, we made a number of
predictions about what the year held for the Linux community. One of those
read as follows:
Desktop Linux will be taken far more seriously by the end of the
year.... At that point, the Linux desktop will have almost
everything needed by a large number of desktop users. More
specialized applications will take years to fill in, but the basics
are coming into place.
Normally we don't say much about our past predictions, in the hope that our
readers will forget them as soon as possible. We may not do any worse than
those analyst groups that sell their predictions printed on heavy paper,
but we still find ourselves embarrassed by the things we say at times. In
this case, however, we just might have gotten it right.
The latest development on the Linux desktop front is SuSE's announcement of the "SuSE Linux Office
Desktop," a new version of its distribution which is due out in January.
This distribution is, of course, aimed at the desktop market; it features
(relatively) easy administration, a full set of office productivity tools
(based on StarOffice), and CrossOver Office for those proprietary
applications that simply cannot be done without.
SuSE, of course, is not alone in its new emphasis on the desktop. Red Hat
Linux 8.0 includes a reworked, friendlier desktop. Distributions like
Lycoris and Xandros are aimed at desktop users; Mandrake Linux, of course,
has always had this emphasis. There is a Debian Desktop
Project out there. Linux systems can even be purchased at outlets like
Not too long ago, even the strongest Linux
advocates mostly agreed that Linux was only suited to
server-oriented tasks. Now, more and more people think that Linux is ready for
desktop tasks, and, perhaps more to the point, that there is money to be
made in desktop Linux.
One might well wonder why desktop Linux is coming into its own now. There
are several possible reasons:
- The set of free desktop applications is maturing. Tools like
OpenOffice, AbiWord, Gnumeric, Mozilla, Konqueror, etc. have reached a
point where they are good enough for most users. The feature lists
may still fall short of the proprietary competition in some cases, but
most of the truly important features are there.
- The Wine project, in the form of products like CrossOver Office,
has, after many years, reached a point where it can run the
proprietary applications desktop users rely on. The availability of
these applications makes the Linux desktop that much more valuable.
- The difficult economy and Microsoft's licensing schemes have made
companies more interested in ways of saving money.
- People are finally beginning to notice that Linux users don't
have to spend their time fighting the virus of the week.
- Linux has clearly survived the dotcom crash - a fact which still
surprises many people. Fears that Linux will vanish like so many
other highly-hyped technologies are fading away.
The theory of "disruptive technologies" states that a new technology does
not have to be better than the one it replaces - at least, not in every
way. It is enough to offer advantages, financial and otherwise, that are
sufficiently compelling to get people to make a change. Linux (and free
software in general) have a lot to offer in cost savings, security, rapid
and open development, freedom from vendor lock-in, etc. Increasingly,
Linux also has applications that perform widely useful functions, and which
are becoming easier to use. Many of these applications are on their way
toward becoming the best available, free or otherwise. We are, it seems,
reaching that point where the balance begins to tip. This may truly be the
beginning of the era of the free desktop.
We should not lose track of the fact that a great deal remains to be done
before free desktops can truly achieve World Domination, however. Linux
administration is getting easier, but remains difficult. Linux
applications still lack features that many users want. A visit to any
computer store will show that there is a whole range of applications that
are still absent on Linux: where are the children's games, menu planners,
language courses, tax return preparers,
home remodel designers, and makeover assistants for
Linux? When your Linux system will help you look like the Cosmo Girl,
we'll know we have truly arrived. But that day will remain distant until
Linux becomes a more friendly platform for proprietary applications.
It is also worth noting that development on the Linux kernel has emphasized
performance on very large systems just as it looks like the Linux desktop
is going to take off. Performance on smaller systems is supposed to be
addressed during the stabilization period. Testing by desktop users will
be an important part of that process; as more people test out the
development kernel in the coming months, it becomes increasingly likely
that the next stable kernel release will meet the needs of desktop users.
The true triumph of the free desktop is still probably some years away. A
great deal of hard work remains to be done. But the results of years of
effort by thousands of developers determined to improve the Linux desktop
experience are beginning to be felt in a serious way. It is going to be
fun to watch where things go from here.
Comments (11 posted)
The MITRE corporation has just released the results of a study it performed
on the use of free and open source software (which it calls "FOSS") within
the U.S. Department of Defense. It is an interesting look at how the DoD
uses free software, and what would happen if an anti-free-software policy
were to be adopted. The full study is available as a 160-page PDF file
you'll find a rather shorter summary of what it says.
The question that this study was meant to answer seems to be "should the
military ban the use of free software?" The conclusion they came to is
Neither the survey nor the analysis supports the premise that
banning or seriously restricting FOSS would benefit DoD security or
defensive capabilities. To the contrary, the combination of an
ambiguous status and largely ungrounded fears that it cannot be
used with other types of software are keeping FOSS from reaching
optimal levels of use.
Looking at one area in particular, the report continues:
The main conclusion of the analysis was that FOSS software plays a
more critical role in the DoD than has generally been
recognized... One unexpected result was the degree to which
security depends on FOSS... Taken together, these factors imply
that banning FOSS would have immediate, broad, and strongly
negative impacts on the ability of many sensitive and
security-focused DoD groups to defend against cyberattacks.
The report looks at free software licenses in considerable detail in a
deliberate attempt to address a number of institutional fears about those
licenses. Worries about licensing, say the authors, have led to a
suboptimal level of free software usage. It is a reasonably
straightforward and accurate study; for added fun, they look at the EULA
for Microsoft's "Mobile Internet Toolkit" and compare its terms with those
of free licenses. "However, unlike the Microsoft MIT EULA, the GPL
places no constraints on software simply running on the same system, and
actually goes out of its way not to intrude on other licenses outside of
The report includes a survey of how free software is used within the DoD
now. They break that usage down into four categories:
- Infrastructure, using tools like sendmail and apache.
- Software development, especially with gcc and Perl.
- Security, including intrusion detection systems, security
analysis tools (i.e. SARA and Snort), and secured operating systems
like OpenBSD. "Yet another important way in which FOSS
contributes to security is by making it possible to change and fix
security holes quickly in the face of new modes of cyberattack. This
ability, which allows rapid response to new or innovative forms of
cyberattack, is intrinsic to the FOSS approach and generally
impractical in closed source products."
- Research, which benefits from Linux clusters and the general
culture of free software.
The report authors looked at costs, of course:
More often than not, the strongest deciding factors for choosing
FOSS products were capability and reliability, with cost being an
important but secondary factor.
They note one other important factor regarding free software and costs:
Without the constant pressure of low-cost, high-quality FOSS
products competing with the closed-source products, the
closed-source vendors could more easily fall into a cycle in which
their support costs balloon and costs are passed on to their
The report concludes with three recommendations that, they say, would help the
DoD make optimal use of free software. They are:
- Create a "generally recognized as safe" list of free software. 115
free applications found by the survey would be the starting point for
this list. Suggested "applications" include, however, Linux, OpenBSD,
NetBSD, and FreeBSD, so this list would be pretty general.
- Develop generic infrastructure, development, security, and research
policies. These policies would promote the use of free software in
situations where it is deemed appropriate.
- Encourage use of FOSS to promote product diversity.
"Acquisition diversity reduces the cost and security risks of
being fully dependent no a single software product, while
architectural diversity lowers the risk of catastrophic cyber attacks
based on automated exploitation of specific features for flaws of very
widely deployed products."
Finally, a set of appendices provides lists of free software applications
in use within the DoD, and the full text of a large number of free software
If the DoD was seriously considering banning free software, one can only
hope that this report will put an end to such thoughts. Through a great
deal of detailed research, the report's authors have demonstrated that the
Department of Defense is already heavily dependent on free software, and
would be badly hurt if use such software were forbidden. Increasingly,
free software is crucial part of the systems we all use, and that, of
course, is a good thing.
Comments (7 posted)
It's time for our weekly report to our readers. Read on for the latest
subscription counts and a few bits of site news.
As of this writing, we are getting close to 2200 subscribers. That still
leaves us far short of our medium-term goal of 4000.
Things are headed in the right direction,
however; with continued support from our readers, we hope that we will get
to where we need to be before too long.
We are also encouraged by a small increase in the rate of corporate
subscriptions. They still fall short of our hopes, but there are signs
that the bureaucratic wheels are beginning to turn. If you work for a
company that could benefit from a subscription, please consider talking to
them about setting one up.
This week we were also able to announce a
group subscription for the Debian project, which has been funded by HP.
Debian developers are encouraged to read the announcement for information
on how to get access to this subscription.
For those of you who have been requesting the ability to pay with American
Express: we have finally managed to get that set up. Progress on setting
up a Euro-zone bank account has been slower; it looks like that will not be
a viable option anytime soon. The best approach for
accepting funds from Europeans without credit cards may turn out to be to
those people send us checks. We're still working on that one, though.
There has been a small stream of requests for a stable URL for the latest
free version of the Weekly Edition. That has now been implemented; the
current free weekly can be found at:
Of course, lwn.net/current continues to refer to the
most recent (subscription) Weekly Edition.
We have been having some trouble with sites blocking mail from the LWN
(things like the various LWN mailing lists and subscription notices). That
mail originates from our production server, which is donated to us by
Rackspace. Some people, evidently, have received a lot of spam from
Rackspace-hosted systems, and have simply blocked the entire Rackspace network.
Rackspace tells us that they shut down spammers as soon as they know of
them, but it's an ongoing battle. Meanwhile, we are looking into other
generating and routing mail so that this problem, hopefully, will be behind
For those of you making your holiday shopping lists: LWN gift certificates
will be available shortly. The work is mostly done, but won't be completed
at this point until after the weekly publication cycle. Stay tuned for the
That is the LWN news for this week. Thanks, as always, for your support.
Comments (19 posted)
For the second week in a row, we have no "letters to the editor" page,
since nobody sent us any letters. The reduction in readership caused by
the subscription gate probably has a lot to do with that. Still, we would
like to hear from you; if you have comments you would like to see
published, please feel free to send them to email@example.com
Comments (5 posted)
Page editor: Jonathan Corbet
Inside this week's LWN.net Weekly Edition
- Security: Protecting DNS; new vulnerabilities in inn, Kerberos, and Zope.
- Kernel: The kernel gets crypto; Reiser4; <tt>sys_epoll</tt>.
- Distributions: Clusters: Mandrake's CLIC and openMosix
- Development: The Debian Desktop Project, ALSA 0.9.0rc5, SapDB on CVS,
BusyBox 0.60.5, PostScript margins, Mod_python 3.0 beta 4,
LSB v1.3 draft, Sweep 0.5.10, Galeon 1.3.0, KDE 3.1 RC1,
Python Gamelets, SBCL 0.7.9.
- Press: Linux vs. Windows installation, Open Source is good for America,
SourceForge grows to 500K users, Congress takes aim at open-source,
Linus on the status of 2.6, Linux for tech writers.
- Announcements: TV Linux Alliance, AMD Developer Center, Building Online Communities,
KOrganizer Workshop, LAD Techies Database, UMeet conference