http://www.eweek.com/article2/0,1759,1585522,00.asp
While Longhorn has been discussed repeatedly in this blog, an aspect which has not received as much attention as it should is that of backwards compatibility. As the indexed article suggests, a lot of things which worked with previous Windows versions will not work with Longhorn. Balancing the needs for operating system effectiveness [and, increasingly, security] with those for gegacy support is a contentious quesiton at the best of times. But in this case, abandoning legacy support may make sense to Microsoft in economic terms as well.
An alternative to abandoning legacy support is to retain it with the added reliability of Linix, as discussed in this article:
http://www.eweek.com/article2/0,1759,1586641,00.asp
The fact that Linux can run legacy Windows applications with ease may be one of its major competitive advantages in the Longhorn future.
http://www.informit.com/articles/article.asp?p=170772
While much of the current security furor concerns itself with Windows systems, the plain fact remains that other systems, like UNIX, can also be vulnerable to the wily hacker. Given that UNIX systems often run high-value assests, ensuring security on this OS also needs investigation and effort. The current article shows how user home directories are a security vulnerability on a UNIX system, and since the concept of eliminating all users is a tad counter-productive, the author gives some practical recommendations on how to secure these home directories.
Related links in sidebars address other articles dealing with UNIX and Linux security.
http://www.masternewmedia.org/2004/04/04/nine_rules_for_good_technology.htm
Problems with the Open Source Software model have been mentioned in this blog in previous entries, and one recurrent theme is the aura of legal uncertainty which currently surrounds OSS. The article indexed by the URL is a long, thorough, sound discussion of the legal issues in which Open Source Software is enmeshed, and should go a long way, if information is necessary, to enabling orgnizations to set policy on these issues.
http://www.microsoft.com/technet/itsolutions/migration/linux/mvc/win2kcd.mspx
While most of the buzz these days is about migrating from other operating systems to Linux, with Windows being one of the major feedstock pools for this, in fact the migration may proceed the other way, in some cases. Microsoft's site, at the indexed URL provides a complete guide to Linus to Windows migration, inluding a virtual conference covering migration in extensive detail, white papers on pre-, post-, and ongoing-migration issues, case studies, analyst reports, and additional resources for specific platforms and activities.
If you want to go this way, Microsoft wants you to be sure to know how to do it. The fact that this in-depth how-to covers 6 pages of extensively illustrated and annotated articles is testimony to that.
http://www.aci.com.pl/mwichary/guidebook/index
This website preserves and showcases Graphical User Interfaces, with a featured interface of the day, featured component, and featured icon. News, a featured site, and contact information round out the home page. This is a good way to compare the look, if not the feel, of different OS without having to install them on disk partitions, and should find a place in many teachers' toolbox.
http://weblog.siliconvalley.com/column/dangillmor/archives/010227.shtml#010227
The requirement for Liunux to offer a better user experience to challenge Microsoft for the SOHO market in a big way has been made in this blog before. One of the key components of this is a high quality GUI, another notion that is not exactly new to this blog. Installation issues have been adduced as a real deal-breaker in particular.
One of the more contentious issues relating to Linux desktop use has been about cost. Yet there is no question that Linux has been making progress as desktop alternative. All of which serves as an extensively self-referential introduction to this article by a Linux skeptic, who has nevertheless kept trying, and who now considers that the gap is indeed narrowing rapidly enough to start thinking about Linux as a desktop alternative.
http://www.linuxinsider.com/perl/story/33301.html
Interesting overview of 8 textbooks which are popular for teaching IT to MBA students, and the degree to which they minimize Linux, or are positively hostile towards the alternative operating system. Textbooks, by their nature, tend to the 'plain vanilla', and it is not surprising to find that the sorts of errors for which humanities texts get castigated in vain also crop up here. It is certainly worth noting by IT educators the degree to which text bias exists.
Interestingly enough, if you move towards books intended for computer science types [for example, general titles on operating systems], there Linux and UNIX get all the coverage they deserve, which is extensive.
To turn a phrase: "I mild less who writes the nation's laws, and more who writes the nation's business texts".
Here is something which could have any number of major educational uses: the UNIX Forums [sic]. In addition to network computing topics:
UNIX for Dummies Questions & Answers
UNIX for Advanced & Expert Users
C Programming in the UNIX Environment
IP Networking for Dummies Questions & Answers
Network Security for Dummies Q & A
many of which are obviously directed at the newcomer, special interest topics include Linux, SUN Solaris, HP-UX, AIX, OS X, SCO[!], and BSD. Treated as well are general topics, downloadable UNIX and Linux benchmarks, and UNIX Standards.
http://news.com.com/2100-7344_3-5177257.html
Novell, which last January bought the major Linux distributor, SuSE, as well as the Ximian desktop, has adopted the strategy of coupling the desktop and office productivity software to its operating system. This takes a leaf from Microsoft's book, since it is this strategy which largely accounts for Microsoft's dominance. If correctly implemented and marketed, such a software suite should be extremely interesting to companies looking to avoid the pricing and intellectual property straightjacket Microsoft now has on them. The nature of Linux suggests that the same constraints will not (and cannot) be implemented by Novell.
Even though this represents a major departure for Novell, the company still intends to support its venerable NetWare network operating system [which currently has about 5% market share], and is considering developing a new version, itself tightly integrated with Linux.
For another viewpoint, which characterizes Novell's announcement as an indication that stand-alone NetWare is going away, see:
http://www.eweek.com/article2/0,4149,1552516,00.asp
http://techupdate.zdnet.com/techupdate/stories/main/Six_barriers_to_open_source_adoption.html
A look at the hurdles which must be overcome in implementing Open Source solutions [particularly Linux], as well as some evaluation of just how high these hurdles are:
* Lack of formal support: in fact availability of ongoing support will be essential to enterprise acceptance of Open Source, and serious solution developers are already providing this.
* Rapid pace of change: represents a balance issue -- it is good to have bugs and vulnerabilities fixed quickly, as Open Source has shown it can do, but the sorts of timeline pressures inherent in Windows versions is entirely absent.
* Roadmapping problems: in fact having a roadmap does not mean that you are going to get there. The Open Source community tends to be better at long-range roadmapping.
* Functional gaps no doubt exist now, but they are in the process of being addressed, and should be closed over time.
* Licensing issues remain a problem, and new licence tools may have to be developed.
* Independent software vendor endorsements may be of limited impact.
In general, there are concerns, but no showstoppers.
http://www.linuxjournal.com/article.php?sid=6811
'Port knocking' -- only allowing systems to connect if they implement a sequence of closed port access attempts -- is an idea which could help VPN security and similar implementations in Linux. This article shows how do to it -- it represents a useful addition to the whole armour of security which administrators must implement these days.
http://www.pcworld.com/news/article/0,aid,114418,tk,dn012304X,00.asp
Weaning oneself off Windows and migrating to another operating system [Linux, in the case of the indexed URL] is not something to be contemplated by the technologically faint of heart. Face it, people don't want to have to learn something different, even if in the end, it is also better. The problems and rewards are discussed here, and could serve as a source for internal position papers for organizations who would rather switch than fight.
For a more in-depth look at migrations, check out this article:
http://www.tomshardware.com/howto/20040329/index.html
which promises to be the first in a multipart series, and which provides checklists for use during actual migrations.
IBM has already provided a 9-part roadmap to moving from Windows to Linux:
http://www-106.ibm.com/developerworks/linux/library/l-roadmap.html
A similar take on OS X [made more similar, of course, by the BSD-UNIX roots of the latter operating system] can be found here:
http://www.pcworld.com/news/article/0,aid,114464,tk,wb020204x,00.asp
http://www.eweek.com/article2/0,1759,1545612,00.asp
The indexed article examines the 2.6 major revision of the Linux kernel and finds it has made significant improvements for enterprise support. Of course in the production environment, you want an integrated distribution, not merely a kernel, and one example, 'Fedora', is reviewed here:
http://www.eweek.com/article2/0,1759,1542890,00.asp
The equivalent offering from SuSE is discussed and reviewed in these two articles:
http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-145-1-1-618817-6964-1
http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-145-1-1-618817-6967-1
On a related note, in the educational environment, the ability to run a Linux distribution from CD can be a valuable attribute in teaching as well as for laboratory work, and a two-part article on this can be found here:
http://www.extremetech.com/article2/0,1558,1457025,00.asp
http://www.extremetech.com/article2/0,1558,1544559,00.asp
with the drawbacks and advantages of this approach clearly detailed.
http://www.linuxinsider.com/perl/story/33089.html
Seasoned computer professionals can recognize the difference between Linux and Windows seven times out of nine, but that is not quite the issue being raised in this article, which is differentiation -- the change in value an organization realizes from implementing one solution rather than the other. Unsurprisingly given its venue, the article comes to the conclusion that the market-driven nature of Windows results in bass-ackward engineering, striving for features first, and layering compatibility afterwards.
One thought strikes immediately: it may well be less Microsoft's being driven by the market which is the problem, if any exists, and more Microsoft's misperception of what the market it. This is particularly the case, I would argue, in the case of operating systems, where I think most IT professionals would choose stability over features any old day.
http://www.nwfusion.com/best/2004/0223os.html?fsrc=rss-microsoft
Benchmarking and comparative testing are one of the most hotly disputed topics in IT -- it have been ever thus. The authors of the indexed article effectively admit that in their introduction, which pittedWindows Server 2003 against Novell NetWare, Red Hat Linux, Apple's OS X, and SUsE Linux. The result found that Server 2003 was the best performer.
Naturally, this result will be disputed by all the losing parties, often quite vigorously. It is, however, an interesting case study to assign to students, since they should read the report, look for any evidence of bias, and be able to give some estimate of how reliable the reported outcome really is.
http://news.com.com/2100-7344_3-5166155.html?tag=nefd_top
Article analysing server sales, showing that Linux servers enjoyed high-end growth towards the last quarter in 2003. That revenue built 63% to $960 million [still rather small compared to Microsoft figures of $3.9 billion]is testimony enough to the rapid adoption rate of Lunux, but that unit shipments only grew 53% suugests that more powerful versions of the Linux product are being served. This in turn suggests a growing role for Linux servers in high-perfomrance computing.
Inescapably, Microsoft views Linux as a top threat, particularly with the latter's higher growth rate. A rebound in IT spending is clearly on the way as part of these developments.
Related articles are linked to the one cited.
http://www.arstechnica.com/reviews/004/software/kde-3.2/kde-3.2-01.html
One of the things that may disorient Windows veterans on first encountering GUI Linux is that there is more than one 'desktop environment' [and in fact, understanding all the windows managers and their various options can be a major task in itself]. KDE - the K Desktop Environment - is one of the major components in the Linux GUI world, and this highly detailed and deeply linked article gives a fine idea of its past, present, and potential future.
Covering the project structure, new developments in the February 2004 release, the internals of KDE, the development environment, licence issues, and where to get the KDE, this article is a major source of information for anyone interested in the Linux desktop GUI.
An effective GUI is essential for Linux success in the desktop arena in general, and the SOHO environment in particular, and with this latest release, it certainly seems like the KDE has provided Linux with this requirement.
http://www.nostarch.com/litw.htm
No Starch Press has released its user guide: Linux in the Workplace online under the GNU FDL, which means you can follow the link given on this site and access HTML and .PDF versions of the book at no cost. The guide itself is meant for desktop Linux users, and skips over the administrative bafflegab which afflicts most users with a severe case of eyeball-glazing. The K Desktop Environment is emphasized, and the level is set to the complete beginner
This therefore is an excellent resource to learn about practicing with a major Linux GUI [and make no mistake, Linux has to offer a GUI if it is to succeed on the desktop, either in the organization or at home]. The book's intended audience makes this eminently suitable as reference for classes covering introductory computer applications, since both operating system and common productivity programs are covered.
http://www.frozentech.com/content/livecd.php
This is simply an annotated list of live links to the dozens and dozens of CD-ROM-based Linux distributions out there. Sortable columns comprise ISO image size, name, and primary function of the distribution. This is an ideal first stop if you are seeking a particular type of Linux distribution [e.g. one with a small installed footprint] and don't have any names in your head.
Those teaching Linux could use this as a source for comparative projects relating to distributions.
http://ct.com.com/click?q=ea-fjLyQhV9N0j5j379g_F_TXW~t_PR
Clustering is a key strategy for managing server farms in large-scale production environments. The Windows way of doing this is relatively straightforward, but Linux can do this as well, and this white paper "GPFS Primer for Linux Clusters" provides details of the General Parallel File System Version 1, Release 3 which helps underpin this ability.
United Business Media's CMP division has launched a set of tightly focussed searchable Web pages called 'pipelines', which index news, trends, how-to-do-its, products, white papers, webcasts, sponsored links with downloadable software, and a glossary. Those of specific interest to most applied IT teachers are:
http://www.securitypipeline.com/ covering desktop, network, and infrastructure security plus policy & privacy.
http://www.linuxpipeline.com/ covering core Lunux, applications, enterprise open source, and business.
http://www.networkingpipeline.com/ covering security, infrastructure, wireless, and voice/data integration.
http://www.serverpipeline.com/ covering entry-level, mid-range, and high-end servers, plus their supporting technlogies (including operating systems).
http://www.itutilitypipeline.com/ covering utility computing and services, grid computing, and enterprise systems.
http://www.desktoppipeline.com/ covering desktop operating systems, application software, and hardware as these relate to all current desktop OS.
Additional pipelines address small business, mobile computing, and storage issues. These look like excellent information sources to benchmark and revisit, for students and teachers alike.
http://www.linuxdevices.com/articles/AT3855888078.html
Considering the importance of kernel versions to the Linux world, and the fact that there is an extensive installed base of the 2.4 kernel, some advice from an expert on the issues in migrating to the 2.6 kernel is not out of place. That is exactly what this, the first in a series on 2.6 kernel migration, is all about.
It provides a useful source of detail to supplement instruction on kernel versions in Linux and why these are useful and necessary.
Migrate on!
http://www.wired.com/news/infostructure/0,1377,62016,00.html
Effective mimicry is often a key element in the Darwinian struggle, which gaves more point to this LinuxWorld report indicating that Linux is steadily moving to the desktop, gaining a Windows look-and-feel as it does so. Related stories are also listed on this site.
I have consistently advocated that anyone bringing out a Linux distro which is as easy to install and use as Windows [and achieving that goal is not hard] could corner the home market in an extremely short time, if marketed correctly. Implemening a GUI which will work for the Windows-aware crowd is one important key to this strategy.
http://www.techcentralstation.com/012204A.html
Article looking skeptically at the concept of open source, particularly as it relates to Linux. While it acknowledges the strengths and accomplishments of the open source movement, it also makes clear that ongoing maintenance is an intensive and costly chore. Moreover, software is a unique case where open source freedom can even attempt to apply -- it does not represent an extensible economic model.
The final claim, that open source theorists advocate a form of socialization of creation through government subsidy seems both wrong-headed and misplaced. It is by no means clear that all, if many, of the open-source community are advocating this. Nor is it clear that there is something inherently wrong with the diversion of government resources to creative community support -- it was on this basis that the InterNet itself was created, for example.
The broader, more supportable point made here: there is no one model of software development which is correct in all circumstances.
http://www.microsoft.com/mscorp/facts/
As Mark Twain said: "There are lies, and there are damn lies, and there are statistics". Microsoft is feeling sufficient competitive pressure from Linux that they have started dishing out "the facts" which show that Linux is more costly while performing less effectively than Windows. True, false? There is some educational value in using sources like these as a case study in source bias, even if the facts are correct.
This site also misses the point that many are disenchanted with Microsoft for a variety of reasons other than cost -- and indeed, I would myself support an alternative OS of proved reliability even if the running costs [apart from the conversion costs] were equal, for many of those same reasons.
http://www.linux-mag.com/2003-01/kernal_01.html
One of the more metaphysical questions relating to Linux is: "what is it?" -- this because the only portion of the OS which is truly, madly, deeply Linux is the kernel. All of the rest of the supporting 'pariphernalia' represents Something Else -- almost universally GNU, but not necessarily. The sorts of users which Linux must serve if it is to become a viable desktop alternative to Windows will suffer severe M.E.G.O. in relation to such a debate, and this is worth hoisting inboard by Linux enthusiasts.
That said, and it certainly has been, by me at least, the Linux kernel development has some of the same flavour as watching WINDOWS open into new versions, in terms of a spectator sport, and this article discusses some of the kernel foundations and what directions are likely for its future development.
http://www.preshweb.co.uk/linux/howtos/dos/
While I regard DOS as such a historical relic that I do not even provide an indexing term for it in this blog, there may be times when you have to run a DOS applicaiton on a Linux system, and you don't need the cost or power of a full-blown OS emulator like VMWare. This article describes how to use a free DOS emulator along with a copy of the FreeDOS operating system, complete with active links.
Much of the specific focus is on making games work, but it is still a well-written and helpful primer for anyone wanting to run DOS under Linux.
Perhaps nothing symbolizes the Linux-Windows divide as much as the fact that the former recompile their kernel about as much as the latter recover from crashes. While I don't think this is the sort of thing we want most home users to be have to be doing, the fact remains that for any degree of competency in Linux, familiarity with kernel building issues is a must.
Of course, a hand-built kernel implemented by someone who is savvy about hardware and security will be both fast and resistant to disasters, which is another bonus, for sure.
But I still think most people are going to want to build their own kernels about as much as they will want to change the piston rings on their cars, and that too is a fact of life.
No sooner do I write a couple of screeds fulminating about the advantages of the Linux desktop for all and sundry than I come across, courtesy of the wonderful Lockergnome Linux Fanatics listserv, the site for the Desktop Linux Consortium, which, one strongly suspects, is a consortium of parties interested in desktop Linux.
Since it is just at the 'getting started' stage, if making the waters of Linux rise floats your boat, why not mosey on over and hop on in?
http://www.linux-mag.com/2003-09/viruses_01.html
The lesser vulnerability of Linux to viri and similar malware is a plus, but "lesser" does not mean "none at all", as this article, which discusses Linux viri, clearly indicates. It provides a good primer on how Linux can be affected by a virus, with some case studies, and it also touches on the added problems involved with running a Windows emulator.
All this means is that the shouting about the 'weaknesses' of Windows, which has come from several sides in the OS debate, increasingly comes down to the issue of a software monoculture. Even though I accept that by design and heritage Linux and OS/X are somewhat more resistant to malware writers [and I think I am toning down this appreciation a bit], the fact remains that the major reason why Windows systems are most affected is because of their majority status.
Taken to its extreme, if alternative OS deployed in homes and offices shrug off most of the Windows badness, then we have to hope that their deployment in fact does remain restricted, precisely so the 'target' offered to malware authors is not as tempting, allowing us to compute in peace.
http://www.tldp.org/HOWTO/Hardware-HOWTO/
Rejoice or regret the fact that those installing Linux must have a more intimate knowledge of hardware than their Windows/Mac counterparts, the fact remains for most distros that this is still true. The above guide is source of information about Linux and hardware compatibility. Here are some others:
http://hardware.redhat.com/hcl/
http://www.linux-mandrake.com/en/hardware.php3
http://www.eskimo.com/~lo/linux/hardwarelinks.html
Whatever your opinion on this [and I feel it is a Good Thing in those months it is safe to eat oysters, and a Bad Thing otherwise], Linux will never get the sort of home penetration that Windows 98 did until home versions 'hide the hardware weeny' as effectively as W98. Since I think there is a big market there, if properly developed, such a prospect bears careful consideration.
A blog which serves as an OS tweaking site with help on a wide variety of operating systems, plus some resources for imaging and Visual Basic. Searchable with news articles and downloads. Well laid-out, with an interesting block showing the IP address you are using to contact the site, the browser being used, and what language is set in the browser.
Keeping up with developments in IT has been aptly compared to trying to drink from a firehose -- the water is there, but there is just so much of it! Just keeping tabs on Linux developments is enough to leave your day hourless, so here is a Web site, linuxpipeline, which helps to channel the flow.
Covering news, trends, demonstrations, products, applications, and offering a free newsletter and a glossary, this site probably deserves bookmarking.
Incidentally, I was alerted to this site by Scott Finnie's Newsletter:
http://www.scotsnewsletter.com/subcenter/subscribe.asp?type=subscribe
which is a sufficiently valuable resource that I can plug it here without shame. It is free, but donations are solicited, and it is some measure of the exteem in which I hold it that someone as notoriously reluctant to part with a buck as I has nevertheless contributed cash to this in the past, and will do so in the future.
http://www.eweek.com/article2/0,4149,1399254,00.asp
While short-term issues tend to get short shrift in this blog, I am consistent in making exceptions, so here is a prediction on Linux for 2004. The reason why it is included is the emphasis on the fact that Microsoft is pricing itself out of the market, which is certainly true, and something to consider. But I also wonder if Microsoft is not also shutting itself out of the market with its DRM and other forms of overzealous intellectual property schemes.
As the article states, grandma really may not care about all of this, but I would think the sorts of issues repeatedly raised in this blog by someone who is in no way an anti-Microsoft zealot probably give pause to others higher up the corporate food chain. As for the home market, a turnkey Linux installation which supported e-mail, Web browsing, multimedia, and a simple office suite, while hiding the gory details, certainly would bid fair to supplant XP Home.
The revelation in the article that the single most expensive component in a low-end PC is the operating system certainly represents a change in the way things are done [and if nothing else, something of a vindication of the theories George Guilder championed over a decade ago], and perhaps one the implications of which Microsoft [and indeed everyone else] has not truly read, marked, learned, and inwardly digested.
http://www.linuxworld.com/story/38073.htm
Debate between two commentators on problems with the Open Source community, relating around issues of:
1) Developer redundancy
2) Propensity for Open Source feuds
3) Misdirected developer attention
4) "Us vs. Them" mentality
5) Microsoft as the Beast of Babylon
In fact, the commentators wind up essentially agreeing on 4) and 5) above; the spirited community input into the discussion section enlarges aspects of this debate further.
The range of Ziff-Davis/EWeek topic centers is certainly comprehensive, and each features news, reviews, opinions, and analysis, in the following categories:
Database http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3850-1
Desktop http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3853-1
Developer & Web Services http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3856-1
Enterprise Applications http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3859-1
Linux and Open Source http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3862-1
Macintosh http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3865-1
Messaging & Collaboration http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3868-1
Mobile Devices http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3871-1
Networking http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3874-1
Security http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3877-1
Storage http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3880-1
Windows http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3883-1
Wireless http://eletters.wnn.ziffdavis.com/zd1/cts?d=75-95-1-1-618817-3886-1
With this cornucopia, there should be little reason for an assignment or a research paper to lack content.
http://www.businessweek.com/technology/content/nov2003/tc20031119_9737.htm
Article suggesting that Linux's competitive position may be much stronger than financial measures would suggest. The reason: Linux is developed in a decentralized way, which helps maximize the scope and speed of innovation. In making this argument, the article is in line with some received wisdom about complex systems operations, so it certainly serves as food for thought.
http://networking.ittoolbox.com/news/dispnews.asp?i=105191
Article pointing out that as Microsoft products become incrasingly expensive and obtrusive in the application of user verification measures, Linux looks more attractive as an OS alternative. The big block to this is that people want to use the same software at home as they do at the office [and since this represents a valuable form of free self-improvement, as an office manager I would go out of my way to encoursge this], and Linux has not made a big corporate desktop entry yet.
One possibility would be the development of a critical mass of home users who could demand Linux in the workplace and have the numbers to back them up. One of the major attractions to Windows in the past was its lack of copy protection which made home users [most of whom, let's face it, are not going to pony up $500 for a copy of OFFICE] able to use the same software at home and in the office. If this conncection is broken, it puts Microsoft's corporate position at considerable risk.
Which suggests, more than ever, that Microsoft's draconian pursuit of every possible purchase may be ultimately self-defeating.
URL indexes a single floppy distribution Linux firewall designed for Ethernet connections to the Internet (cable or xDSL), allowing connection sharing. This would be a good example of a class project which students could use for home purposes as well, providing extra motivation.
http://www.newsfactor.com/perl/story/21431.html
An article looking at a range of myths concerning Linux on the desktop, suggesting that cost advantages of the OS alternative will be limited at best, because of the cost of support. Commentary and related links are provided with this story.
Another article on Open Source [which is not coterminous with Linux, but actually is its superset] myths, is avalailable here:
http://www.cio.com/archive/030104/open.html
along with sidebars and links, makes the following points:
* Low price tag is not the chief attractor -- functionality is.
* Although migration imposes costs, ultimately savings are real.
* While there is application support, it is more diffuse than in the closed-source world.
* There are no major legal barriers to Open Source adoption, despite the SCO brouhaha.
* Open Source can support mission-critical applications, and is doing so, for example, in the banking industry.
* Open Source is ready for the desktop.
The bottom line is that Open Source software makes business sense, and deserves consideration. Coming from an authoritative neutral source like CIO.com, this is a significant statement of the case.
http://certcities.com/editorial/features/story.asp?EditorialsID=64
In the Microsoft or Cisco world, it is straightforward to specify certification types and the paths among them. However, the Linux world is somewhat more complex -- this article profiles the various certification options available for Linux.
http://www.newsforge.com/os/03/10/30/0537250.shtml
Article on the 'Panther' upgrade of the Apple OS/X, describing its features and comparing it with Linux. Appended commentary gives some other information and opinion about this latest OS upgrade.
http://www.nwfusion.com/news/2003/1020linux.html
Despite the progress made in Linux development, users still want more: better tools for system management, extended driver support, enhanced security, and more commonality amongst distributions [in my opinion, one of the major FUD issues with Linux, for people who do not understand the OS well].
In addition to the main article, there is a selection of related links.
http://www.kbst.bund.de/Anlage303777/pdf_datei.pdf
A detailed, warts-n-alll look at the issues and problems involved in migrating from Windows [both as an operatings system and a support plaftorm for major service applications] to Linux. Equally valuable is the acceptance of the large degree of variables which can apply in this stuation, suggesting a multiplicity of migration paths rather than just a single conduit from A to B.
http://www.theregister.com/content/61/33292.html
When the numbers are foggy, what's a poor crystal-baller to do? Enthusiasts on both sides of the Windows-Linux divide have been forecasting the death of the opposition in the server market for some time now. This article indicates the complexity of the environment in which such claims are being made.
It all rather reminds me about the old joke about the two-car race held in the Soviet Union, in which the Chevrolet won and the Moskovitch lost: in discribing the event, Tass indicated that the Moscovitch came in second, whereas the Chevrolet was just ahead of the last-place finisher in the race!
http://informationweek.com/story/showArticle.jhtml?articleID=15202136&pgno=1
Entended article indicating that incompatibilities between Linux and Windows systems continue to be a major concern of administrators, particularly those in enterprise systems. In fact, some 33% of respondents separately segment the Linux and Windows environments. The former are the Web and similar "outward-looking" systems and the latter are the desktops.
This is not a situation which should be expected to be resolved Real Soon Now.
http://www.osnews.com/story.php?news_id=4867
An admittedly rough and ready comparison of Red Hat Linux 9 vs Sun Solaris 9, coming to the conclusion that the Red Hat product wins by a nose, and is cheaper to boot. Obviously, more examination of this comparison would be helpful, but if it were to hold up, so much the worse for Sun, since its proprietary OS is more expensive than Linux, especially if the number of processers rises over 1.
http://catb.org/~esr/writings/taoup/html/
I do not program I admit
Not even one little tiny bit
I'm not ashamed, I'm not amazed
To say it leaves me whollly dazed
You may think me quite the rogue, m'aam
But this does not mean I ever program.
OK, so I don't, but heaps and heaps of top-ranked sysadmins do so, and in some sense this is what distinguishes a real Robert E. Lee from a Martin Sheen look-alike in the networking world. Here is a free on-line book The Art Of Unix [sic] Programming by none other than the head honcho of the Free Software movement, Eric Steven Raymond.
'Nuff said.
http://itmanagement.earthweb.com/secu/article.php/3086051
http://www.cioupdate.com/trends/article.php/3086581
http://www.cioupdate.com/budgets/article.php/3087231
A three-part article [oddly enough not internally linked] about Linux in the enterprise, dealing with security, conversion, and cost issues, intended as a CEO backgrounder. This provides a useful summary for evaluating the desirablility of Linux as an enterprise platform,
This is a SANS report [as distinguished from an AVEC report, which is something else altogether] on the top 20 InterNet vulnerabilities, 10 from the Windows side of the house and 10 from UNIX/Linux. That IIS should continue to hold top position in the Windows listing should not come as a flabbergasting surprise to anyone.
http://www.theregister.com/content/56/33226.html
A common riposte to the continuint vulnerability of Windows to viruses is the claim that if other OS represented the same sized "target" as Windows, they too would be infested. This article suggests otherwise: the basic structure of OS X and Linux help make them more resistant to viruses from the get-go. There is considerable resonance, mentioned here, with the "monoculture" problem, which does not exist for either of these other OS.
If true, then this is another cost factor which needs to be evaluated when looking at the practicality of replacing Windows in an organization.
http://utilitycomputing.itworld.com/4604/030924linuxgrid/index.html
Article examining the degree to which Linux is making [and can make] contributions to grid computing, explaining the pros and cons. In addition the site indexes a number of white papers and articles relating to grid computing.
http://www.extremetech.com/print_article/0,3998,a=58657,00.asp
From my viewpoint, this article with a title originated by Walt Kelly strikes the nail solidly on its most puissant point. Linux developers need to focus on application production, in a finished manner, for desktop acceptance to be widespread.
I receive a Linux listserv from LockerGnome [itself a source of many useful items in this blog], and a continuing, multi-part series [up to #7 and still going] wherein the editor wrestles with installation problems in getting what should be a simple and basic piece of software to work in a manner recalling an XFL game in the height of winter makes this point plain. When I read this latter listserv, I sat back and wondered why anyone would want to go through such a frustrating, hair-tearing experince.
By the sounds of this article, I am not the only one so wondering.
http://www.infoworld.com/infoworld/article/03/08/29/34FElinux_1.html
Another article on the costs of switching to Linux, with a number of interesting observations, including one suggesting Linux is not always the cost solution. Points out as well that TCO needs to be related to ROI, rather than considered as a singular entity. If ROI rises faster than TCO, and there is a causal relationship, then "cheaper" is not always "better".
http://www.eweek.com/article2/0,3959,1234349,00.asp
One more look at the cost issues surrounding Linux, this time focussing on the out-of-pocket costs involved in deploying Red Hat Linux.
Some Web resources dealing with Linux commands:
Alphabetical Directory of Linux Commands
http://www.onlamp.com/linux/cmd/
LinuxCommand.org
http://gd.tuwien.ac.at/linuxcommand.org/
Linux Commands and Shell Commands Libraries
http://gd.tuwien.ac.at/linuxcommand.org/
Some Useful Linux Commands
http://www.er.uqam.ca/nobel/r10735/unixcomm.html
An HCL for Linux site, broken out by device type. When a specific product is selected, buying information and review links are supplied, as well as links to Linux-friendly PC purchase. This can save a lot of heartache for the do-it-yourselfer.
If I had just had this before buying that 32 MB PCI grapics card...
http://entmag.com/news/article.asp?EditorialsID=5910
Rivalling the argument about how many VC angels can dance on the head of a pin has been whether Linux or Windows has a lower TCO -- this article comes quite close to saying "both". Besides making some sensible analytical remarks about costs and their estimation, it also [along with commentaries in response] demonstrates that such issues as IP and Product Activation represent real factors to be considered above just cost issues.
This won't settle this debate, but it can make the discussion more informed.
http://entmag.com/news/article.asp?EditorialsID=5911
At the beginning of WWI, most of the troops involved said "home by Christmas" -- it did not quite work out that way. Nor will the war among Linux, UNIX, and Window be resolved by a knock-out blow any time in the immediate future.
The difference being, this war may ultimately benefit all computer users, something which assuredly cannot be claimed for "the War to End All Wars".
http://www.cioinsight.com/article2/0,3959,1192684,00.asp
Examines the problems Linux has experienced with negative competition, and points out that while the SCO suit is high-visibility, the shark in sheep's clothing is really Microsoft. Suggests as well that should Microsoft win out in this struggle, the long term consequences would make eating Grandma look like a picnic.
http://www.technologyreview.com/articles/wo_lefkowitz072503.asp
Article recording an interview with Robert Lefkowitz who makes the point that open source software may be attractive for business not because of low cause of acquisition, but because the nature of the open source support community remarkably lowers TCO. Given an acceptable level of technical performance, this may indeed be a devastating weapon against the closed software paradigm represented by Microsoft.
http://cda.mrs.umn.edu/~mine0057/fs.pdf
A short paper ["New Advances in the Filesystem Space"] on the drawbacks of current file systems, particularly as these relate to POSIX, and what advances are being made in this area. It is worth keeping an eye on this area for new metaphors underlying a changed way of doing things.
http://www.osafoundation.org/desktop-linux-overview.pdf
A downloadable .PDF file taking a look at how Liux has progressed as a desktop competitor to Windows. The conclusion is reasonable and not particularly startling: Lunux is increasingly suitable as a desktop OS, and will make inroads on the Microsoft monopoly, but is not going to produce any sort of revolution in this area over the short term.
http://linuxshop.ru/linuxbegin/win-lin-soft-en/
A somewhat rough-around-the-edges resource grammatically, but nevertheless useful, source for Linux software which is functionally compatible with the Windows equivalent [think "Office"]. In addition to those programs which can work on more or less a 1:1 replacement basis, Windows applications which can run under WINE are also covered.
http://saintaardvarkthecarpeted.com/linux_vs_sco_matrix.html
The welter of claims and counterclaims made by participants and guilty bystanders in the SCO/Linux case needs some straightening out, and this matrix not only provides a list of alternatives and possibilities, but also an outcome set [which, oddly enough, is the same for all cases -- no effect on Linux]. In addition, links to supporting material for the citations in each matrix cell are provided.
http://www.acm.uiuc.edu/sigmil/RevEng/
A work-in-progress, admitted warts and all, representing an online book covering reverse engineering in both Linux and Windows, built along general approaches. One of the follies of current IP litigation is to outlaw reverse engineering, which can be a most valuable form of education and discovery for those involved in computer programming and IT applications. The long-term pernicious effects of such a narrow stance can only be expressed by forebodings, but the learned opinion with which I am familiar suggests strongly this IP stance is Not A Good Thing [nor it is, contrariwise, A Good King, either].
So here, perhaps, are Things Man Was Not Meant To Know.....
http://eletters.wnn.ziffdavis.com/zd/cts?d=75-1-1-1-618817-13-1
Linux kernel v. 2.6 is still undergoing the throes of birth, but promises to add a number of desirable improvements. Here is an article about the state of play in the kernel development.
AMD and Intel Ship 64-bit Chips: http://eletters.wnn.ziffdavis.com/zd/cts?d=75-28-1-1-618817-1171-1
Red Hat, Dell and Others Support Intel's 64-bit Madison: http://eletters.wnn.ziffdavis.com/zd/cts?d=75-28-1-1-618817-1171-1
Does the G5 Really Matter: http://eletters.wnn.ziffdavis.com/zd/cts?d=75-28-1-1-618817-1174-1
The whole bandwagon towards 64-bit computing seems to have sprouted afterburners. In part this results from the ease with which Linux/UNIX variants can support 64-bit processors [they've been doing it for years], and Apple's venture results from the fact that OS X has BSD UNIX roots. Given all that, Intel simply could not afford to hold back from the fray, and the more progress and support that is given to hardware and software developments relating to this, the sooner we all will have supercomputers on our desktops.
Then the question to answer will be: "What do we need these for?" -- which, when you think about it, is really a pleasant position in which to be.
A neatly arranged searchable message board site covering tips for Windows, Linux, and Apple OS, plus miscellaneous issues. Comprehensive with archives, downloads, FAQ, fora, news with archive, a glossary, reviews, surveys, tips, and web links, this looks to be a useful site for a wide variety of computer users.
http://www.nwfusion.com/reviews/2003/0623linuxreview.html
With all the distributions of Linux out there, which one is best? Foolish question, grasshopper! It all depends on what you want to use it for. But for enterprises, the scope is somewhat narrowed, so here is an article which compares Red Hat with several United Linux entries [with the main focus on SuSE] and also gives some glimpses as to how all this stacks up against Windows 2003 Enterprise Edition.
To save some suspense, in case time is too pressing to actually view the article, it was Red Hat by a nose.
http://www.forbes.com/technology/2003/06/16/cx_ld_0616linux.html
Despite a somewhat negative title ["The Limitations of Linux"] this short article looks at some of the pros and cons of the Linux OS for the corporate environment today. The conclusion: Linux has a place, but that place is not all-encompassing.
http://www.eweek.com/article2/0,3959,1121367,00.asp
I can remember, when I first beheld the IBM S360 some 35 years ago, yearning to have one of my own -- little did I dream. Now, of course, reasonably powerful computers are available for everyone, but just as you can never be too rich or too thin, your coumputer can never be too powerful. The popular image of the "big machine" in film and fiction was just that -- a machine housed in massive caverns sized on a titanic scale [for example, Colossus in The Forbin Project.].
And there have been, no doubt, many impressive and gigantic supercomputers, ranging from Crays to ASCII White. Yet again, as has been true so many times in IT's past, at least some of the palm goes to the lowly and humble, as it turns out that it is quite cheap to arrange clusters of Linux machines which are not just resepctable, but in the upper echelons of the supercomputer ranks. There is no reason why such developments cannot continue.
All this suggests to the extent that we allow ourselves to be computationally challenged by raw power requirements, we may well be reading off the wrong disk drive.
While the news that the Linux Router Project (which provided a single-floppy routing capability in Linux) has been terminated by its author in a state of some bitterness itself represents a sad outcome, some wider reflections are more troubling. The question being: how can Linux developers support themselves? The for-profit model of proprietary software answers this question for its developers without doubt.
True, some people make a living at Linux, and some people don't want to make a living at it, but just want the make the OS live. Perhaps that will be enough, but there are times, like this, when optimism is a bit harder to come by than it normally would.
http://news.com.com/2100-1016_3-1017180.html?tag=lh
Version 2.4.21 corrects some errors and adds support for new hardware, most notably the AMD 64-bit Opteron processor. This might just steal a march on Wintel Itanium, since the AMD is backwards compatible with 32-bit applications. I am sure someday we will look back on 32-bit computing as being limited and quaint, just as we regard 16-bit computing now.
http://http://nefiles.uiuc.edu/rhasan/linux
The tendency in information technology is to look forwards, not backwards. But remember that "the future of the present is in the past". What is now depends crucially upon what was then, and knowing this is helpful in all sorts of ways, not least because of the UNIX tradition of maintaining traditions, sometimes just for tradition's sake. So here is an illustrated and documented history of Lunux, showing how far the OS concept has come in a decade.
http://arxiv.org/abs/cs.DC/0306058
One of the joys of Linux, I am told, is how you can gang a whole lot of little computers to create a sort of Compzilla. I suspect the folks at CERN have more than a little bit of experience with this; the abstract contains a link to the full paper. The noise of 1,000 fans all whirring together at once must be something to be heard to be believed.
http://cl.com.com/Click?q=88-ec5PQQKmi5OsSUzRjhGvQVQSUDrR
A useful summary by an interested participant, Hewlett-Packard, on the "Linux Strategies And Solutions" offered by five major suppliers of Linux systems. Oddly enough, Microsoft is not on this list of names....
The OS files is one of those sites, well worth bookmarking, which takes your breath (and time) away. From W95 to "Blackcomb", MS-DOS to FreeDOS, Linux in a variety of distros, BeOS to Netware, there is something here on just about every operating systems running on a PC. In addition to reading about it, there is a links page to free OS that you can download and play with.
Nothing quite like whiling the time away twiddling with an obsure OS....
My attention was grabbed by Brad Gadberry's comment [no mean IT commentator, he] about the relative difficulties of using Windows and Linux, noting that you can have bad Windows experiences also, and that emphasizing the minor problems in Linux makes the OS sound like risky business.
My first response was "Well, yes" -- in the sense that Windows is easier to use than Linux, so that when something goes wrong with it, a relatively inexperienced person can effect a repair more easily than with Linux [to be clear, I am not talking Win 9x here -- it is simply unfair to compare such a limited OS with Linux]. So in that sense, the perception of "greater risk" in Lunux is not necessarily wrong -- my Ford and my Ferrari may both suffer from a problem with their brake calipers, but reparing the latter is liable to be more fraught with angst.
But when I thought about this further, I asked myself: "What does it mean to 'use Linux'?". This is by no means as silly a question as it sounds, because "using Linux" can mean one of (at least) two things:
1) To interact with and employ the capacities of the operating system through the command line interface; or
2) To interact with and employ the capacities of the operating system through the graphical interface.
Now Windows, of course, is primarily graphic [though many useful command-line elements like scripting are possible within it], so when we talk about "using Windows", there is usually no doubt about what we mean. I would argue with equal force that in terms of complexity and power, Windows and Linux are roughly similar [i.e. this difference is not even a single order of magnitude]. Currently, however, the power of Linux is most readily tapped through the CLI, and this is exactly the interface which is most intimidating to most computer users, so that even executing a trivial correction may appear to be non-trivial.
There is, I would argue, some need for precision when talking about "using" operating systems, just to make sure we are not comparing hawks with handsaws.
http://cl.com.com/Click?q=04-k9W0IFQjFfCUUR1_QHNpWzpOYWeR
When a company gets sufficiently worried about the competition to actually lower or eliminate profit margins in order to land the sale, the company is taking its competition seriously. Whether you view this as an anticompetitve tactic or a new-found appreciation of how Microsoft endangered by the Linux legions, the fact that statements like this can be made is an eloquent indicator of how things are changing in the OS arena.
http://cl.com.com/Click?q=04-k9W0IFQjFfCUUR1_QHNpWzpOYWeR
There is a lot of interest in saving IT dollars by moving from a proprietary system to one which is open and low-cost. Naturally, Red Hat is only too happy to provide information on how to do this: "Solaris to Linux Migration: An Introduction".
Note that registration is required to view this paper, but this site is sufficiently useful that I find it worthwhile to be registered.
Despite the title, this includes Linux tips as well, contiributed by the site users themselves. These range in profundity from basic to advanced, and can be viewed storted into categories.
http://www.linuxworld.com/site-stories/2003/0401.tsu_p.html
Another viewpoint in the "is Linux ready for the office desktop sweepstakes" brouhaha. I find myself in entire agreement with the author -- the geek viewpoint that like spinach, installation pain is fundamentally good for you may be right -- but as far as gaining converts for the desktop, it is dead right -- like the man going on a green light when an out-of-control truck is coming through the red. What most IT people [myself included, much of the time] fail to understand is that people want to do things *with* their computrers, not *to* them.
That's a major culture gap which has to be crossed daily, and lots of Linux mavens are having real trouble with it. Be assured, Bill Gates & Co. have no trouble with it at all, which may go some way in explaining relative market shares for the desktop.
http://eletters.wnn.ziffdavis.com/zd/cts?d=75-18-1-1-618817-502-1
http://eletters.wnn.ziffdavis.com/zd/cts?d=75-18-1-1-618817-505-1
http://eletters.wnn.ziffdavis.com/zd/cts?d=75-18-1-1-618817-508-1
Contradictory positions on the future of Linux are ever with us, like mildew beneath the bedspread, but the latest SCO shennanigans have brought the pot to an intense boil. John Dvorak [who is, on IT matters, always right when he is not wrong] bellows gloom and doom [the comments section is well-worth following up here] in the first of the above entries.
In the second and third articles, continued improvement in the Linux desktop is profiled, suggesting conttinued wind beneath the penguin's wings. For my own part, there are not words fit for a family newsletter [which this definitely is *not*] to express my opinion of it. I expect IBM to win big on this one, with SCO becoming just the memory of another teapot tempest, and Linux moving from strength to strength. I say this based on the evidence, since I am by no means a Linux enthusiast.