|
| Thu, Aug 28th | home | browse | articles | contact | chat | submit | faq | newsletter | about | stats | scoop | 23:14 UTC |
|
login « register « recover password « |
| [Article] | add comment | [Article] |
Ed Sawicki (a professional instructor/lecturer who has used NetWare and NT since their beginnings and Unix since 1982) comments on the state of Open Source software in the corporate world of NT and NetWare. He's also used Coherent, QNX, Linux and *BSD. An assembler is his favorite programming tool. He's an open source advocate and a frequent critic of the existing computer establishment. He is the president of the Accelerated Learning Center. For his editorial, hit the details link. Links: Accelerated Learning Center Copyright notice: All reader-contributed material on freshmeat.net is the property and responsibility of its author; for reprint rights, please contact the author directly. PROTECTING YOUR CAREERby Ed Sawicki - ALCFrom time to time, people ask me if there is objective information that compares NT, Linux and NetWare. The requests usually come from people in organizations that are converting from one of the operating systems to the other and want the information for one of two purposes.
Most of the information on the subject is published by the vendors, market research firms or the magazine publishers. Clearly, information supplied by the vendors is suspect. A Linux or NetWare enthusiast views any words on the subject from Microsoft as an outright lie. NT enthusiasts feel the same about Novell but are often left wondering about Linux since there is no one vendor to distrust. The objectivity and competence of research firms and magazines has been called into question recently in light of the "benchmark wars" that have produced literally unbelievable results. Typically, Microsoft makes a claim on their web site which is usually backed up by a report from some market research firm or independent testing organization. When they claim that NT outperforms NetWare, Novell cries foul and responds with counter-claims backed up by other marketing organizations that have pro-Novell inclinations. When they claim that NT outperforms Linux, the Linux community responds but sans support from research firms. The trade press adds to the confusion by publishing articles that are highly polarized. Some articles are quite good and it's clear that the author has a good grasp of the issues. Unfortunately, there is far more written by people who are either clueless or have a definite political agenda. Some authors simply follow the herd - praising or criticizing a product or vendor depending on what's fashionable at the time. This has led to the following not-so-funny jokes.
Q:What's the best way to increase your productivity?
Q:What's the best way to increase your company's overall
productivity?
Without objective information, we can't expect management to make decisions based on technical merit. So, they're left with marketing hype. The best marketeer wins their business. Perception becomes the reality. HOW DOES THIS IMPACT YOU?I remember the good old days where the computer folks in a company were the experts. Management respected their opinions and relied on them to set the corporate computing direction. They could spend time evaluating technologies and products to determine whether and how their company could exploit them. They felt important and job satisfaction was usually high.These days are over for many companies. Upper management now makes important decisions without help from their own computer literate people. It's reminicent of the Vietnam war where the Johnson/McNamara White House made the day-to-day decisions rather than the on-site commanders. Like the Vietnam soldiers and airmen, today's computer folks are expected to implement whatever management decides whether it's the best course of action or not. The question is whether this state of affairs is good for you and your career. If you enjoy working with technology, chances are good that you won't be happy working with stuff that you think is technically inferior. But hey, it's your job and many people aren't happy with their jobs. That's why we call it "work". But your career is a different story. Happy or not, you need to know that you'll have a job in the future. If your employer has you working on things that you think will not be marketable skills in the future, you're putting your career at risk. Far fewer people nowadays are in cradle-to-grave jobs. We must protect our futures by developing skills that our future employers will need. But exactly how do you determine, for example, what operating systems will be popular in the future? If products become popular for reasons other than technical merit, then there's no way to know. The marketplace is too fickle to be predictable. Unless you can keep current on all possible contenders, you're gambling with your career. If, for example, you now spend all of your time managing a NT-based network with an Exchange-based email system, what do you do when the company decides that converting servers to Linux is a better deal? Given the popularity of Linux these days, this example is a particularly plausable one. THE SOLUTIONThe solution is for products to succeed or fail in the marketplace based on their technical merit. Then you can keep your career on track with less risk of choosing the wrong horse. That is, if you can recognize technical merit when you see it and therein lies the problem. There are too many examples of the excellent losing to the mediocre. 100VG losing to Fast Ethernet and Banyan VINES losing to virtually everything else are good examples.What's needed is a source of technical information that is free from political bias that technical folks can use to stay on top of the details and the facts. Information that is easy to grasp for those, such as upper management, who don't want to know or can't comprehend the gory technical details. While this seems like an impossible goal, things have happened recently that may now make this possible. OPEN SOURCE AND PROJECTSThe Open Source Software revolution (of which Linux is the best known example) has created an attitude that says "If we can't trust vendors to act in our best interest and build software products that improve our productivity and lower our costs, then we'll build the software ourselves." The result of this attitude is free software that is often better than many or most vendor products. Samba and Apache are excellent examples.The stunning success of Open Source Software has inspired people to take this attitude beyond software development. Many people have started "projects" in other areas. Projects have been started that create standards for both software and hardware. These standards increase compatibility among vendor products and reduce our costs. As grassroots support for these standards takes hold, vendors are compelled to follow them or alienate a growing portion of their customer base. I, for example, will no longer buy a hardware device that does not have support for Windows and Linux and, where appropriate, for NetWare because I use all three operating systems. This morning I learned that the vendor of a video chip set refuses to release the technical details to the Linux community. This means that Linux video drivers must see this card as a lower performance generic video card. I'll no longer buy video cards or computers based on the chip set (or that vendor) and I'll encourage others to do the same. If enough people participate in the boycott, we'll force the vendors to have "open hardware" standards. See the power of projects? It's time for a project that evaluates network operating systems and publishes objective comparisons so we don't have to rely on the vendors and the the press. People can use these comparisons to help their careers and help their companies decide on their future direction. Members of the press can use the comparisons to get the story right more of the time. NTIQUENTique (the NT critique project) documents the differences between NT, Linux and NetWare. You can get more detail about this project by visiting the NTique web site.[Comments are disabled]
[»]
Drivers & Microsoft To the first two comments, I would like to say the following:
[»]
re: sheeeesh The author did not ask for suppport on every platform but rather asked
for specs onthe hardware. The Linux community is one of give me the
information and well write it ourself. We don't need support form the
hardware vendor, we just need to be able to write our own drivers. If the
hardware vendors release there specs to the public then they do not have to
write drivers for any platform, rather the platform maker can write the
drivers, which shoudl be better. After all who knows the platform better
then the platform programmers?
[»]
Gut Instinct Like the early pioneers of sea travel like Magellan and Columbus, most of us should consider ourselves cutting a path towards a fresh digital future. Like our predecessors before us, we must acquire insticts about the technology that is being literally thrown en masse` at us. You have the Whizzer corporation creating their super-dooper-whiz-o-matic slicer/dicer programs. While Spacely Sprockets displays their brand new all-in-one-der rack mounted router/csu/firewall/vpn gadget, we must gain a 'feel' for the industry and find out what's hot, and what's not as early as possible. Part of an early explorer's success was gauged on how well he read the tides, and the winds. As someone shared earlier, the winds of change are flowing in this digital highway. Our success depends on our ability to read these changes and adapt our strategy as practically as possible. Refrain from going out and spending company dollars on those wizz-o-matic's and start researching top-contenders with the companies 'big picture' in mind. Some technologists have to worry about saving their career. Not only through the development of new avenues, but also in the amount of manpower and resources are put into sole-sourcing tasks to the "whizz-o-matic". I have personally made mistakes in believing that this new technology is the 'sure-fire' winner only to find that a tremendous amount of effort was wasted. Adaptation is the key. Keep your head out of the clouds, and your feet on the ground. When you feel the winds shifting directions, go. This can only come with experience and perseverence. Abraham Lincoln failed his first business venture at the age of 22. He ran for city legislature at 23 and failed. He failed again at business at the age of 24. His sweetheart died when he was 26. After two more decades of failures, he eventually became the President of the United States. We must adopt the same principle of adaptation, and perseverence. Protecting your career is your second priority. Being a good servant is your first. There's my two cents. Aloha
[»]
Protecting your career It's Noam Chomsky time! This "I'm reasonable, you're reasonable, let's be reasonable" approach suddenly brings tobacco industry strategies to mind. Microsoft's increasingly shrill propaganda campaign against Linux tells us they see the writing on the wall and now their "agent of reasonable influence," Sawicki, is saying, "Hey, there's room here to co-exist-- let's just decide what it is that *you're* good at, and what *we're* good at, and we won't really have to bother each other. Isn't that reasonable?" The reality extends beyond technical merit, although whatever relative Linux shortcomings that do show themselves are typically corrected in short order. The reality is that Bill Gates and his minions declared war on mankind and unblinkingly used scorched earth tactics for a decade to wreck our industry and destroy consumer choice. Screw Bill Gates. I won't use NT under *any* circumstances. The sun is seting on the Microsoft era. Hitler should have chosen to be "reasonable" *before* he got to Stalingrad. No thanks, Sawicki.
[»]
Lack of originality and other remarks The "About" page on the NTique web site uses information from my article - not the other way around. One of the goals of the article was to let the world know about the NTique Project. I'm one of the project's founders. Like any good programming project, reuse of code (or text) is not interpreted as a lack of originality. Why the fuss? I don't think I said that Linux relies on Open Source projects. However, commercial or not, open source software has significant benefits over closed source software. One only has to look at the recent uproar over Microsoft's NT, CryptoAPI and the NSA key affair to see the dangers of closed souce. Yes, I agree completely that management's reliance on credentials rather than on merit is a problem. I've been a critic of Novell's certification program from the start. However, this is not the problem I wrote about. Management often ignores the advice of their technical folks regardless of whether they have the credentials or not. Ed
[»]
Lack of originality or what? The article made me want know more about ntique.org. Guess what I found there? A large portion of the article was taken word for word from the ntique.org about page. Hmmmmmmmmmmmmmmm?
[»]
Binge and Purge It seems that's what we have to do if we work in this biz. I'm just about ready to yank my 6 years of VMS from the resume.....
[»]
This applies directly to the This article backs up the popular stories of not enough people being
skilled in software. The problem isn't a shortage of people or technology
but the increasing reliance of managers on paper and credentials and the
complete disregard of merit. Managers are more and more refusing to look
at anything that doesn't have a certificate or a proper degree or flashy
marketing.
[»]
Linux is coming of age A flaw in the argument is that Linux relies on Open Source projects. The
commercial scene of Linux is the most exciting. Getting more and more
companies to port their software over to Linux will ensure that the
comments shared in the article are obsolete.
[»]
Feh The above commenter clearly misses the main point of the article, which is how to avoid becoming obsolete. This IS a serious issue if you make your living with high-technology, since the way the winds blow does change quickly and often. (For example, six years ago, getting your CNA or CNE was all the rage, and generally brought with it the impression that you're set for life when you've got it. How well are those IPX gurus faring now in today's Internet-obsessive job market?) ...and as far as his opinion of driver support is concerned, I really don't care if the manufacturer of the hardware won't support every single operating system under the sun. I don't think they have to. They can put out drivers for the platforms their software developers are comfortable with to increase the ease of use with that platform, or they can hire people to make drivers for the platforms they want to cozy up to where they don't have the in-house expertise. I am well aware that these costs are a serious issue in the total cost of the final product, but responding to simple inquiries with the information required for a third-party to develop their own drivers costs about the same as any other fifteen to twenty sheets of fax time. Saying "no" to someone who is willing to do the work at no cost or obligation to the company is usually going to be a flawed response. When the cost of releasing the information is so miniscule, compared to the relative gain of having a product functioning under yet another (even moderately popular) platform and increasing the utitilty value of the product, then refusing to release the information to interested developers is a bad buisness move and either reflects a general myopia of corporate management, or bad buisness planning that got them into a situation where they were prevented from releasing the information by an NDA. Either way, it's one rather clear sign that a vendor is not making the right decisions (and increasing the chances that they, too, could become obsolete rather soon).
[»]
seeesh Thank you for you another over biased essay. If a vendor decieds not to support hardware under every OS in the world they dont have to and many wont. The cost of supporting more than one OS is extreme....also, if you dont like it code your own drivers! I will continue to use the hardware I want even if its only supported by one OS.
|