quite a bit, and the dialog boxes have a much nicer font; tutorial has been updated to explain the new setup method, and to cut down on toner usage when printed; bootstrapping example for Octave (now at 2.1.69)." Read the release announcement on the distribution's home page.
Another new version of SLAX is out, just two days after version 5.0.3. From the changelog: "SLAX updated to Slackware current; fixed Kopete, it doesn't crash anymore (updated libxslt); recompiled KDE, added cyrus-sasl support; fixed SLAX installer, added remount, sync option; added MPlayer 1.0pre7; firewall is not started automatically anymore."
Foresight Linux 0.8
A new version of Foresight Linux has been announced on the project's Wiki pages: "We are proud to present you Foresight Linux 0.8! Reacting on critics we increased the speed of the installation process. It is up to 60% faster! We also included a new boot splash and did lots of cosmetic things. Old and new highlights are 'Conary' (revolutionary package management program) and nice tools like Beagle and F-Spot. Also included is the latest version of Oversite systems management tool, which now has a GUI configuration interface. It is still under development, but is currently very functional for keeping you system up to date!" Here is the full release announcement.
Mandriva Linux 2005 (PPC)
The PPC edition of Mandriva Linux 2005 Limited edition has been released in the form of three ISO images: "The new release has been on the mirrors for a few days already but here's the unofficial announcement: Mandriva Linux Limited Edition 2005 for PPC is available! It looks like we have a new name, an inflated version number, and a new way-too-happy penguin. Also Linux kernel 2.6.11, GNOME 2.8,3, KDE 3.3.2, X.Org 6.8.2, GCC 3.4.3, and much more on three 700MB ISO images. If that's not enough, there's 7+GB of packages on the mirrors." Find more information in the release announcement.
Frugalware Linux 0.2
Frugalware Linux 0.2 has been released: "The Frugalware Developer Team is pleased to announce the immediate availability of Frugalware 0.2, our second stable release. A short list of the most important improvements and bugfixes since 0.2rc2: updated Mozilla Suite to 1.7.7 and MPlayer 1.0pre7; new packages - irda-utils (utilities for infrared communication between devices) and frugalrledit (runlevel editor for Frugalware Linux); added patch to binutils to ignore as_needed libs that were not needed. Please refer to the Frugalware ChangeLog for more information." Here is the full release announcement.
Frugalware 0.2 - although loosely modelled on Slackware Linux, Frugalware Linux comes with its own graphical utilities and sophisticated package management
(full image size: 333kB)
Onebase Linux 2005
Onebase Linux 2005 has been released: "The Onebase Linux Project is proud to announce the 'Project Re-launch' by introducing version 2005. After 5 months of intense development and testing we made a hallmark improvement and brought numerous features. Our focus was not on bringing yet-another-Linux by simply repackaging or showcasing déjà vu software. Instead, Onebase Linux 2005 brings a new face of Linux with its improved file hierarchy, boot system and addition of Onebase Portal. The following link gives information about the prime features present in this release." Read the full release announcement for additional details.
* * * * *
Development and unannounced releases
|Upcoming Releases and Announcements
Daemon News reports about the delay in the much awaited release of FreeBSD 5.4: "FreeBSD Release Engineering Team's Scott Long says that they are a bit behind on FreeBSD 5.4 release. The stability problem reported several weeks ago in a particular high load, high profile environment has been resolved and the release process is under way now." FreeBSD 5.4-RELEASE is now expected to hit the download mirrors on 7 May; here is the updated release schedule.
The Vidalinux development team has published details about the upcoming release of Vidalinux 1.2: "This is the full list of features that VLOS 1.2 will have. The release date is June 20, 2005, this release will include a high range of features that will make this distribution very powerful: GNOME 2.10; KDE 3.4.0; kernel 2.6.11; bug fixes and security updates; reworked Anaconda installer with new features; two CDs installation; new Portage GUI application with lot of new features; new artwork and logos; PPC platform support; 3D accelerators drivers installed by default (ATI, NVIDIA); new binaries repositories and VLOS rsync server; more wireless-friendly...." Find more information in this announcement.
StartCom Enterprise Linux 4.0
StartCom has announced that development has started on StartCom Enterprise Linux 4.0, a distribution rebuilt from source RPMs of Red Hat Enterprise Linux: "StartCom Enterprise Linux 4.0.x is under development and will be released soon, starting with the x86 (i686) architecture, followed by a release for AMD's Opteron 64 bit (x86_64) CPUs. You may want to browse the RPM trees at one of the download mirrors to see the progress for this release, code named "Barak", meaning lightning in Hebrew. " No fixed roadmap has been given, but you can read the announcement on this page.
* * * * *
Summary of expected upcoming releases
|Web Site News
Donations: NdisWrapper receives US$250
The recipient of our April 2005 donation is the NdisWrapper project. For those that are new to this web site, just a quick re-cap: DistroWatch.com donates 10% of income from advertising on the site to various open source software projects on a monthly basis. Readers are welcome to nominate their favourite open source projects - either by mentioning it in the forum below, or by direct email (which can be found at the bottom of every page).
If you are using a wireless network card on your Linux-based operating system, it is likely that you have heard of NdisWrapper. Since many wireless network card manufacturers do not release specifications of their products, the NdisWrapper project has developed a Linux kernel module that loads and runs Ndis (Windows network driver API) drivers supplied by the vendors. This project has been a life-saver for many notebook users who would otherwise not be able to take advantage of their notebook's built-in wireless network capabilities under Linux. It goes without saying that the NdisWrapper developers have considerable expenses while testing various wireless network cards and we hope that our donation will help them cover some of the cost.
As always, our donations programme is a joint initiative between DistroWatch and LinuxCD.org, which contributes US$50 every month. LinuxCD.org is an online store selling low-cost Linux/BSD CDs - they have the largest selection, inclusive of all the latest releases, and they offer the lowest prices. Next time you need to order your favourite Linux or BSD CDs, get them from LinuxCD.org.
Here is the PayPal receipt for our donation:
This email confirms that you have paid OSDN / VA Software $250.00 USD using PayPal.
Transaction ID: 9RA696345P350230W
Total: $250.00 USD
Item Title: Donation
Invoice ID: 234835
Message: This is a donation by DistroWatch.com as part of our programme to support OSS projects. Keep up the good work!
Here is the list of projects that received a DistroWatch donation since the launch of the programme:
* * * * *
New distribution additions
* * * * *
New on the waiting list
- Decebal Linux. Decebal Linux is a new source-based distribution with tightened security; it includes the ProPolice stack smashing protector and other patches. The distribution's web site is still under construction and lacks more detailed information, but a release candidate of Decebal Linux 0.9.0 is already available for download and testing.
- MOPS Linux. MOPS Linux is a new Russian distribution based on Slackware Linux. The product includes several packages that don't feature in Slackware itself; these include geodata libraries, PostgreSQL, Squid, Xmail, Webmin, OpenOffice.org, and other software.
- Muriqui Linux. Muriqui Linux is a new Brazilian distribution based on Debian, but with the port of Red Hat's Anaconda installer. It is developed by Instituto Doctum de Educação e Tecnologia
- Pentoo LiveCD. Pentoo is a Linux live CD, which boots into a Linux environment on any x86-based PC. It is similar to KNOPPIX, except that Pentoo is based on Gentoo Linux. The main purpose of Pentoo LiveCD is penetration testing.
- SUTLinux. SUTLinux is a Thai Linux distribution and live CD based on Thailand's popular OpenTLE project.
* * * * *
DistroWatch database summary
- Number of Linux distributions in the database: 402
- Number of BSD distributions in the database: 11
- Number of discontinued distributions: 49
- Number of distributions on the waiting list: 110
|Tips, tricks & hints (by Robert Storey)
I was at much pains to describe to him the use of money, the materials it was made of, and the value of the metals; that when a Yahoo had got a great store of this precious substance, he was able to purchase whatever he had a mind to; the finest clothing, the noblest houses, great tracts of land, the most costly meats and drinks, and have his choice of the most beautiful females. Therefore since money alone was able to perform all these feats, our Yahoos thought they could never have enough of it.
-- Lemuel Gulliver, in Gulliver's Travels
* * * * *
When I was just a young lad, I thought the term Yahoo! was simply what a cowboy yelled while trying to ride a wild bucking bronco. I wasn't until much later that I read Jonathan Swift's Gulliver's Travels (a great book) and found out what Yahoo really means. However, to most people nowdays, Yahoo! is the name of a web portal and email service.
Thus, no prizes for guessing that FetchYahoo is the name of a utility that can fetch mail from your Yahoo account, thereby eliminating the need for you to use a web browser for mail access. FetchYahoo downloads the mail from your Yahoo inbox and delivers it to your local machine's mail spool (usually found in directory /var/mail or /var/spool/mail depending on your distribution). You can then read your messages with a mail client like Kmail, Mutt or Sylpheed (assuming you've set these up to read local mail - more on that below).
FetchYahoo sounds a lot like Fetchmail, and you might wonder why you couldn't just use the latter to do the same thing. In fact, you could - if you pay for the privilege. Yahoo no longer offers free POP service, but it is available for a fee. FetchYahoo is a Perl script that cleverly circumvents this need. Of course, it may well be worthwhile paying for Yahoo's mail service (to gain more server space and other features) - that's just a decision you'll have to make.
Few (if any) distributions come with FetchYahoo pre-installed, but it's easy to obtain. Users of Debian-based distributions can do an apt-get install fetchyahoo, but it should be simple enough to find a RPM binary as well. In FreeBSD, the port resides in /usr/ports/mail/fetchyahoo. If your distribution doesn't have a packages, just go to the source. Being written in Perl, FetchYahoo is easy to install on Linux and BSD (indeed, people have even gotten it to run on Windows).
Once installed, your next adventure is to configure FetchYahoo to your needs. The installation procedure should have left you with a sample configuration file - at least in the case of Debian, the file is found here:
Copy the fetchyahoorc example file to your home directory, rename it .fetchyahoorc and set it's permissions so that only you (the user) have read/write access:
chmod 600 .fetchyahoorc
Then with your favorite editor, open the file. Most of the settings are optional and self-explanatory, but a few are absolutely mandatory. You will, of course, need to fill in the following settings with correct info:
###### SHOULD configure these ######
username = yahoo-user-name
# this can be a password or an md5_hex hashed password
password = yahoo-password
The next section is concerned with configuring your mail spool. You will probably want to use mail spooling (that is, Fetchmail-style local delivery of messages to your machine), but another possibility is that you will instead choose to forward the messages elsewhere (that's configured further down in the .fetchyahoorc file - we'll get to that shortly). For example, if you normally collect all your email from my_name@my_isp.com, you could forward all your mail to there and disable spooling.
###### mail spool, mbox file and procmail configs ######
# set use-spool to 0 to disable outputting to a file/filter
use-spool = 1
# if spoolName ends with a / we output in maildir format to that directory
spool = /var/mail/local-user-name
Note that on many distros, the mail spool may reside in /var/spool/mail rather than /var/mail. As for the local-user-name above, that's your usual logon name (so in my case I set this to "/var/mail/robert"). If you've never set up mail spooling before, check your /var/mail (or /var/spool/mail) directory and see that your logon name is there. If not, as root create the necessary file and set the correct ownerships and user permissions (in my case):
chown robert:mail robert
chmod 660 robert
Finally, note that mail spooling requires that you have a Mail Transfer Agent or MTA (such as Sendmail, Postfix, Exim, etc) installed, configured and running in the background. In years past, setting up a MTA was a big trauma, but these days many distributions do it for you. Configuring a MTA is beyond the scope of this article, and the only thing I'll say about it now is that I recommend Postfix. Take a look at the Postfix website if you need more info. Also, you'll need to set up your mail client (Kmail, Sylpheed, etc) to collect mail from the spool file.
At this point, you've done enough to start downloading your mail from Yahoo. However, you do have a number of other useful options further down in the file, and it's worth taking a look at those. You may, for example, want to empty the Bulk folder automatically, since it tends to be filled with nothing but spam. Do that by changing the following setting to 1:
# set this to 1 to enable emptying the Bulk folder before fetching messages
empty-bulk = 0
Although not strictly necessary, you can also automatically empty the Trash folder. FetchYahoo gives you the option of emptying the trash before or after you check the mail, but I found that emptying the trash before checking didn't work. Emptying it after worked fine. You'll find the settings here:
# set either of these to 1 to enable mail checking before or after downloading msgs
empty-trash-after = 0
empty-trash-before = 0
I did mention above that you could forego mail spooling altogether and instead choose to forward your mail to your regular email account at your ISP. Take a look at the following settings if you wish to enable this:
###### mail forwarding configuration ######
# set use-forward to 1 to enable mail forwarding
use-forward = 0
# set mail-host to your smtp outgoing mail server
mail-host = outgoing.example.com
# the e-mail addresses you want mail forwarded to
send-to = firstname.lastname@example.org , email@example.com
# the e-mail address used as the from address, this should probably be at the
# same ISP as the outgoing smtp mailhost specified above
send-from = firstname.lastname@example.org
There are still more possible tweaks, but the above should be enough to keep most users happy. To actually download (or forward) your Yahoo mail, just run fetchyahoo from the command line. Hardcore geeks might want to set up FetchYahoo as a cron job to check for mail at specified intervals.
One final tidbit before departing - many have wondered if there is an equivalent utility for Hotmail. Indeed there is, called Gotmail. However, Hotmail is owned by Microsoft, and consider this message from the Gotmail maintainers: "Although the team behind Gotmail tries to keep it working, every few months Hotmail makes changes which break this until they have time to fix it. Have a backup plan for email access in case that happens."
An inadvertent and totally coincidental oversight by Microsoft, I'm sure.
-- Robert Storey
|DistroWatch in the News
DistroWatch on Microsoft.com
Folks, we have done it - DistroWatch is now linked directly from a web page at Microsoft.com! It is a page discussing computer security (of all pages) and although the article has generated negligible amount of traffic for this web site, it still represents an interesting point of view. Naturally, don't expect much flattery on account of Linux, but don't dismiss the article either - the author presents some interesting points. More details in Security Management: When an Issue Affects Multiple Products.
That's all for today. We hope that you enjoyed this week's DistroWatch Weekly!
If you've enjoyed this week's issue of DistroWatch Weekly, please consider sending us a tip.
(Tips this week: 0, value: US$0.00)
|Reader Comments - Jump to last comment
1 • As Always.. (by sn0n on 2005-05-02 10:17:17 GMT from United States) |
... another great article :-D
you really should go daily with these things.. ;-)
then i wont have so much boredom at 3:15 am ;-)
2 • Incorrect Link to Rubix (by IMQ on 2005-05-02 10:40:40 GMT from United States)
The link actually points to Arabian.
3 • too many distros (by Scott on 2005-05-02 10:42:50 GMT from United States)
Yes, there are way too many Linux distributions! Several years ago, I had fun testing various ones in Distrowatch's top 100 list. Now days, I pretty much just pay attention to the top 25 distros in that list, and even then, I rarely ever install anything below the top 10 on any of my machines, as I just don't have the time to try them all out anymore. If there's ever going to be any measurable gap closed between the number of Linux users and the number of Windows users, there needs to be an industry-wide focus on just a top few REALLY GOOD distros. I would say just stick to the top 10. The company's that sell their own flavor of Linux can continue to develop, but the volunteers really should try to stay focused on just a few of the best free ones. I would like to see Distrowatch have two seperate lists. The top 10 truly free distro's. And a top 10 list of non-free, commercial distro's.
4 • Number of distributions (by Shabani on 2005-05-02 10:48:26 GMT from Canada)
It's useless to ask people to contribute to an existing project instead of starting their own. This complaint has been voiced many times to no avail. I just hope that one day, people will realize that having 400 linux distros and 11 BSD ones is nonsense. By the way, I still haven't figured out what are those 11 BSD distributions, beside Free/Open/NetBSD and Monowall.
The apparent lack of testing of GCC 4.0.0 points to another problem : it's hard to understand why some distribution maintainers feel compelled to release their product 2 or 3 times in a week. For instance, it's common to see different versions of Slax or Damn Small Linux released in less than 7 days. To be polite, I will just say that quality writable CDs aren't cheap.
To be a little out of topic, I didn't give my email address because I don't know if spammers read this newsletter and I'm tired of being bugged by people trying to sell me Viagra, put me in contact with lonely housewives or give me the latest Wall Street hot tip.
5 • Too many distributions (by Jan Stedehouder on 2005-05-02 11:10:07 GMT from Netherlands)
Definitely I would encourage more cooperation and less personal scratching of itches, not just at the distribution level but also at the package level. Linux/ Open Source is all about choice, about the ability to take the source and improve on it. As such, a great concept with very good results. However, when really looking deeper into the various distributions the real differences are extremely small and boil down to the personal preferences of their followers. At the software level we can see thousands of pieces of software at place like Sourceforge and Freshmeat, many of them abandoned or under supported. Through the internet communities can be build with ease, but -please- focus on joining one of the existing communities before starting your own.
6 • About the high number of distros. (by Steve on 2005-05-02 11:32:06 GMT from United Kingdom)
This causes confusion – and the way to fight confusion is to introduce simplicity and consistency.
A rating scheme! Not for the features – rather, the way they implemented. For example, with 1 top / 5 lowest:
1. functional stability / uptime from F5 (broken!) to F1 ( no defects)
2. security hardness from S5 to S1 (SE like, no root user etc)
3. security fix responsiveness from R5 (takes over a year) to R1 (top – within 2 days)
4. size and strength of vendor commitment from V5 (one person) to V1 (10,000 plus dedicated staff)
5. target market T1 – corporate infrastructure, T2 – web / internet infrastructure, T3 – specialist industry, T4 and T5 can't think of anything. Hobbyist?
So a supplier of a perfect product for the corporate market would have a rating of F1S1R1V1T1 let's make that 1-1-1-1-1.
You would not want to bother with a distro rated as 5-5-5-5-2 !
Rating would happen by a poll at Distrowatch – it should be kept unpublished until it has at least 50 poll ratings (for some statistical strength).
My feeling is that this idea, right now, is getting somewhere - but is not actually good enough yet. It needs more work – improvements and more ideas plus extra simplicity.
7 • PearPC (by Reuben Perelman on 2005-05-02 11:47:04 GMT from United States)
I think Distrowatch should donate some money to the PearPC project legal fund, to sue some GPL violating rip-off of their programing.
8 • distro numbers (by mark on 2005-05-02 11:54:40 GMT from United States)
I think that the numbers are high
But there is some good in this
Look at windoze when your alone
You become the only target
For the last 3 months I have run centos4
B4 that it was mephispro
B4 that fedora with alot of tests along the way
I learned alot from them
But the reason Linux is not King isnt cause there are too many
Its alot of things doze fud,lack of drivers,status quot
But the big one is when you go to best buy what do you see?
If there was mainstream (hp,compaq,dell,emachines) choice
That was 75 bucks cheaper then the rest
Dont you think things would be different
just my 2 cents
ps suse 9.3 with the multimedia upgrades from them for mp3/movies
is very nice, but each of the distros I have run was good this is just
better; and that is the other story linux is moving pretty fast
9 • Let all the flowers bloom, BUT... (by pp on 2005-05-02 12:01:41 GMT from United Kingdom)
Competition is good. I welcome all the new distros. But at the same time, my choice will always be 90% determined by the answer to the following questions:
1. Is this distro on a secure FINANCIAL BASIS now and in the next 3 years? Does it have a REVENUE STREAM or not?
2. Is there a large USER BASE to turn to?
The rest 10% is about features.
As someone here wrote: "stick to the top 25". Good advise, but I'd be even more critical if you don't want to be reinstalling every year. Stick to Fedora, Mandriva or SUSE. Ubuntu maybe ok for now, but it's a house built on swamp. Where are the revenues? There aren't any. Debian? Fast as a Soviet planning comission.
The wild competition satisfies techies and produces titans once in a while, but if you are looking for an operating system, you can't lose by being conservative.
10 • A Linux distribution is like a religion ... (by Todd on 2005-05-02 12:13:35 GMT from United States)
A Linux distribution is like a religion. If you've ever tried to suggest to another person that his or her choice of a distro might not be the best, then you know what I mean. Even if you haven't, you have probably come across a "distribution opinion war" on one of the mailing lists or public forums. But that's OK. We should be passionate about things we love, even if it's just a mass of programming code. What follows are facts and figures about Linux distributions. Personal opinions may vary, but facts are a lot more difficult to dispute...
Not to split hairs, but if you substitute "OS" for "Linux distribution" you have a much more viable statement. :)
11 • Re: Let all the flowers bloom, BUT... (by Benedict on 2005-05-02 12:18:52 GMT from Belgium)
Debian? Fast as a Soviet planning comission
A bit later:
but if you are looking for an operating system, you can't lose by being conservative.
You are contradicting yourself here. By producing an excellent stable distro Debian is being conservative as opposed to the quick release distro's. So by your own definition, Debian should be in your shortlist of distro's you mentioned
12 • Many Distributions (by Flavio on 2005-05-02 12:37:29 GMT from Brazil)
It is my opinion that one the greatest stuff by Linux is many distros... If any distro doesn't have what you want you can build one for you...
13 • stick to the top (by Leo on 2005-05-02 12:39:17 GMT from United States)
I guess this is how free software eveolves: in an evolution, bazaar like way ( see http://www.catb.org/~esr/writings/cathedral-bazaar/)
As a practical little piece of advice, for a regular user, always stick to the top distros. I like to refer people to DW's Top Ten. We need more of that.
If you wanna play around, learn and/or develop, there are hundreds of distros around, this is good. And if it wasn't for this caothic evolutionary process, the Mandrake folks wouldn't have forked RedHat to create a really useable distro, Knopper wouldn't have created Knoppix, an extremely succesfull project ...
Let it roll guys/gals !
14 • re: Leo (by Jan Stedehouder on 2005-05-02 12:48:08 GMT from Netherlands)
Well, Klaus Knopper did create something new, something that did not exist before: a live CD based on Debian. He followed the spirit of OSS development: create something better. By know, scores of others have done little more than imitating his achievement by creating their own live distribution.
In the bazaar, everything is controlled or followed by the maintainer, who relishes in getting feedback. The individual with an itch should either create something new, take over an orphaned project or support the maintainer in scratching. In a world where each wants to be a maintainer of his/her own project the bazaar model fails due to lack of proper peer review.
15 • to many distrobutions? (by superkitty at 2005-05-02 13:12:13 GMT from United States)
Absolutely too many distros out there.
In the top ten you have 3 Debian based distros and of course Debian. It makes me wonder if the people working on the children of Debian might have been able to help Debian improve a great deal. Each of them have good qualities, it would have been nice to see them stick with Debian, then MAYBE Sarge would already have been released
16 • Onebase Linux 2005 (by Anonymous Penguin on 2005-05-02 13:28:00 GMT from Italy)
When I read that the new Onebase Linux had been released I jumped of joy: I had read at their site the Onebase was beeing completely rewritten and I was hoping for a worthy contender among the source based distros.
But my joy was very short-lived.
This is what I read when I was trying to download it:
>>Inorder to download Onebase ISOs, you need to donate. Read below on why you have to do this.
» By donating $14.00, you can download for 7 days any Onebase Product ISO that has been released.
» There is no hidden or recurring costs after downloading and installing Onebase Linux. <<
Whaaat? Is that a joke? If you want to sell your distribution, don't talk of "donations" !
And the 7 days clause is particularly absurd.
But then it goes on:
» Despite this if you are not willing to pay this donation amount, Please PM the forum administrator to apply for a free copy² <<
With other words either pay or beg!
17 • Lot of distros (by Grmblmbl on 2005-05-02 13:30:33 GMT from France)
There are many distros but I think it can be usefull. We can consider that there are a dozen of stable and mature distros, and hundreds of experimental ones. I agree developpers should try to join an existing distro that nearly does what they want instead of developping their own. But diversity is good and sometimes something new can appear :)
I agree with Steve about scoring distros in order to see the strongest.
I think it could be usefull to have a tool to compare 2 (or more) distros : version of packages, size, price...
I also need a form to help me choose the distro I need : I check the packages I want the most recent version of, I choose live or not, I chose cd or net install...
And I need a better paid job but I'm not sure you can help :)
18 • No subject (by Marauder1 on 2005-05-02 13:34:47 GMT from Canada)
GPL is about give and take ....
Someone gives a distro and the other takes it.
The other takes the distro and improve it.
Someone gives a distro and the other takes it.
The other takes the distro and improve it.
Someone gives a distro and the other takes it.
The other takes the distro and improve it.
But the way, if i like the Mona Lisa with a moustache ...
19 • re: re: Leo (by Leo on 2005-05-02 13:38:28 GMT from United States)
I'll have to disagree when you say "The individual with an itch should either create something new, take over an orphaned project or support the maintainer in scratching." Most of these people are doing this on spare time (I know 'cause I did it myself for a long while). These people should do what they feel like doing.
If you are talking about commercial distros though, yeah, they should have a clear focus and clear targeted market. But this is not what we are talking about here.
In the end, interesting projects/distros will end up attracting users and developers. Only a few. Only the ones that make more sense ...
20 • Too Many Distros (by GWJ Mateo on 2005-05-02 13:39:24 GMT from United States)
I don't think there are too many distros per se, but too many distros trying to do the same thing.
If there is no significant difference in package selection, the only thing that separates Mandriva, Fedora, Suste, Slackware, Mepis, Gentoo, Ubuntu, and the rest of the top 25 is the package management set and whether you have to pay for the pleasure.
To my mind, there should be, max, three or four of any particular subset of linux usage: 4 all rounders, 4 pure desktops, 4 minis, 4 pure servers. Anyone that wants a particular feature beyond that should devote their effort to making one of the leading contenders in that area better, rather than further fragmenting Linux development.
Of course, I'm waiting for someone to come up with the uber-gaming Linux distro so that I can dump windows for good, so maybe I'm part of the problem.
21 • Are there too many Linux distributions? (by Bob on 2005-05-02 13:46:35 GMT from United States)
My god, yes! My question is: What is unique about each distro? I would love to see a list of features from each distro that is NOT in any other distro. Sometimes it just seems like a very subtle difference. I personally would prefer a "firefox" type distro which had "extensions" and "skins", etc., that would make a distro easily customizable, without having to resort to making my own.
It just seems like there are a lot of workshops making wheels out there, with no one looking at each others.
22 • Too many distros-NO (by Dave Thacker at 2005-05-02 14:24:34 GMT from United States)
If I was to volunteer for Debian today, it would probably take me a year or two to be allowed to do anything useful. Same goes for Gentoo. They have standards to maintain and reputations to protect. I understand that.
One of the freedoms of GNU/Linux is the freedom to learn by doing, That means I don't need the permission of Debian, Gentoo, Ubuntu, Slackware, or Red Hat to work with the OS/software and learn from the experience. If I choose to put up a web site and distribute my work, then great. If not, I have still learned something. I will have learned far more than I would have spending time playing "Distro Politics" and trying to get approved to do something cool.
Saying there are too many distributions is like saying there are too many kinds of cuisine! Can you imagine how pathetic life would be if we could only eat McDonalds? Yechh!
23 • Distros / Ranking / Which one to choose (by Vic at 2005-05-02 14:29:27 GMT from Luxembourg)
Refer people to the DistroWatch Top Ten is not a good thing. But that's exactly the problem, you will have to find the distro that suites you best. I like Kanotix, Arch and SuSE, I don't like Mandrake, Ubuntu and others. In other terms, it is difficult to find the right one and one cannot just stick to the Top Ten since one does not know about other distros. OK, for a newbie SuSE and Mandrake will be fine (and Kanotix!), but personally I won't test SuSE (had enough of them) or Mandrake (screwed up my HDD due to a bug).
Nevertheless, there are too many distros, for as it is so hard to find the right one, yet the huge amount of available distros gives you possibilities you couldn't dream of!
Perhaps a new ranking system would be great, but I haven't been thinking of the details...
24 • Distrowatch Weekly and me... (by J.Gabriel on 2005-05-02 15:29:44 GMT from United States)
I just want the Lad and the DsWW staff to know that I look forward to Monday's Weekly edition more than Slashdot or any other news site period.. You are now my favorite site..
That said could I ask for search sorting based on default desktop. It would be great to take an older machine and find all the IceWm or Xfce distros out there to choose from.
Keep up the fantastic work..
25 • yup, too many distros (by Anonymous on 2005-05-02 15:33:12 GMT from India)
perhaps developers who create new distros may think of contributing to debian, and maybe we will see a shorter release cycle ;-)
Distrowatch link on MS site, thats a milestone!
26 • multimedia (by austen on 2005-05-02 16:12:31 GMT from Canada)
Maybe there are too many distros for most purposes, but it's better than too few. I appreciate any distro developer who commits to gnu/linux multimedia.
Take Demudi. Having a complete audio package with a tweaked kernel in a single installation is great! And apt makes it easy to update. But others may prefer redhat/fedora based systems... and some may just want to shell out for boxed products like "studio to go."
At least in multimedia, a distro should be designed for the application. I want a system without bloat. That's a huge advantage for linux.
27 • No subject (by fiksve on 2005-05-02 16:22:14 GMT from Norway)
Its not that simple. If you are a budding developer you are not likely to be invited to any interesting parts of existing projects. And if I am to spend my spare time developing something, it better not be about improving an installer or maintaining the AMOR packages!
This counts double if I have a revolutionary idea I want to implement in the core of the distribution. Even if they did like the idea it is unlikely that I would play a big part in incorporating it. Whats the point?
If I ever start on my own distro I will want decisionmaking power, something wich can only be achieved by forking a project and make it my own. The only real question here is wether something like this should be released to the public, and possibly wether the public should be less enthusiastic every time they hear about some thing great and new. If it it good, it is bound to end up in the mainstream distroes sooner or later anyway. Thats the beauty of GPL I suppose :)
28 • The number of distros isn't the real problem. (by DaveW on 2005-05-02 16:27:49 GMT from United States)
The problem is too many distros that are essentially the same. There are only a few archetype Linux distros out there from which almost all others are derived: RedHat, Debian, Slackware, LFS. Too many new distros differ from the rest only because they include Blackjack instead of Solitaire, or Gnome and not KDE. Most of those differences can be negated with a simple download of a few new programs.
As the number of distros grows, evolutionary sorting fails because there will never be reviews/comparisons done on most of the new ones, so no one will know when something truly different is going on. IOW, there aren't too many distros -- there's too little information.
I'm thinking maybe DistroWatch, as the premier Linux info source, might be able to help. Maybe add a Hotlist of distros that really have something special about them, new ideas worth exploring -- distros like OneBase and Arch and, earlier, Knoppix and Kanotix come immediately to mind. To qualify for the Hotlist, distro makers would have to produce a convincing fact-based argument as to why their work offers something truly different beyond just the mix of included packages.
That's more than Ladislav could be expected to do, but something that DW users could take on as a group. Just an idea that needs further development, but maybe worth some thinking about? It does seem like Linux has reached a stage of development where some more evaluation and pruning is needed.
29 • two categories of distro (by ybouan on 2005-05-02 16:41:52 GMT from United States)
I think they are two categories of distros: live and not live.
Live CDs are mostly run as-is so there will always be hundreds of them and they can still be usefull.
The only way to reduce these numbers is for a distro like morphix to become a standard with each and every itch scratcher publishing a "distro build script" instead of a full distro. This would be kind of like the Thin Station online build system. Now I don't know Morphix to well and have not looked at it in a while so this may already eist.
For other distros since they are instalable I see no reason to have so many. A lot of the new ones I see are internationalized versions of mainstream ones.
The devs making them should just contribute translations to other projects.
30 • Never too many choices! (by R on 2005-05-02 17:00:40 GMT from United States)
Someone once told me "get any two experts together, and watch a hockey game brake out." Open Source gives people choices. The freedom to try new things without someone telling you no. You start regulating Distro's, you destroy what Open Source is about. You destroy freedom.
From my experience getting people to try Linux. I've found that choices scares Window users. They always say "what app am I supposed to use?" That never scares them back to Windows, What sends then back is the lack of Drivers, and all this Win-Hardware junk. Take care of this hardware issue, People will switch.
31 • So what? (by Max on 2005-05-02 17:39:29 GMT from Australia)
The only reason we know of all these distros is because Ladislav had the idea of creating this wonderful website. Otherwise, most of us would not know about probably 90% of the distros featured in here. And then nobody would be bothered... And so what if there are so many? People can create whatever they like, that's the whole idea of free software no? And people will then decide if your creation is any good or just "another" distro. Who is to say, "no you can't create your distro"? You guys? Duh!
32 • way too many distros (by way too many distros on 2005-05-02 17:43:52 GMT from United States)
subject says it------more than anything im tired of the all everything distors----just give me a kde only desktop distro, or a gnome one for gnome people
33 • Ndiswrapper (by sclebo05 on 2005-05-02 17:47:40 GMT from United States)
Good call on donating to the ndiswrapper project. I am using it right now to connect my otherwise useless netgear wireless card.
34 • RE: too many distro's (by Richard at 2005-05-02 17:47:43 GMT from United States)
While there might be advantages to having fewer, more differentiated distro's I'm not bothered by the number. I download most of the general interest releases when they appear on Distrowatch. And burn them to a CD-RW. CD-RW's are cheap around my area...About 30 cents a piece. I usually have a spare system around that's is getting ready to be donated, so I test the installations on that. Live CD's I use on any of my personal systems. If it doesn't boot or locks up during install on a known working system that I can get into under Puppy then I just erase the CD-RW and move onto the next one.
I like the wide variety Linux and BSD provide. If I want to do some video or sound editing I can boot up Dyne:bolic. IF I just to surf around or test the stability of an old PC I boot up puppy, Berry, kanotix or Austrumi. When I want to try more in depth tasks I have Suse or more recently PCBSD. Next week I may have a different distro I prefer for my surfing/productivity work. And of course if the game makers ever settle on supporting a *nix that would kill the last of my widows systems.
35 • Amount of Distro's (by NixUser on 2005-05-02 19:12:36 GMT from United States)
"But there is a downside to it that really needs to be addressed. That is that there are now too many distributions and the ability to identify exactly what Linux is and to 'sell' it may become more difficult when it should be getting easier. I think developers need to ask themselves whether it might be more advantageous for the community as a whole for them to volunteer on existing, popular distributions rather than to create their own."
Most distro's are to scratch a developers own itch. I personally think that the number of distro's uncategorically identifies Linux. It is identified by:
2) Freedom (beer,speech)
Thought the above traits, you have everything from:
3) Multi-media centers
4) Technical workstations.....and.....
As for the number of *BSD releases; most are firewalls, if I am not mistaken. Your going to have the major 4 or 5 releases:
1) FreeBSD: speed / flexibility
2) NetBSD: portablity
3) OpenBSD and its clone (Mir): security
4) DragonFlyBSD: Extension of the 4.x kernel, taking new ground.
5-11) Firewalls: Uhhhh firewall. ;-)
More specifically, these 5 catagories of *BSD; each unique to itself. The *BSD landscape isn't all that vast. Well, I guess, I forgot the mini-BSD releases. O'Well, I guess, I might have my shot in the foot. Anyway, I see the over all scene for *BSD quite compact. Perhaps it's a 50/50 split.
36 • No subject (by Anonymous on 2005-05-02 19:15:53 GMT from United States)
"Thought the above traits, you have everything from:"
"Through out the above traits, you have everything from:"
"O'Well, I guess, I might have my shot in the foot."
"O'Well, I guess, I might have shot myself in the foot."
37 • ONEBase (by Morgan Schulz on 2005-05-02 19:59:05 GMT from United States)
Has anyone tried onebase yet? It seems to take on a lot of the issues that I have with the current state of linux. Beyond that, $14 to download a fully functional distro doesn't seem all that unreasonable, although they should probably offer to send you the cd for that cost. It's certainly better than making you join the "club" so you can get updates for your system.
Easy unatended installs, easy disk imaging and deployment, and restore points are places where windows has us beat. I know these *can* be done in linux, but not simply. These tasks are important in IT, so lets make them fast and easy.
Just my $.02
38 • RE: ONEBase (by Morgan Schulz ) (by Anonymous Penguin on 2005-05-02 20:26:48 GMT from Italy)
"Has anyone tried onebase yet?"
I wanted to, but I didn't. I like honesty (don't call a "donation" what in fact is a price) and I like to spend money (either by buying or by donating) on open source projects which are useful to me. How do I know before trying?
Anyway if you take a look at their forum, people seem to be having quite a few problems, including a few users who have a lot of experience, not just with Onebase but with Linux in general (I know them)
So if you are willing to spend (and possibly waste) $14 , go ahead by any means.
If you want to play safe, try free distros: Kanotix highly recommended.
39 • Mandriva 10.2 (by Leo on 2005-05-02 20:43:25 GMT from United States)
I have updated Mndrake 10.1 -> Mandriva 10.1 using urpmi the other day. My impressions are almost the same as expressed in the latest review of Mandriva 10.2 posted here in DW.
* Great stability. I upgraded over 800 packages all at once, with no intervention of my part, it all just worked. And it's been running fine ever since. Mandriva is a pleasure to use. Things just work for the most time.
* Mandrake/iva has evidently decided to be one step behind the bleeing edge, to provide more stability. Good ? Bad ? Whatever. It's their strategic decision. I think it is mostly good. Although they should (IMHO) keep the "community" releases in the bleeding edge every six months, and a corporate, very solid release once a year. They lost the bleeding esdge, and they may loose some of their mindshare to (K)Ubuntu, we'll see.
* The artwork is very little profession in 10.2 - There is penguin which has been characterized as drunken, dead nad stoned. To me it is the latter. THis is the main background. Horrible choice. And the word MENU in a rough font on top of the Main Menu button in the panel is really really bad - These things have been discussed a lot in cooker, but they still went through.
Other than that, Mandriva is perhaps the best distro around
40 • too many distributions? No! (by Walt Huntsman on 2005-05-02 20:59:58 GMT from United States)
I tried posting this before, but for some reason it did not take. So here goes take two.
I agree with J Gabriel's request for a search by window manager. As a user of an older laptop, I try to stay away from KDE and Gnome, to the point that I don't even load KDE/Gnome apps or libraries. They just seem sluggish on my 400Mhz Celeron (even though I have 256mb RAM).
As far as too many distributions, I don't think so, especially as I haven't quite found my perfect distro. Because I have such a slow dialup connection, I concentrate on smaller distros (~125mb range max). However, in that small space, I want to be productive (word processor like AbiWord, email program, web browser, file manager, XMMS, GIMP or other image manipulation program, GUI text editor like SciTE, working dialup, etc.) I tried a number of such distros, but I'm still looking around.
I've tried DSL (wouldn't boot when I first tried it), Austrumi (no working dialup), Puppy (same problem), BeatrIX (slow), SLAX, no working printing), and currently Feather, which works for most things.
One other thing I'd like is to find applications that look good, work good, yet don't require KDE or Gnome libraries, or Python libraries, or - you get the idea. (One of the few such apps that comes to mind is JPilot.) Can't apps be made that look good and work good without requiring a bunch of additional libraries be installed? How about being able to include just the piece you need to make it work?
I don't mind downloading once in a while, but with a dog-slow connection, I don't like to make a habit of it. Downloading SLAX took 18 hours over several days. Any thoughts or ideas? Upgrading my machine or connection are now options at present. Thanks.
walt_huntsman [at] myrealbox [dot] com
41 • Vitality rating needed- amount of distros is good- could use more! (by AQ on 2005-05-02 21:02:34 GMT from United States)
I of course believe that choice is never unwelcome.
But I do think that their should be a better categorization of the vitality of distributions.
Without choice, Scientific Linux would not come into play, or alot of others that are specific to certain fields of interest, such as dyne:bolic. These might not be the most popular, but that doesn't mean that they are necessarily suffering from a lack of vitality.
I just simply don't care about popular culture where the majority dictates the products which are available to the minority. And that is what gnu/linux gives us, the opportunity for all to have what they want.
However, when looking at GNU/Linux in terms of that freedom, something that can become a much greater factor of the strength of the distribution would be some ranking of its vitality level.
42 • Too Many Distros (by David Matson on 2005-05-02 21:54:25 GMT from United States)
I find that many distrobutions are just a showcase of maybe two or three new features/applications. It's rare that a distrobution is much more than that. They (their homepages) always make it sound like the new distrobution is much more of a new mindset or something, but I usually come out saying something like, "I liked the way they did their menus", or "I liked their package management", or "I liked how they added a GUI tool to simplify that difficult application"... As a side note, I feel that the separate, non-compatable, and non-intuitive package-management schemes are a huge reason why many *nix newbies/investigators loose interest shortly after installing.
43 • Number of Distros Again! (by Steve on 2005-05-02 22:29:15 GMT from United Kingdom)
OK, so, perhaps another idea.
Who has played Epic's Unreal Tournament? (??)
Those who have.. know that, under the game being played, is a great game engine which can made to do many many things just using a high level script (usually supplied by 3rd parties). In fact the game can be _as a user experience_ utterly changed in any way - perhaps so changed as to be quite unrecognisable.
These are "mutators". There are literally hundreds - the game was designed to accept these and was a great strength, as enthusiasts could spend time and energy building their own special contribution. Like plug-ins except 1 level more capable; they can change code.
Idea - why not have a Gnu/Linux "core distro" and to personalise from to a final distro - using mutators. Build on a good core system to be - whatever you want.
So there would be a minimal handful of distros; all others grow from these via using mutators.
For example: Say a solid LiveCD was the base which could then be installed by a mutator to HD and in so doing, it gains the aspect the user has choosen e.g. web server.
The mutator could be no more then a script to grab the right elements from current repositories (e.g. latest samba, KDE etc) and to merge these onto the LiveCD image writtent to disk.
Want to update the core system? Fine, put in a new LiveCD and run the same mutator. Now you have a new full suite fully up to date and the mutator has carefully kept all your files in your home dir etc etc.
I'm not a coder. Is this possible? If Epic's UT can do this, why can't Linux?
Bye for now...
44 • Re: Onebase (by andrew on 2005-05-02 22:54:38 GMT from Australia)
I agree that business of "compulsory donation" is a bit unfortunate, but on the other hand, it is just 14 dollars - for Heaven's sake let's not be so cheap! So I got over it and paid up, because I find Onebase interesting, if sometimes frustrating. But it is one distro that is trying to do something else than just grab Anaconda from here, Arch from somewhere else....
45 • mutators (by gnobian_ken00bie on 2005-05-02 22:57:05 GMT from United States)
Actually, the mutator idea is very close to Morphix. There are quite a few tools available to customize it fairly easily and several other distros are based off of it. Ubfortunately, Morphix creator Alextreme has been busy with a lot of other matters - like his degree - and his web hosting company flaked on him, so the development of the main Morphix versions stagnated for quite awhile. Seeing that the official Morphix CDs are so outdated has probably undermined a lot of the interest. Still, it's a great project.
46 • advertising mepis (by im-ka on 2005-05-02 23:12:32 GMT from Sweden)
this is how coolblue (the webmaster of mepislovers.com is cblue.... hmm) advertises mepis on the ubuntu forums
47 • RE: andrew (by Anonymous Penguin on 2005-05-02 23:20:16 GMT from Italy)
"I agree that business of "compulsory donation" is a bit unfortunate, but on the other hand, it is just 14 dollars - for Heaven's sake let's not be so cheap!"
You don't know me because here and at OSNews I use a nickname which is different than the one I normally use at distro forums.
Yes I agree, $14 is not a lot, but for me it is a matter of principle: I have just donated (a *real* donation, not compulsory) for the second time in 20 days to my favorite distro: Kanotix.
But I am never again going to make the same mistake I made with SkyOS: being a paying beta tester!
And you know what I mean, because you are one of the "experienced users" I mentioned above who are having problems right now.
48 • Re: advertising mepis (by steve on 2005-05-02 23:29:45 GMT from United States)
It is quite sad that Cblue needs to promote Mepis in that way. I know the Mepis lovers are very bitter about Ubuntu and they are always bringing up about how Mepis is superior through other forums, slashdot, distrowatch, etc.
They even try to manipulate the distrowatch ranks. Just have a look at the their forums as the topic keeps coming up a lot.
I tried Mepis and Ubuntu; and still didn't like it.
I have yet to find the perfect Linux distro, hopefully that ends soon.
I looked at OneBase and it seems rather cool, but I won't jump too soon.
And PC-BSD looks really promising, especially with their package installer.
49 • RE: advertising mepis (by Anonymous Penguin on 2005-05-02 23:31:25 GMT from Italy)
I find it funny, but it is kind of giving Ubuntu fans a taste of their own medicine: they are *absolutely* everywhere advertising Ubuntu as the Almighty God of all distros and operating systems.
50 • Re:andrew (by andrew on 2005-05-02 23:52:50 GMT from Australia)
"You don't know me because here and at OSNews I use a nickname which is different than the one I normally use at distro forums."
I remember you from Onebase forums - your country of origin gives you away:) Hi!
"Yes I agree, $14 is not a lot, but for me it is a matter of principle: I have just donated (a *real* donation, not compulsory) for the second time in 20 days to my favorite distro: Kanotix."
That's fair enough. I also donated voluntarily to distros I really like. I just decided that while Onebase kind of twisted my arm, in the end I was willing to donate to them too, because I find it interesting and different enough that I would like it to continue, and the developer clearly put a huge amount of work into this release, and the price is reasonable ... so what the heck.
"But I am never again going to make the same mistake I made with SkyOS: being a paying beta tester!"
Well, as it happens I paid for SkyOS as well because I want to see diversity - I believe it is good for me as the consumer. I'm willing to fund development of SkyOS to the tune of what was it - 30 bucks? since I get the finished product as well as acess to all betas. But that's another matter.
"you are one of the "experienced users" I mentioned above who are having problems right now."
But to put it in perpective, they are not show-stoppers. The system is up and running... I'm using it right now.
51 • advertising Mepis (by gnobian_ken00bie on 2005-05-02 23:57:11 GMT from United States)
Where have you seen Ubuntu users bashing another distro on a forum for that other distro? Perhaps it has happened. I don't know since I don't view every forum out there. If they have behaved this way though it's pretty tasteless.
Anyway, as far as I know, I'm the most active Ubuntu user in THIS forum and I have praised some of the strengths of Mepis while pointing out some of my objections. And I've always acknowledged that different distros will suit different tastes, values, temperaments, and needs.
And that's the mentality I've encountered from other Ubuntu users. Some evidence... here's how Ubuntu treats Mepis:
52 • Mepis... ROFL (by AQ on 2005-05-03 00:00:55 GMT from United States)
Now that was a hilarious link.
You'd have to be out of your mind to float something like Mepis to Ubuntu users. Mepis has quite a few parts which are proprietary... and Ubuntu is all about complete openess and community.
I use neither and agree with Ubuntu's method, but that whole scenario is hilarious. I have to wonder how long it will take the Mepis crew to come here and hi-jack this thread.
53 • for anonymous penguin in Italy (by gnobian_ken00bie on 2005-05-03 00:24:11 GMT from United States)
Your point a week or two ago concerning compatibility between Debian and Ubuntu is a legitimate one. I wish there were a better way to do what they're doing, but I can't think of one. That said, I've run vanilla Debian packages on Ubuntu, but YMMV.
However, I wanted to inform you and others of an excellent option - one that reveals why Debian is the epitome of power an flexibility. From within a 64 bit Ubuntu install (or Sage or Said for that matter) it's a fairly simple matter to construct a 32 bit chroot jail with a very minimal install using debootstrap. Then using dchoot, you can execute programs from the jail without having to chroot each time.
Now, here's where I'm addressing your point. You can also make a 64 bit Sid jail under your 64 bit Hoary. And a 32 bit Sid jail. And 64 and 32 bit Breezy jails. And Sarge jails. And then you can run apps from any and all of those without having to reboot or even chroot.
debootstrap and dchroot rule! And I'm glad I put off learning Slackware and Gentoo and the BSDs until I'd learned more about Debian's tools. I do still want to learn those as well. And another poster has convined me to try Scientific Linux too. But I continue to be amazed at the power of Debian.
54 • the power of Debian continued (by gnobian_ken00bie on 2005-05-03 00:30:17 GMT from United States)
Next I'm going to experiment with dkpg-divert. That looks like an equally useful tool for running a mixed system without breaking things.
55 • Micro$oft FUD and BS (by diertowatch reader on 2005-05-03 00:35:59 GMT from United States)
I agree with their statement "I advise you to check up on this topic yourself and draw your own conclusions."
I have that's why I use Gnu/Linux.
DistroWatch on Microsoft.com LOL.
56 • RE: advertising Mepis (by Anonymous Penguin on 2005-05-03 00:58:05 GMT from Italy)
I didn't use the word "bashing", although that happens, alas. I just used the word "advertising"
57 • "foresight linux" / "historical distros" / hurd & haiku etc. (by Anonymous on 2005-05-03 01:03:03 GMT from Italy)
FOR LADISLAV. i have a question and two suggestions:
1. you mention "Foresight Linux 0.8" on the current weekly issue, but it is not mentioned in "how independent your distro is" section (it should have been mentioned there as being based on "Specifix Linux"). why? is it very newly added?
2. i think maybe it's time for distrowatch to add a new section, "historic distros" (or something like this) where you could mention all distros which were discontinued before distrowatch appeared. also, you could add all distros in the "discontinued" section three (or five?) years after they released their last version. what do you think? (i guess you might use http://www.ibiblio.org/pub/historic-linux/README or other links which you may find) i think, sooner or later, distrowatch will have to include a section like this. "discontinued distros" list is more like a waiting list for a "historic distros" section.
3. since distrowatch includes bsd and linux, why wouldn't it include all types of operating systems which are "software libre" (hurd-based or even haiku, for instance)?
58 • RE: andrew (by Anonymous Penguin on 2005-05-03 01:04:43 GMT from Italy)
Hello again :)
Indeed, you and I have very much in common: promoters of open source; promoters of operating systems choice.
59 • RE: for anonymous penguin in Italy (by Anonymous Penguin on 2005-05-03 01:14:28 GMT from Italy)
Thanks for the info. Indeed, you are right, the power of Debian is amazing.
60 • Too many "distros"?? Not really!! (by Ed Borasky on 2005-05-03 01:29:05 GMT from United States)
I don't think there are too many *unique* distros!! There are three commercial distros: Red Hat/Fedora, Mandriva, and Novell SUSE. And there are three "community" distros: Debian, Slackware and Gentoo. Everything else is a variant of one of these as far as I'm concerned. And most of the variants are variants of Debian.
61 • RE: Foresight Linux (by ladislav on 2005-05-03 01:57:17 GMT from Taiwan)
1. The "stats" page you are referring to is in the process of being discontinued and is no longer updated. This was a manually-maintained page, which was too much work, but now that we have everything in a database, you can easily find various aspects of distributions from the search page:
2. As for historic distributions, we do have a list:
3. This has been asked and answered many times. There are only so many hours in a day, so I had to draw the line somewhere. Right now the line is at Linux and BSDs (Solaris was an exception because it was in much demand, much greater than Hurd or any other OSes). Besides, there are 112 Linux distributions on the waiting list right now, so it's not like I have nothing to do...
62 • RE: Ed Borasky -- 'Too many "distros"?? Not really!!' (by Anonymous on 2005-05-03 01:58:42 GMT from Italy)
there are MUCH more than six distros. you are not fair. you are also considering the popularity factor, while pretending you are considering the "uniqueness" factor only. it is not honest.
63 • I stand corrected (by gnobian_ken00bie on 2005-05-03 02:11:22 GMT from United States)
You did say "advertising", yes. Though I also haven't come across Ubuntu being advertised in the forums of other distros. (But again, I may just have missed that.) And there's a fine line between preaching the immense superiority of one's own distro and bashing another, e.g. saying distro X is "far far better" than distro Y or as you put it or likening distro Y to "Almighty God", particularly when one is going into the camp of the distro at the receiving end of the unfavorable comparison.
I'd prefer just to say that no distro I have yet encountered has been a readily adapted to my own particular needs and abilities and values than Ubuntu. But there are a lot of great distros out there and you may find another to be better suited for you. There's no bashing there.
(On reflection, it occurs to me that in calling Debian the epitome of power and flexibility, I may have been guilty of the very thing I am now criticizing. I like the power and flexibility, but it may well be that another distro, perhaps Gentoo, has as much or more. Tempering one's remarks is a good thing.)
64 • "historical distros" vs. "discontinued" (by Anonymous on 2005-05-03 02:24:32 GMT from Italy)
1. i hope we will soon have all of the fine statistics in the new (automated) form, won't we? i am talking about automatically generated synoptic tables, which is much nicer than the google-like search&find.
2. please read my second point again. you didn't seem to have understood it at all (maybe you read it in a big rush). i am mentioning both sections there.
or at least you could complete the "discontinued" section. please follow the link i gave you. you will find old distros that distrowatch doesn't include. some of them are still available for download and may actually be useful for old hardware. maybe distrowatch could also link to them. (an example would be "jurix", on which suse was initially based.)
3. i got your point on "so many hours in a day". but is it really impossible for you to get a team of volunteers? are they hard to find? nobody is offering to help? i very much admire your work, and i don't think "it's like you have nothing to do". but i never implied you should work alone. get a team, don't let yourself become a workaholic :) (btw, congratulations for robert storey)
65 • "package management" criteria (by Anonymous on 2005-05-03 02:35:27 GMT from Italy)
(for Ladislav) why don't you include "package management" in the "Search by Distribution Criteria"?
66 • klaus knopper (by vampire_janus on 2005-05-03 02:38:38 GMT from Philippines)
klaus knopper didn'[t invent live cds. he was in fact working on another live cd before working on knoppix. so knoppix is not the first live cd but arguably, it is the most innovative
67 • site bug? (by vampire_janus on 2005-05-03 02:43:30 GMT from Philippines)
why are the textfields and buttons on top of this page (Type distribution Name, Go, select distribution Go) differently themed than the home page?
68 • RE: "historical distros" vs. "discontinued" (by ladislav on 2005-05-03 02:48:44 GMT from Taiwan)
2. I re-read your point, but I still don't see where I made a mistake. The page I gave you lists distributions that have been discontinued since DistroWatch started, as well as historical distributions that were already discontinued by the time DistroWatch started. I am not prepared to go any further than that, because there is very little interest in those distributions. If I get dozens of email asking that I create a more detailed page about distributions that existed before the year 2000, I might change my mind.
3. As Klaus Knopper once said, "it is sometimes more difficult to manage people than to manage software". Some time ago I made an effort to solicit help on these pages, but my experience with this was overly negative - some people offered to help, some of them even started working on some stuff -- and then I never heard from them again. This has been the case with 90% of the volunteers. Maybe it was my fault - maybe I am not a good manager or can't delegate tasks clearly - I don't know. But the fact is that I am no longer enthusiastic about asking for help.
69 • "package management" criteria (by Anonymous on 2005-05-03 02:59:28 GMT from Italy)
(for Ladislav) why don't you include "package management" in the "Search by Distribution Criteria"?
70 • Add a new category -- /tmp Distros (by William Poetra Yoga Hadisoesen on 2005-05-03 03:39:45 GMT from China)
New distros which don't look like they are going to stay around for a long time should be in that category, and if they manage to stay alive for 2 years or so, move them to the main category. How about that? :)
71 • There's still room for more!!! (by Toscanini on 2005-05-03 03:42:06 GMT from United States)
My old laptop doesn't have a hard drive, but it boots Damn Small Linux into memory, plays music from my fileserver, and never needs a reboot unless I want to try the latest version of DSL. Can't do that with Windows.
My clients bring me their infected computers. I boot them with Knoppix, connect to the network, scan the drives with antivirus software and recover files. Can't do that with Windows.
Many of my servers have been running the same version of Debian GNU/Linux for nearly three years with free security updates, a nearly unlimited selection of software, and absolutely no licensing fees whatsoever. Can't do that with Windows.
My Slackware workstations and servers provide a handpicked selection of best-of-breed cutting edge software, an unbeatable development environment out of the box, and a base configuration so secure that none of them were compromised even when the sole maintainer was unable to release updates for months due to health problems. Can't do that with Windows.
Gee whiz, my car stereo even runs Linux. Too many distributions? This is a sign of success, people, not failure! Along with other free/open kernels, Linux has proven to be a solid and flexible base for a staggering number of applications. Yes, one day you may even owe your life to the fact that someone was able cluster together a bunch of inexpensive PCs using some obscure Linux distribution in order to analyze the human genome or simulate the weather or model new drugs. The only reason there are so many Linux distributions is because there CAN BE. Can't do that with Windows!!!
Linux was created to be useful tool. And it is. It has already fulfilled (and probably surpassed) the ambitions of its creator and contributors. I'm always amazed when people suggest that "fractionilization" will hurt something that has already diversified, evolved, and succeeded beyond anybody's wildest dreams. It's like saying that ice cream will lose popularity because it comes in too many flavors, or that pizza will go out of style because it has too many toppings. Well let them eat vanilla and pepperoni, I want mango and roasted red peppers!
72 • No subject (by Anonymous on 2005-05-03 04:05:01 GMT from Italy)
thank you for your answer and i am sorry about this situation (lack of volunteers).
i took a look at the search page and i admit it's much better than the manually maintained pages. i would have the following suggestions:
1. please include search by "package management"
2. use synoptic tables as much as possible. recoding the manual tables to automatically generated ones would be very nice. SOME criteria fit the table-like design much better than the search-engine style. for instance, it's much more pleasant to see all of them at the same time and to make comparisons (like in the manually maintained "how independent" and "package management" sections). actually, some things become much more obvious when listed in synoptical tables.
i refer to those tables pretty often. they are one of my main targets while visiting distrowatch, so please don't distroy my pleasure: keep the (recoded) old design too, along with the new one! automatization doesn't necessarily mean changing design.
3. maybe you could add the option to search for distros NOT matching one of the criteria. for instance, people may want to find all "desktop" distros which are not BSD or which are not live cd.
4. i think it's vital to change the "distribution category" selection type from menu-style to (multiple) box-checking. that way people could select distros symultaneously belonging to two or more categories. most of the categories are not mutually excluding or "disjunctive". for instance, people may want to find distros which are both "multimedia" and "live cd".
this may also be the case with "desktop environment" -- what if somebody wants to find distros which support both kde and icewm? (maybe even the "based on" criterium: there are distros which are based both on knoppix and debian)
i think a good model would be rpmseek.com which has a(n advanced) search menu based on box checking (see http://search.rpmseek.com/search.html?hl=com). this way it allows for just ANY kind of search combination.
since everything is automatical anyway, i don't see why not making it as flexible as possible and why not adding these search features. this would be very nice, since so many people are very much interested in distrowatch for its statistics and management of categories.
the fact that they don't compile a list of wishes like mine doesn't mean they won't greatly appreciate these features. some people simply accept what they get and implicitely visit less frequently. i chose to suggest improvement.
P.S. i wonder why it shows i sent two of my previous messages twice? i only sent each of them once
73 • the previous message is for LADISLAV (by Anonymous on 2005-05-03 04:07:19 GMT from Italy)
the previous message includes some suggestions for LADISLAV
74 • mepis vs ubuntu (by im-ka on 2005-05-03 06:59:30 GMT from Sweden)
i've posted the same on the mepislovers forum and i really like the reaction there. it looks like someone has used cblue's name, so apologies if my presuming was wrong.
75 • and... (by im-ka on 2005-05-03 07:15:39 GMT from Sweden)
76 • too many distros? I don't think so. (by catalinux on 2005-05-03 07:27:11 GMT from Romania)
So, was a post with a link about "The cathedral and bazaar."
The model IS the true model of Open-source development.
No problem about number of distros.We cant and must not influencing
this thing .A bazaar must be a bazaar.Cause never neeed another like -Microsoft "Cathedral".
my 2 cent words
77 • Too many distros...etc (by Ninad on 2005-05-03 07:27:24 GMT from India)
First this is a great site. I visit it almost every day and now I long for this newsletter. Great stuff Ladislav.
Gnu linux is about choice and fredom and stop any one from creating his/her own distro.Without this freedom to create/innovate there would not be any choice and it would be like Windoze.
I would not have found out about mulinux or Slackware when I got hooked to Linux in 2000. I needed some distro which would give me a taste of Linux and also would run on windows on a P100 16 mbRAM and 4.3g hdd
Both these distros support UMSDOS.
Now I do not use mulinux as I have upgraded my PC.
I learnt from this initial stint how to configure X, sound etc
A number of 400+ active distros is high though.But we would not have choices without these.
A rating based on certain parameters would be good
I sometmes wonder how many people are using the 400th distro.
Also where can I find a list of all the active distros on the DW list
Let the distros keep coming and let this site run infinitely
78 • Re: PearPC (by Anonymous on 2005-05-03 08:51:11 GMT from Germany)
I think you better support the gpl-violations.org project which has succeeded enforcing the GPL for several times now already.
79 • Agreed - waaaaaaay too many distros (by Foobarbaz on 2005-05-03 09:23:42 GMT from New Zealand)
I have to agree - there are way too many distros. I think it would be a good thing to suggest to would-be new distro-makers that they join an existing distro. ( Not that they'll all listen, but anyway .... ).
80 • Number of distributions, many irrelevant distributions (by Geert on 2005-05-03 10:24:28 GMT from United States)
In fact there are probably only 6 or seven important distributions with commercially relevant market share and proffessional support. All the others are just the market in action. What is strange is the attention paid to them in the media.
It seems also not "right"to consider the recent e.g. debians as seperate distributions: they are rather a repackaging. The installed thing is still a debian.
Like political parties: there are everywhere lots of parties, but in most countries the top 4 occupy 80 % or more of the field.
81 • too many distros? two sides of the story (by im-ka on 2005-05-03 11:45:34 GMT from Sweden)
on one hand, it would be nice if all the brainpower spread for creating and maintaining 300+ distros would focus on let's say 10-20.
on the other hand, every single distro makes some users happy and having so many distros is the effect of floss/gpl. and i think we all agree that it's the free software principle that has brought gnu/linux so far.
82 • Re: Too many "distros"?? Not really!! (by Ed Borasky on 2005-05-03 14:39:51 GMT from United States)
OK ... let me define "distro": It must include at least the following:
1. A Linux kernel (sorry, BSD folks -- if I knew more about your efforts, I'd include you, but I don't),
2. The GNU tool set of software (compilers, shells, etc.),
3. A corporate/for-profit or non-profit parent development organization,
4. A package management system and a package repository, and
5. A mechanism for keeping security/defect fixes current on installed systems.
Let me start with Debian. They have a non-profit parent development organization, a Linux kernel, the GNU tool set, a package management system and a security/defect fix update mechanism. Every Debian-based distro out there, regardless of whether they have implemented their own organization, toolset, security mechanism, etc. still depends on Debian! So, looking at the Top 10, I've flagged everything that depends on Debian:
1 Ubuntu: Debian
5 MEPIS: Debian
6 KNOPPIX: Debian
7 Debian: Debian
8 Kubuntu: Debian
9 Damn Small: Debian
Wow! Six out of the Top Ten distros are really Debian! They depend on Debian, if you install them to your hard drive, you use Debian's package management, usually Debian's kernels and GNU toolset, and in some cases Debian's security update service.
So what's left?
2 Mandriva: Mandriva
3 SUSE: Novell
4 Fedora: Red Hat
10 Gentoo: Gentoo
I had to go to the Top Twelve to pick up Slackware!
11 SLAX: Slackware
12 Slackware: Slackware
And, as I scan down the list, I keep seeing Debian-based distros. Maybe there are really only three distros, classified by their package format: Red Hat/Mandriva/SUSE et. al. (RPM), Debian (deb) and Gentoo (source/ebuild) :)
83 • too many distros? (by fab on 2005-05-03 19:26:15 GMT from Germany)
I am not sure if there are really too many different distros out there. One thing is for sure: the more distros we can use, the more innovation we might experience. Almost every distro out there has its special advantages and/or disadvantages and also its very special "niche" where it fits. Even if it is "only" learning something about linux-distro-building. But there is a downside.
The more people that decide to run their very own distros might learn certain stuff the hard way, thus loosing some time on problems that other distros solved earlier.
If the new developers would join one of the smaller distros, their very special ideas/innovations will almost surely be appreciated by the "not so big" developer staff. And many smaller distros would love to get additional help, because the number of developers constantly rises and falls. Some developers want to work onanother distro, some might get a new full-time job with one of the big distros like novell, redhat or mandriva and others still might want to start their own distro. This leaves many smaller distros in a constant situation of shouting desperately for developers in order to release a new version. And if no one reacts to this shout, some distros will die sooner or later. What a shame...
A little bit of merging rather "similar" distros would be a good thing imho. Maybe we should have "only" 200 distros around... Still enough for testing, isn't it? ;)
p.s: A propos: shout for developers... is there any site where potential developers can see which linux-distros need help in which area? Something like a central forum for linux-jobs would be a cool idea, right?
84 • im ka, ubuntu, mepis. (by gnobian_ken00bie on 2005-05-03 21:12:14 GMT from United States)
Thank you for clearing this up. it's refreshing to see this sort of conflict resolved in a civilized manner.
85 • why not? (by im-ka on 2005-05-03 21:18:36 GMT from Sweden)
"p.s: A propos: shout for developers... is there any site where potential developers can see which linux-distros need help in which area? Something like a central forum for linux-jobs would be a cool idea, right?"
well, let's do it. i think it would be useful. building appealing websites with cms is not a problem these days, i'm sure that someone or some organization would donate some server space. all we need to do is contact developers and tell them about it.
86 • No subject (by speel on 2005-05-03 22:33:49 GMT from United States)
"I think developers need to ask themselves whether it might be more advantageous for the community as a whole for them to volunteer on existing, popular distributions rather than to create their own."
rofl are you kidding me?? these wonderful distro makers want to make a name and with out it there annonmys :D :D
87 • Yes there are too many and No there are not enough distributions (by x on 2005-05-04 00:13:03 GMT from United States)
On the surface, it may appear as though there are too many distributions. I wonder how many of the advancements Linux, as a family of operating systems, would have advanced if there were only a dozen or so. It is inevitable that many distributions will sink into oblivion.
Some of these distributions were developed for specific situations and are great for those in a similiar senario. These may never have a large following. A few of these may just address hardware specific issues( embedded ).
System usage is another reason for the diversity in distributions. Users that want to use a system for audio and video will not need many of the programs included with the more mainstream versions.
The average user may not know all of the applications they need and which ones can eliminated . A focused distribution can fill this need.
Others were/are people wanting to do something different. Once those new items are incorporated into one of the more mainstream distributions, the innovator becomes obsolete and ultimately unused.
I am sure that several distributions were developed with unrealistic expectations and abandoned when reality hit.
Finances play a big part in the survivability of any distribution. Even if all hardware and labor is donated, there are still monitary issues that need to be addressed.
I see the operating system industry heading in a direction similiar to the restaurant industry. There will be some big players and a lot of small players, each catering to their marketing niche. A substantial number will fail. Those that survive will change due to competitive conditions in the market. Distributions merging is already happening, two or three become one and suddenly you have one or more candidates for the historical archives.
Successful distributions will be the ones that can cultivate long-term developer and consumer bases. The unsuccessful ones will still be contributing to developement, by leaving a record of what did not work.
If the distributions can maintain common standards, then the only problem will be which of these four hundred + is right for a particular user. This is where Distrowatch steps in. I only wish more packages were listed. Maybe Ladislav can clone himself multiple times and fulfill this little disire, but there are more important things.
Thanks again to all of those who make Distrowatch possible.
88 • Not too many distros (by warpengi on 2005-05-04 01:42:01 GMT from Canada)
Let's reuse the car analogy as happens from time to time with computers. Do we hear people saying their are way too many different kinds of cars. Too much choice just inhibits the car market from really developing. Ha, never heard that!
People have learned to enjoy the variety of choices and take pleasure in shopping for and comparing cars. Not everyone but most people. Back before redhat became RHEL there was Suse, Mandrake and Red Hat; green, yellow and red. Mostly the same under the hood yet the arguments about which was better could easily compare to the Ford, Chevy, Dodge arguments that I used to hear gowing up. You want an OS to take your data from here to there. Being foolish humans we also want to feel that we have the biggest, best, fastest, newest, whateverest. That's fine and that's why there is room for all the different distros. Whether you want a brand new Lexus or a 73 Mustang convertible that you built up from scratch in your garage the choice is yours.
Maybe all you really want is a go-cart for the dirt track out back or a Fischer-Price for the kids to run around on the patio, a motorcycle for the summer long weekends or a Hummer to hog lanes on the main drag. The choice is yours and Linux distros have all of those and more.
We don't have to beat Microsoft and dominate the market with "THE ONE" Linux. We have already bested the beast and are free. What more could we want?
89 • which distro will catch Ubuntu now? (by tonio on 2005-05-04 02:31:03 GMT from United States)
Several weeks ago we saw Ubuntu overtake Mandrake now Mandriva for the top ranking in Distrowatch. Many people predicted that this was going to happen. Now, which distribution(s) will try to bid for the top spot?
Many Linux/BSD's will get only more hits once people find out about what Microsoft wants to do for their next releases of Windows including Longhorn. Many people fear that Linux will no longer be free and that someone maybe even Microsoft will fight back and take control of linux. This should never happen, but many people fear that it will. There are going to be more lawsuits coming, but Linux should withstand all the pressures and continue to become better.
Best Regards to linux and bsd users
90 • to Tonio (by gnobian_ken00bie on 2005-05-04 04:47:08 GMT from United States)
Fortunately, there's no way for M$ or any other company to take control of GNU/Linux. The GPL makes that impossible, which is why M$ has called it "viral" and a "cancer".
Theoretically, it might be possible for someone to take a BSD and add so many proprietary improvements to it that it was vastly superior to any existing BSD. In actuality, as long as there are people who consider freedom itself a selling point, many would still choose the Free versions. But it could at least happen theoretically.
But with GPL software, the improvements must be shared as well, at least if their incorporated into GPL code. (I won't go into the linking to GPL libraries substleties.) Once the freedom is granted, it can't be revoked. At least not without securing the permission of all of the previous contributors - a daunting prospect even for a multi-billion dollar monopoly, especially since so many of the contributors are passionately committed to software freedom.
Lawsuits are another matter, but it appears even SCO has conceded that the Free Software Foundation's GNU contributions are unimpeachable on copyright grounds. And SCO isn't doing well at all against Linux either. Patents are another matter, but a lot of ridiculous software patents have been recently rejected by the courts.
We need to stay vigilant. But never afraid.
91 • LINUX/FOSS vs MICROSOFT Patents, licences & "Free Trade" (by ROBERT HUNTER on 2005-05-04 10:19:51 GMT from Australia)
The Australian Federal Government is pushing ALL educational institutions in Australia to adopt a monoculture of Microsoft products. Of course, this move is being vigourously opposed by the Open Source community, and non-Microsoft commercial software vendors. The exact motivation is not clear, certainly, the impedending so-called "Free Trade" agreement with the USA may have something to do with it. Plus Microsoft spreading FUD stories, citing the SCO/IBM bunfight, telling the OZ govt that Linux is a communist plot, a virus, the TCO of M$ is far better than Linux, citing so-called "indepedent" studies from M$ sponsored firms, and so on. Meanwhile, most of the main Australian trading partners, like asia and China etc, are adopting Linux wholesale...so in a few years our children will be IT illiterate, and unable to compete in a heterogeneous international IT environment. Am I afraid? Shit yes! But there are a lot of timid Aussie companies out there that will just "go with the flow". With M$ offering indemnity insurence, corporate sponsorship, etc. Then the use of Linux, BSD, OSS, etc may dwindle. For myself, I don't care, they can make using Open Source a criminal offense for all I care, it won't stop me using it in accord with the GPL. But these big transnationals are persistant, and immortal, and immoral. I think that the European Commission have got the right idea by not bowing down in the face of Microsoft's bullying tactics by requiring M$ to release certain portions of their TCP/IP stack coding. If M$ refuses, then the Euros will probably ban M$ from bidding in government software contracts of all European community member governments. I think the real killer is that we can't even tell if Microsoft is stealing Open Source code! So many people I talk to are apathetic about the whole thing, even some IT professionals who should know better! The money markets and the mass media are controlled by so few people, and now computing and the internet is going the same way! Sure, most silly patents applications made by M$, which have no grounding or relevence to Intellectual Property are being rejected. But, adding it all together, Open Source is under threat. So why don't we talk about this, instead of endless drivel on how there are too many or too few Linux Distros!
People, get a grip!
92 • Please help me in LINUX (by Zulfiqar Ali on 2005-05-04 13:38:38 GMT from Pakistan)
I am Zulfiqar Ali Form Pakistan live in city Karachi. I saw your Web Site and I am very nuch pleased to read many nice informations , thanks for your good works, and I shall be geting infoemation for this Web Site.
What I do to get more information on web please tell me more about LINUX.
thank you very much,
With best wishes
Zulfiqar Ali Hunzai
93 • Wonderful! (by William Roddy on 2005-05-04 18:19:53 GMT from United States)
I think the best reason to come down on the side of many, as opposed to few, distros, is DistroWatch. If there were few distros, there would be no DistroWatch. If there were no DistroWatch, there would not have been such a wonderful (and civil) exchange of views, such as there was this week.
Terry Pratchett calls it The Law of Unintended Consequences. Lots of distros equals DistroWatch. And right now, DistroWatch is the best place for people from all over the world to gather and to find portals into the wonderful worlds of GNU/Linux, BSD, open sources, live CDs, user interfaces, free software, and so much more.
From 400 mhz machines, to 64 bit machines, there are free solutions.
I learned about Fermi Linux from DistroWatch, which led me to learn about Scientific Linux, the latest incarnation. And though I have absolutely no need for an enterprise Linux, it is painfully clear to me that, with Scientific Linux being used by by countless thousands of the top scientist throughout the world, on servers, desktops, 32 and 64 bit machines, with scrutiny that would frighten most developers to death, there is no need to pay for a certified enterprise solution to computing needs. I'm using Scientific Linux 40 now -- with another distro on another partition --- just because it works perfectly, and because of the excitement I get from watching scientists pursue perfection.
Are there too many distributions? It is like asking if there are too many different kinds of animals in the world. Some people are allergic to cats. And I would love to own a giraffe some day, to what end, I have no idea.
As one writer pointed out, Microsoft is a monoculture and look what that has bred. Disease spreads in monocultures. Variety breeds strength.
Ladislav, thank you, a million times over, thank you.
And finally, gnobian_ken00bie, I don't know who you are, but I am glad you are a contributor.
Thank you all for your efforts and for the joy you bring.
94 • to Zulfiqar Ali (by gnobian_ken00bie on 2005-05-04 21:32:36 GMT from United States)
One of your best friends in learning about Linux - or as some prefer GNU/Linux - is www.google.com
This site, Distrowatch, is of course also a great resource. And the pages and forums of whatever distro you choose should be helpful.
http://www.tldp.org/ The Linux Documentation Project is a great resource.
If you want to learn the command line, I've found this to be a great place to start: http://www.linuxcommand.org/
But ultimately, you'll find google is invaluable. Surf the web for awhile each evening, following links, bookmarking resources as you find them, and you'll quickly discover that the web is filled with many many wonderful Linux resources.
Enjoy and welcome.
95 • thank yous (by gnobian_ken00bie on 2005-05-04 21:37:21 GMT from United States)
Thank you, Robert Hunter for the heads up concerning the goings on "down under". I confess vast ignorance concerning your beautiful country. I had thought your government was more tech savvy since Bob Hawke's initiatives. A good reminder of the importance of vigilance.
And William Roddy, thank you for your kind words - and for piquing my interest in Scientific Linux. I've enjoyed your contributions as well. I'm just an enthusiastic GNOME-loving GNU/Linux padawan, nobody special, but eager to learn and share.
And Ladislav, thank you for a wonderful website.
96 • Your recommendation (by William Roddy on 2005-05-05 00:59:17 GMT from United States)
gnobian_ken00bie, I just clicked on the Web site you recommended, http://www.linuxcommand.org/, and it is superb! Just what I needed. Thank you.
97 • too many distros? (by Steve on 2005-05-05 02:38:36 GMT from United States)
I's say that there are not enough as I have yet to find the perfect one for me. Mainly I want one that runs smooth, easy install, good package system, not a lot of junk software I'm never going to use, and easy 3rd party software compile and installablity. I started with redhat, but hated how they would move and rename libraries, which makes it so damn hard to compile software. Debian good once I finially got it installed but a lot of the software is really old. I currently use slackware.
98 • SLAX (by Max on 2005-05-05 06:43:16 GMT from Australia)
There we go, another SLAX release...
99 • Re: historical distros (by Ariszló on 2005-05-05 20:30:03 GMT from Hungary)
Here are some pictures of Yggdrasil Linux:
100 • The Rise of Ubuntu (by Leon Brooks on 2005-05-05 23:58:19 GMT from Australia)
I'm guessing that if there is diddly-squat difference in package lists, the progress of Ubuntu through the ranks - wartys and all - is a clear clarion call for "excitable" Debian-based distributions.
101 • Too many distros etc (by mikkh on 2005-05-06 11:47:46 GMT from United Kingdom)
One of the comments, states he rarely looks outside the top 25, which is pretty near sighted, not to mention idle.
I've tried most of the top 100, even the micro distros like puppy and DSL and based on Knoppix variant number 399.
Quick digress.... Based on, doesn't always mean inferior to, in fact quite the opposite in a lot of cases.
The following all outshine their more famous parent IMO
Blag 3000 (Fedora core 3)
Vector SOHO (Slackware)
Practically anything Debian based (Debian)
I've had flirtations with lots of distros and marked several as ones to keep an eye on, and quite a few as offering nothing special/poorly implemented - Ubuntu being one of those! Yoper and Vidalinux being two of the ones to watch. I keep coming back to a favourite few and they all live out of the top 25.
Conectiva is hard to beat as a solid, easy to install fast distro that is slick and obviously Mandrake haven't bought into it for nothing.
Currently lurking well below half way in the top 100, but IMO deserving at least a top 10 place - in fact I'd put it in the top 3
Conectiva uses synaptic for both the install and for package management (despite being RPM based) and the only minor niggle is the default repositories list contains very slow mirrors. A bit of detective work soon uncovers faster ones and my sources.list is available here http://homepage.ntlworld.com/michael.hopewell/sources.list
for those not sure how to insert new mirrors
Just copy my sources.list over yours in /etc/apt
I've sometimes have up to 7 distros installed, but 2 that always remain are Conectiva and Vector SOHO.
Slackware has been a favourite for a while, but it's now been replaced by Vector SOHO, because it seems more functional and better suited for a home desktop. It only uses 1 CD as opposed to Slackwares 2, but seems to cram more useful software in and be more usable from the word go.
Another one to recently impress me is Blag 3000. I always preferred Gnome when I first started tinkering with Linux, but have recently started to use KDE a lot. I actually prefer IceWM when it's done correctly, but most distros just toss it in as an afterthought and do little or nothing to make it usable. If you're one of those who thought icWM was a poor fourth behind KDE, Gnome and XFCE, take a look at Libranet or Vector SOHO which both do very good iceWM desktops. Back to Blag..... It uses a nice customised Gnome and like Conectiva, uses synaptic, which is one of the reasons I like it.
Too many distros are let down by poor package management and a factor often overlooked in distro reviews is how easy it is to use on a day-by-day basis. I find Suse's Yast annoying and slow for instance and Gentoo just annoying. Life's too short for all that compiling nonsense and the dependencies often seem ridiculously large for a distro that claims customisation to your own needs as a selling point. On a fresh install I wanted links ( text only browser) to review the documention again. Emerging links required 15 dependencies including xorg !! Oops, waffling again, back to Blag.... again
Here's a version of Gnome that I actually like using, and the repositories list just needed a few ticks adding in synaptic to have a decent software base to choose from - no detective work needed this time :o) What did impress me was the little test I do in Gnome only distros like this i.e installing k3b and seeing how much of KDE it makes me have. I've seen some where it practically ( or does) require the whole of KDE. Not with Blag though. kdelibs was already installed and just needed updating before it installed K3b.
Streamrunner, which I'd never seen before is installed by default and gives you access to thousands of online radio stations - nice touch :o)
Before I finish waffling, a quick word on OneBase.......
I wasn't too proud to plead poverty, and requested a free download.
I got a reply within an hour and was soon downloading the 120MB net installer ISO. OneBase was always on my one to watch list, and I was disappointed to see it go inactive/defunct a few months ago. It's a sort of Gentoo, without the pain. Not for the total noob, but it's straightforward if you have a minimal amount of Linux experience. After you've fetched the basepacks and kernel with two simple commands and rebooted a quick olm -b kde gets you KDE 3.4 and a nice fast distro. -b = binary but if you're feeling in need of a walk -s fetches the source and gives you a more optimised system.
102 • No subject (by Anonymous on 2005-05-07 09:39:41 GMT from Italy)
ladislav, have you read my suggestions about the distrowatch statistics? i would very much like to know your opinion on that. i worked on that list and i care about it. please read and answer.
103 • RE: Do Commercial Vendors of Open Source benefit the GNU/Linux Community? (by ROBERT HUNTER on 2005-05-08 10:26:47 GMT from Australia)
I've seen some comments about the utility [or lack of it] that commercial Linux distros offer the Linux and other OSS communities. There are distros like White Hat that build Red Hat Ent source code for folks that can't afford, but perhaps need an enterprise-style OS. [Such as NGO, charities etc]. All well and good. And it's good that we have commercial vendors around who can sell us support, for those who haven't got the time or motivation or whatever to become linux gurus themselves.
Also, having governments and companies as well individuals who are interested in Open Source makes them have a vested interests in the survival of Linux etc. Having a huge mass of ordinary folks using FOSS is great, but having governments and companies on-side is good too, as these linux users can help defend Open Source rights and freedoms.
I started to use Linux via RED HAT [Distro 4.2, I think], but of course, there are many wonderful distros out there, Gentoo, Debian, Slakeware, SuSE, and many more. And I have tried a fair number of them. Where would we be without the thousands [millions?] of wonderful people out there, writing code or documentation, or producing informative sites like distrowatch.
The there is RED HAT's fantastic RHCE programme. To the best of my knowledge, Red Hat is the ONLY vendor that has produced a practical "hands on" exam for an Open Source product. [of course there is CISCO's CCIE exam, but that is 100% proprierty]. What a change from those MCSE exams, where you just cram and pass, and have absolutely NO skills to run a network!
Of course, there is LPI, and these people have done a marvelous job trying to simulate a "real" computing skills test.
Do I get pissed at Red Hat...of course. Sometimes they do stuff which is a bit "Microsofty". At one stage, they refused to give paying students a copy of RHEL-3 to their fee-paying students! Imagine paying for a AUS$3000+ training course, and a AUS$950 exam, and you can only use RHEL-3 in class! And they were training US to administer their OS! What a pill! Red Hat have now reversed this situation, and give copies of RHEL3 or whatever to their students.
But passing a hands-on exam like the RHCT and RHCE gave me a lot of self-confidence. I felt a lot more confidence that I could tackle a real world problem, much more than with an exam like CompTIA's A+ PC hardware exam, or the MCSE exams.
So yes, I would say that commercial linux vendors do add to the richness of the Open Source experience. We have so many choices: those that like a "purist" solution go with Debian, or Gentoo, those who can afford it can patronise SuSE, Mandriva, Red Hat or whatever.
The only downside I can see from all this "distro-diversity" is that it creates some gaps sometimes. Particularly in hardware support. Why or why do folks buy hardware that cannot be used in ANY computing platform! We should DEMAND that hardware manufacturers stick to standards so that ANYONE can use ANY hardware with ANY operating system. [Even ..yuk!...Microsoft..]. Mostly, this is due to manufacturers producing cheap shit that does not comply with standards and they don't publish their [mostly stupid] customisations. So let's vote with our feet, and our money. I once saw a LINUX magazine promoting Microsoft wireless mouse/keyboards to install on a Linux machine. [www.linuxformat.co.uk]. What idiots! I can't count the number of hours I have spent on fiddling with "Winmodems" and "Winprinters" etc, getting them to work for my clients. And of course, I am by no means unique. So why would Linux users want to fill Microsoft coffers after they play these sort of games with us? And still when I go to Linux confs and meets I see ton's of "Microsoft" hardware! Are we masochists or what? Sure, there's a technical challenge, and a bit of fun, in hacking an X-Box for Linux,or FreeBSD, but what about the bigger picture?
The best hardware of all is stuff which you can shove into a computer and it just WORKS! Like my CISCO wireless PCMCIA cards. They just work! No fuss! This is how it should be. They are a bit more expensive, but such quality and open engineering should be supported by us! Then there is the Speedtouch USB ADSL modem. Again, it is easy to get working. And there is lots of this stuff out there. So let's start voting with our wallets.
Number of Comments: 103
Display mode: DWW Only • Comments Only • Both DWW and Comments
|• Issue 700 (2017-02-20): RaspBSD, Debian replaces Icedove with Thunderbird, Fedora's licensing guidlines, tips for switching shells, finding battery charge, getting IP address and killing processes|
|• Issue 699 (2017-02-13): Clear Linux, GhostBSD network utility ported to FreeBSD, Ubuntu coming to Fairphone, elementary OS crowd funding an app store|
|• Issue 698 (2017-02-06): Solus 2017.01.01, comparing containers with portable applicatins, Tails dropping 32-bit support, Debian Stretch enters freeze|
|• Issue 697 (2017-01-30): Subgraph OS 2016.12.30, running Ubuntu on an Android phone, Arch Linux phasing out 32-bit support, Linux Mint testing updated LMDE media|
|• Issue 696 (2017-01-23): GoboLinux 016, remotely running desktop applications, Solus adopting Flatpak, KDE neon using Calamares, TrueOS tests OpenRC|
|• Issue 695 (2017-01-16): Zorin OS 12, Peppermint team fixes installer bug, Debian refreshes Jessie media, Ubuntu improves low graphics mode, Exciting things coming in 2017|
|• Issue 694 (2017-01-09): MX Linux 16, Fedora considers systemd security features, DragonFly BSD to support massive swap space, Ubuntu Touch roadmap, Puppy's newsletter, sudo's password prompt|
|• Issue 693 (2017-01-02): Comparing small distros, fig language, video driver comparsion, Debian+PIXEL, Wayland on FreeBSD|
|• Issue 692 (2016-12-19): Bodhi Linux 4.0.0, Cappsule containers, Calculate's new Utilities package, Solus and Ubuntu MATE build new application menu|
|• Issue 691 (2016-12-12): SalentOS 1.0, openSUSE improves YaST, Fedora considers slower release cycle, KDE neon gets LTS branch|
|• Issue 690 (2016-12-05): Fedora 25, Ubuntu adopts rolling HWE kernel, running Android apps on GNU/Linux, Haiku working toward EFI support|
|• Issue 689 (2016-11-28): openSUSE 42.2, Fedora's upgrade path, plans for Korora 25, transitioning from PC-BSD to TrueOS, Webconverger's reproducible builds|
|• Issue 688 (2016-11-21): Endless OS 3.0.5, KDE neon fixes security hole, FreeBSD's Quarterly Status Report, Rolling release trial #2 concludes|
|• Issue 687 (2016-11-14): NAS4Free 10.3.0.3, Fedora gains MP3 playback, budgie-remix becomes Ubuntu Budgie, Ubuntu flavours compared, Rolling release trial #2|
|• Issue 686 (2016-11-07): FreeBSD 11.0, rolling release trial #2, Debian announces supported architectures, Simplicity switching to antiX base, farewell to Mythbuntu|
|• Issue 685 (2016-10-31): elementary OS 0.4, SUSE gains ARM support, Mint improves language support, Dirty COW explained, Rolling release trial #2|
|• Issue 684 (2016-10-24): Ubuntu 16.10, Linux popularity in different markets, Fedora runs on Raspberry Pi, Ubuntu features live kernel patching|
|• Issue 683 (2016-10-17): Refracta 8.0, making packages for distributions, Alpine switches to LibreSSL, 386BSD website publishes classic code|
|• Issue 682 (2016-10-10): KDE neon 20160915, Android-x86 6.0, Fedora warns of update bug, HandyLinux drops English translation, LXQt benchmarks|
|• Issue 681 (2016-10-03): OpenBSD 6.0, DragonFly BSD to support LibreSSL in ports, systemd denial of service bug, upgraded Mintbox Mini|
|• Issue 680 (2016-09-26): Uruk GNU/Linux 1.0, blocking applications at the firewall, Lenovo controversy, Ubuntu running on the Nextcloud Box|
|• Issue 679 (2016-09-19): OpenMandriva 3.0, 32-bit vs 64-bit performance, openSUSE updates, KaOS unveils first run wizard|
|• Issue 678 (2016-09-12): Apricity 07.2016, Mageia adopts DNF, KDE neon to use Wayland, FreeBSD updates Linux compatibility, creating cron jobs|
|• Issue 677 (2016-09-05): Peppermint OS 7, Manjaro updates leadership, TrueOS becomes rolling release, organizing files, creating torrents|
|• Issue 676 (2016-08-29): Korora 24, Fedora 25 to use Wayland by default, Linux turns 25, PC-BSD becomes TrueOS, finding software licensing information|
|• Issue 675 (2016-08-22): Gentoo LiveDVD "Choice Edition", moreutils, Ubuntu improves terminal convergence, MATE packaged for Openindiana, FreeBSD improves video support|
|• Issue 674 (2016-08-15): Zenwalk Linux 8.0, Ubuntu phone follow-up, Lubuntu transitioning to LXQt, Steam running on FreeBSD|
|• Issue 673 (2016-08-03): noop linux and EasyNAS, Debian's GnuPG switch, Fedora "Flock", using "nice"|
|• Issue 672 (2016-08-01): Ubuntu Phone 15.04, Solus embraces rolling release model, interview with Jane Silber, FreeBSD Quarterly Report|
|• Issue 671 (2016-07-25): Slackware 14.2, Point Linux 3.2, OpenBSD disables usermount, KaOS releases significant changes, Fedora 22 reaches end of life.|
|• Issue 670 (2016-07-18): Linux Lite 3.0, Bodhi team plans 4.0.0, pfSense changes licensing, running software across distributions, Linux Mint upgrade path|
|• Issue 669 (2016-07-11): Linux Mint 18, proving a system is secure, LibreSSL in FreeBSD, Ubuntu plans phasing out 32-bit, pfSense status report|
|• Issue 668 (2016-07-04): Fedora 24, Linux Mint plans for 18.1, FreeBSD and DragonFly BSD improve their file systems, comparing Flatpak, Snap and AppImage|
|• Issue 667 (2016-06-27): GeckoLinux 421, Fedora supports Flatpak, Solus unveils new features, running GNU/Linux on tablets|
|• Issue 666 (2016-06-20): Comparing more live update methods, Ubuntu's snap packages, Antergos drops 32-bit media, GeckoLinux unveils Rolling edition, learning Linux resources|
|• Issue 665 (2016-06-13): BunsenLabs Linux Hydrogen, Fedora 24 delayed, NetBSD grows in size, Clonezilla questions|
|• Issue 664 (2016-06-06): Sabayon 16.05, Debian updates install media, the cost of free software, Qubes explains secure build process|
|• Issue 663 (2016-05-30): Comparing live update methods, Ubuntu MATE's progress, distros debate systemd change, DistroWatch turns 15|
|• Issue 662 (2016-05-23): Clonezilla Live, new Fedora community repository, DragonFlyBSD runs Wayland, a live edition of Slackware and kernel components|
|• Issue 661 (2016-05-16): FreeBSD 10.3, OpenMandriva adopts Clang, Debian adds ZFS packages, PCLinuxOS drops 32-bit and comparing CentOS with RHEL|
|• Issue 660 (2016-05-09): Ubuntu MATE 16.04, Mint's xapps, FreeBSD Quarterly Report, Debian updates 32-bit support, addressing GPL violations|
|• Issue 659 (2016-05-02): Ubuntu 16.04, compiling custom kernels, Cinnamon 3.0, Sabayon launches ARM build, Devuan ships Beta release|
|• Issue 658 (2016-04-25): Kali Linux 2016.1, elementary OS 0.3.2, Debian elects Project Leader, Fedora 24 feature preview, Nard reaches 1.0|
|• Issue 657 (2016-04-18): Redox, Linux Mint improves update manager, planned Fedora 24 features, Ubuntu 16.04 getting Snappy packages|
|• Issue 656 (2016-04-11): Qubes OS 3.1, Whonix offers bug bounties, Puppy's family tree, setting up disk partitions and running bash on Windows|
|• Issue 655 (2016-04-04): Parsix 8.5, Sabayon's Community repository, Red Hat offers free subscriptions, Ubuntu tablets, command line tips|
|• Issue 654 (2016-03-28): PCLinuxOS 2016.03, Using signatures to create a web of trust, Arch Linux rolls out Pacman update, GuixSD packages GNOME|
|• Issue 653 (2016-03-21): Antergos 2016.02.21, Debian prepares for election, a Unix-like OS written in Rust, watching Netflix on FreeBSD|
|• Issue 652 (2016-03-14): ReactOS 0.4.0, Debian swaps Iceweasel for Firefox, Fedora moving forward with Wayland, Verifying ISO files|
|• Issue 651 (2016-03-07): Korora 23, Linux Mint improves security, Ubuntu MATE on Raspberry Pi 3 computers, trying different file systems|
|• Issue 650 (2016-02-29): Haiku in 2016, running Android apps on GNU/Linux, 30 years of MINIX, Fedora plans Atomic Workstation|
|• Issue 649 (2016-02-22): Zorin OS 11, openSUSE launches new editions, Linux Mint website compromised, sandboxing applications using Firejail|
|• Issue 648 (2016-02-15): XStream Desktop 153, Raspbian unveils OpenGL feature, free hardware, Ikey Doherty talks desktop design|
|• Full list of all issues|