Wouldn't that be the best place to dissect this issue?
Better still, wouldn't this issue be wholly more appropriate to Groklaw than the current bent of politics and advocacy (ie, ODT vs OXML) that seems to be creeping in there? Given the paper legalese that's already been flying around (as listed on the JMRI site) -- not to mention the significance of the ruling in attacking a popular FOSS license -- I know I could benefit from this dispute being given the 'Groklaw treatment'.
So far, it offers not even a mention in passing....
No previous Linux certification is required to become a Novell CLE, however a candidate will need to have the knowledge that is necessary to pass the Linux Professional Institute Certification Level 1 (LPIC 1).
Two groups, in their own ways, are working to ensure that WSIS encourages the promotion of open source amongst its participating countries.
The Linux Professional Institute and the
Free Software Foundation are two of the many hundreds of non-governmental organizations which have received official status at the Summit. (Here is
Part 1 and Part 2 of the complete list.)
LPI will tentatively be holding a number of events at the WSIS conference in December, including an open source workshop and a certification exam lab; it is also our intention to put a Linux "live" CD in the hands of every WSIS delegate. We will have at least six people at the conference, working to ensure that the delegations are capable of overcoming the anti-open-source FUD which is no doubt going on.
Linuxgruven happens to be the employer of Kara Pritchard, Director of Exam Development for
the Linux Professional Institute. Feel free to check it out yourself on either LPI's or Linuxgruven's web sites.
While she is actively involved in LPI development and holds an RHCE, I don't believe that Kara has taken any Sair exams.
I also note that Linuxgruven has been very supportive of LPI efforts, and is recognized as a
bronze-level sponsor of LPI.
It always helps to read the actual article itself. Nowhere did I postulate or infer forking of DEs (beyond what's been already done) and I certainly said nothing here about forking OSs. I've already talked about forking in a previous column.
In fact, I said very little of all about the actual code of GNOME or KDE, and much about the perceived PR efforts of the two groups. Of course choice is good -- I've also covered that ground previously.
But this isn't about the ability to choose -- the choice of open source UIs existed before the GNOME Foundation or KDE League were ever envisioned. This is all about misguided PR, and the diverting of public attention away from the software itself and towards each group's desire to be a "standard" and backing by big-company endorsement.
If all they want to do is advance Linux as a desktop, both KDE and GNOME (and their supporters) would do better channeling publicity efforts through a neutral umbrella group such as Linux International. By creating their own PR efforts, independent of each other and the general Linux community, they call attention to the fragmentation rather than confronting common foes (Microsoft and Apple).
If you want to disagree with that point, go ahead. But be very clear that I'm not knocking the existence of multiple open source desktop environments, merely the undesired consequences of ill-founded PR initiatives.
OK, the points about the usability problems with Beowulf and commercial software such as databases and web servers are well taken.
But Beowulf isn't the only clustering technology for Linux. How do the business-targeted clustering products from TurboLinux or SGI (or others) stack up compared to the UnixWare clustering/SMP approach?
Apologies if my comment about Lesstif being a compatibility tool was taken as a putdown. It most certainly wasn't.
I was just making the point that, at least in the comments I had received while reseraching the piece, I'd found that new GUI apps were more likely being developed using Qt or Gtk. Lesstif is still very valuable, but mainly by those who have already invested resources into programming for the Motif API.
Your point is valid, and your fears are justified by the experience with other cert programs.
I can offer these reasons why and how LPI is different from MCSE and other certification programs:
LPI started as a grassroots organization, its history goes back to active email discussions from within the Linux user community that grew into a formal body
LPI is incorporated as a non-profit body; legally we are not allowed to make a profit
We are vendor-neutral, and have a number of Linux companies as sponsors (with more coming) and even more on our Advisory Council. The intention is that no one vendor can dominate the LPI agenda.
We are trying as much as possible to keep costs down. We have sought to minimize the number of exams required to demonstrate the required proficiency (MCSE's first level requires six or seven exams; LPI's requires three)
As another example of trying to keep costs down, LPI will not be "expiring" certificates the way other programs do. While we all know that older certifications will be less usable, we will not revoke them
LPI is not in the training or book business. How you get your skill is up to you. Take courses, read a book, or go into a cave with a Linux box for a month. All LPI cares about is that you know your stuff when you take the test.
The exams will be challenging. It will be quite difficult for somebody who doesn't know the basics to pass the first level
We NEVER advocate leaning on certification -- LPI's or anyone else's -- as a substitute for good interviews or reference checking etc. A certification program is helpful but it's not the complete answer by far.
Asked for and received. Use of the word "Linux" in the name "Linux Professional Institute" has explicit permission of Linux International, which administers the trademark on Linus' behalf.
Perhaps it wasn't specifically designed to attract Slashdot's attention, but it's specifically designed to annoy the kind of people who read Slashdot.
No such thing. Mine was an opinion (and clearly marked as "Commentary") that was expressed, based on observations I've made both amongst/. readers and amongst new users. I give quiet a few "Intro to Linux" lectures, and I believe I have a fairly good idea where new users are coming from. They're most certainly on a different wavelength than this bunch. Not better, not worse, just different.
The reaction I've seen to my column here does nothing to change the perception that an awful lot of people just don't get it. Some come across as "I get it, so what?", and that's OK. But some of the nastier replies I've seen (including some really precious ones in private email) have a common theme: how dare I suggest that/. readers are losing touch!
Looks from here like I've touched a nerve. Totally unintentional, but fascinating nonetheless.
I don't care if Evan is a "suprerior techno-geek", but if he can't keep from lying, misrepresenting, and otherwise pressing false conceptions through his articles, I'm going to continue to trash him.:)
Do your worst. I stand behind what I said, and incredibly little has been advanced here that refutes much of what I've said (beyond the name calling).
I was legitimately called into question on issue of fact and one issue of timing;
Mandrake was born earlier than the beginning of this year, as I had said. My mistake; I'd tried to look up some history of the distro and hadn't been able to come up with much, on their website or elsewhere.
It's been pointed out by many that Red Hat 6.1 was in release by the time my piece was published. OK, so ZD doesn't turn around articles as fast as some would like -- guilty as charged. But the point I was trying to make -- that Mandrake is trying to show it's not just a RH follower by coming out with 6.1 before RH does -- still stands.
Are the above "gross errors"? Not at all. Are any of the other points factually wrong? Nothing else has been pointed out.
It's all-too-easy around these parts to disagree with an opinion by calling it a lie. I write opinion pieces by design, yet I try to do research to make sure I can back up what I say. In this case I'd interviewed the presidents of MandrakeSoft, Macmillan, and Red Hat before writing. There's also a lot of background to what I wrote that never made it into print.
As for "false conceptions", you'll have to explain what that means. If you think I've achieved incorrect or inaccurate perceptions based on facts at hand, then we can debate that. But you haven't done anywhere near the very difficult task of proving that an opinion is false.
So go ahead, be very concerned with truth. But also keep in mind that my piece says "Commentary" at the top, that I have done a fair bit of homework before preparing it, and I'll be more than happy to rethink things if someone offers a different intepretation of the same premises.
Aimlessly throwing the word "lie" around accomplishes little of value, and certainly does little to disprove my conclusions. Did I lie, as some have suggested, because I deliberately did a/. search on "Macmillan" rather than "Mandrake"? Not one shred, especially since the Macmillan box of Mandrake has some significant non-technical differences with MandrakeSoft's own package -- bundled-in Linuxcare tech support, for one thing. You may not agree with my opinion that this is a significant distinction, but it ain't no lie.
Anyhow, I still stand behind what I said, minor flaws above notwithstanding. Want to disagree? Fine! But indiscriminant shouting of "liar!" gets one nowhere...
I've had a lot of pieces picked up by/. without mentioning it, that generated a lot more response. I didn't ask for this one to be put in, and really don't care that it's here.
If the/. editors think that a mere mention of their site in an article makes something immediately newsworthy, that's ego at work rather than judgement -- the same ego that seems to think that a troll accusation is "insightful".
In Canada, the communications regulatory body is requiring cable companies to allow third-party ISPs access to its bandwidth. At least here, soon one will be able to enjoy cable bandwidth but not be tied to the cable company as their only choice of ISP.
So let the merger happen, for all I care. So long as cablemodem Internet users can opt out of AOL/@Home, they can do whatever they want...
And Sun didn't even make lxrun, which was invented by an SCO employee. Lxrun was running for many months on OpenServer and UnixWare before Sun took advantage of its open source license to port it to SolarisX86.
I appreciate the new info, but there are still a few issues about Wabi I find disturbing:
Sun released 2.2E for Solaris, but refused to bring those important bugfixes to any other port (including Linux), where it stopped at 2.2D;
When it became clear that selling and maintaining Wabi was more trouble than it was worth, Sun could have and should have open sourced the parts that it owned, to at least provide a safety net for the newly-orphaned users. Instead, it prevented licensees from even doing their own bugfixes.
I understand what you're saying, and grant that the dropping of Wabi wasn't done specifically to hurt Linux users -- but that's indeed what the consequences were. Subsequent actions could have lessened the blow (such as allowing users and/or sub-licensees to maintain the orphaned product), but Sun chose the path of greatest shafting instead.
Part of my original point was that StarOffice is a strategic product for Sun, much like Wabi was. As Linux users got shafted then by the way Sun treated Wabi (deliberately or not), we must watch that the scenario doesn't repeat itself with StarOffice.
I never said that Wabi was cancelled because of Linux. I did say that Sun's move to mothball it, rather than open source a product that was no longer selling, indicates it's more interested in strategically pushing its own plans than it is in helping the open source community.
Contrast Sun's actions wrt Wabi with the decision to LGPL the Willows TWIN software when the company decided it wasn't going to sell. Now TWIN is a part of helping Wine be ready for prime time.
Wabi could have also helped Wine, but Sun chose to kill the software and orphan its users because it wanted to strategically push Java instead of emulation.
Listen to Sun's announcement and hear for yourself. Putting thin clients into the home market is exactly one of Sun's goals with its StarOffice/StarPortal strategy.
Sun's future would have people depending on mega-ISPs (using Solaris servers, of course) that would store your files and serve your apps for you. Think WebTV with decent video.
Sun's intent is quite clear in this regard. The main point of my argument, which will be concluded in a second column appearing next week, is that neither Sun's nor Microsoft's world-view is right for everyone. Thin clients are great in some situations, horrible in others. Linux is flexible enough to serve as an excellent thin-client, fat-client, or server. Neither of the commercial approaches is best in all cases, yet neither does much to acommodate models outside of its own.
Linux is a threat to both Microsoft and Sun because it offers the flexibility to serve either model -- or some hybrid of the two -- very well. Linux isn't trying to push the user into one world-view or another of the way computing should be done.
I don't think it's reasonable to paint all ZDNet authors as Linux hostile because of Berst, any more than calling all of IDG that way because of Metcalfe.
It's not uncommon for publishers to have, on the same payroll, multiple commentators with very differing opinions.
Not all the readers out there are smart enough to understand the message intended by ignoring someone who spews. If someone has an audience that buys his arguments, and no rebuttal comes, this silence could be interpreted as an inability to rationally rebut the original spew.
I stand behind my decision to confront this stuff rather than pretend it didn't exist. Yes, it draws more attention to the original spew, but that attention comes in the context of ridicule rather than authoritative opinion. And it was important (to me, at least) to call attention to the kind of spew that people will start encountering when recommending Linux to their employers or clients.
There are times when you let a ranter flail away in silence. Other times -- especially when the speaker has an audience -- that it's necessary to sound the BS alarm before anyone gets sucked in by sophomric namecalling masquerading as cleverness.
PS: This is not a pissing match between ZD and IDG, it's me alone calling attention to something that happened to come from InfoWorld. I have a lot of respect for Petreley and other writers at IDG.
Slashdot Mirror
Wouldn't that be the best place to dissect this issue?
Better still, wouldn't this issue be wholly more appropriate to Groklaw than the current bent of politics and advocacy (ie, ODT vs OXML) that seems to be creeping in there? Given the paper legalese that's already been flying around (as listed on the JMRI site) -- not to mention the significance of the ruling in attacking a popular FOSS license -- I know I could benefit from this dispute being given the 'Groklaw treatment'.
So far, it offers not even a mention in passing....
No previous Linux certification is required to become a Novell CLE, however a candidate will need to have the knowledge that is necessary to pass the Linux Professional Institute Certification Level 1 (LPIC 1).
LPI will tentatively be holding a number of events at the WSIS conference in December, including an open source workshop and a certification exam lab; it is also our intention to put a Linux "live" CD in the hands of every WSIS delegate. We will have at least six people at the conference, working to ensure that the delegations are capable of overcoming the anti-open-source FUD which is no doubt going on.
To that end, LPI has submitted a commentary on the WSIS activities, now part of the official WSIS documentation, that is stirring some interest. Anyone who is interested in helping LPI's efforts at WSIS is invited to subscribe to the LPI@WSIS mailing list.
The FSF is participating through the WSIS Working Group on Patents, Copyrights and Trademarks; RMS is on the group's steering committee and Georg Greve of FSF Europe is one of the co-ordinators.
A little dose of reality.
Linuxgruven happens to be the employer of Kara Pritchard, Director of Exam Development for
the Linux Professional Institute. Feel free to check it out yourself on either LPI's or Linuxgruven's web sites.
While she is actively involved in LPI development and holds an RHCE, I don't believe that Kara has taken any Sair exams.
I also note that Linuxgruven has been very supportive of LPI efforts, and is recognized as a
bronze-level sponsor of LPI.
In fact, I said very little of all about the actual code of GNOME or KDE, and much about the perceived PR efforts of the two groups. Of course choice is good -- I've also covered that ground previously.
But this isn't about the ability to choose -- the choice of open source UIs existed before the GNOME Foundation or KDE League were ever envisioned. This is all about misguided PR, and the diverting of public attention away from the software itself and towards each group's desire to be a "standard" and backing by big-company endorsement.
If all they want to do is advance Linux as a desktop, both KDE and GNOME (and their supporters) would do better channeling publicity efforts through a neutral umbrella group such as Linux International. By creating their own PR efforts, independent of each other and the general Linux community, they call attention to the fragmentation rather than confronting common foes (Microsoft and Apple).
If you want to disagree with that point, go ahead. But be very clear that I'm not knocking the existence of multiple open source desktop environments, merely the undesired consequences of ill-founded PR initiatives.
OK, the points about the usability problems with Beowulf and commercial software such as databases and web servers are well taken.
But Beowulf isn't the only clustering technology for Linux. How do the business-targeted clustering products from TurboLinux or SGI (or others) stack up compared to the UnixWare clustering/SMP approach?
I was just making the point that, at least in the comments I had received while reseraching the piece, I'd found that new GUI apps were more likely being developed using Qt or Gtk. Lesstif is still very valuable, but mainly by those who have already invested resources into programming for the Motif API.
I can offer these reasons why and how LPI is different from MCSE and other certification programs:
- LPI started as a grassroots organization, its history goes back to active email discussions from within the Linux user community that grew into a formal body
- LPI is incorporated as a non-profit body; legally we are not allowed to make a profit
- We are vendor-neutral, and have a number of Linux companies as sponsors (with more coming) and even more on our Advisory Council. The intention is that no one vendor can dominate the LPI agenda.
- We are trying as much as possible to keep costs down. We have sought to minimize the number of exams required to demonstrate the required proficiency (MCSE's first level requires six or seven exams; LPI's requires three)
- As another example of trying to keep costs down, LPI will not be "expiring" certificates the way other programs do. While we all know that older certifications will be less usable, we will not revoke them
- LPI is not in the training or book business. How you get your skill is up to you. Take courses, read a book, or go into a cave with a Linux box for a month. All LPI cares about is that you know your stuff when you take the test.
- The exams will be challenging. It will be quite difficult for somebody who doesn't know the basics to pass the first level
- We NEVER advocate leaning on certification -- LPI's or anyone else's -- as a substitute for good interviews or reference checking etc. A certification program is helpful but it's not the complete answer by far.
Just check out the LPI web site to find out more.Asked for and received. Use of the word "Linux" in the name "Linux Professional Institute" has explicit permission of Linux International, which administers the trademark on Linus' behalf.
No such thing. Mine was an opinion (and clearly marked as "Commentary") that was expressed, based on observations I've made both amongst /. readers and amongst new users. I give quiet a few "Intro to Linux" lectures, and I believe I have a fairly good idea where new users are coming from. They're most certainly on a different wavelength than this bunch. Not better, not worse, just different.
The reaction I've seen to my column here does nothing to change the perception that an awful lot of people just don't get it. Some come across as "I get it, so what?", and that's OK. But some of the nastier replies I've seen (including some really precious ones in private email) have a common theme: how dare I suggest that /. readers are losing touch!
Looks from here like I've touched a nerve. Totally unintentional, but fascinating nonetheless.
Do your worst. I stand behind what I said, and incredibly little has been advanced here that refutes much of what I've said (beyond the name calling).
I was legitimately called into question on issue of fact and one issue of timing;
Are the above "gross errors"? Not at all. Are any of the other points factually wrong? Nothing else has been pointed out.
It's all-too-easy around these parts to disagree with an opinion by calling it a lie. I write opinion pieces by design, yet I try to do research to make sure I can back up what I say. In this case I'd interviewed the presidents of MandrakeSoft, Macmillan, and Red Hat before writing. There's also a lot of background to what I wrote that never made it into print.
As for "false conceptions", you'll have to explain what that means. If you think I've achieved incorrect or inaccurate perceptions based on facts at hand, then we can debate that. But you haven't done anywhere near the very difficult task of proving that an opinion is false.
So go ahead, be very concerned with truth. But also keep in mind that my piece says "Commentary" at the top, that I have done a fair bit of homework before preparing it, and I'll be more than happy to rethink things if someone offers a different intepretation of the same premises.
Aimlessly throwing the word "lie" around accomplishes little of value, and certainly does little to disprove my conclusions. Did I lie, as some have suggested, because I deliberately did a /. search on "Macmillan" rather than "Mandrake"? Not one shred, especially since the Macmillan box of Mandrake has some significant non-technical differences with MandrakeSoft's own package -- bundled-in Linuxcare tech support, for one thing. You may not agree with my opinion that this is a significant distinction, but it ain't no lie.
Anyhow, I still stand behind what I said, minor flaws above notwithstanding. Want to disagree? Fine! But indiscriminant shouting of "liar!" gets one nowhere...
I've had a lot of pieces picked up by /. without mentioning it, that generated a lot more response. I didn't ask for this one to be put in, and really don't care that it's here.
If the /. editors think that a mere mention of their site in an article makes something immediately newsworthy, that's ego at work rather than judgement -- the same ego that seems to think that a troll accusation is "insightful".
So let the merger happen, for all I care. So long as cablemodem Internet users can opt out of AOL/@Home, they can do whatever they want...
this followup column.
- Sun released 2.2E for Solaris, but refused to bring those important bugfixes to any other port (including Linux), where it stopped at 2.2D;
- When it became clear that selling and maintaining Wabi was more trouble than it was worth, Sun could have and should have open sourced the parts that it owned, to at least provide a safety net for the newly-orphaned users. Instead, it prevented licensees from even doing their own bugfixes.
I understand what you're saying, and grant that the dropping of Wabi wasn't done specifically to hurt Linux users -- but that's indeed what the consequences were. Subsequent actions could have lessened the blow (such as allowing users and/or sub-licensees to maintain the orphaned product), but Sun chose the path of greatest shafting instead.Part of my original point was that StarOffice is a strategic product for Sun, much like Wabi was. As Linux users got shafted then by the way Sun treated Wabi (deliberately or not), we must watch that the scenario doesn't repeat itself with StarOffice.
Contrast Sun's actions wrt Wabi with the decision to LGPL the Willows TWIN software when the company decided it wasn't going to sell. Now TWIN is a part of helping Wine be ready for prime time.
Wabi could have also helped Wine, but Sun chose to kill the software and orphan its users because it wanted to strategically push Java instead of emulation.
Sun's future would have people depending on mega-ISPs (using Solaris servers, of course) that would store your files and serve your apps for you. Think WebTV with decent video.
Sun's intent is quite clear in this regard. The main point of my argument, which will be concluded in a second column appearing next week, is that neither Sun's nor Microsoft's world-view is right for everyone. Thin clients are great in some situations, horrible in others. Linux is flexible enough to serve as an excellent thin-client, fat-client, or server. Neither of the commercial approaches is best in all cases, yet neither does much to acommodate models outside of its own.
Linux is a threat to both Microsoft and Sun because it offers the flexibility to serve either model -- or some hybrid of the two -- very well. Linux isn't trying to push the user into one world-view or another of the way computing should be done.
I don't think it's reasonable to paint all ZDNet authors as Linux hostile because of Berst, any more than calling all of IDG that way because of Metcalfe.
It's not uncommon for publishers to have, on the same payroll, multiple commentators with very differing opinions.
Not all the readers out there are smart enough to understand the message intended by ignoring someone who spews. If someone has an audience that buys his arguments, and no rebuttal comes, this silence could be interpreted as an inability to rationally rebut the original spew.
I stand behind my decision to confront this stuff rather than pretend it didn't exist. Yes, it draws more attention to the original spew, but that attention comes in the context of ridicule rather than authoritative opinion. And it was important (to me, at least) to call attention to the kind of spew that people will start encountering when recommending Linux to their employers or clients.
There are times when you let a ranter flail away in silence. Other times -- especially when the speaker has an audience -- that it's necessary to sound the BS alarm before anyone gets sucked in by sophomric namecalling masquerading as cleverness.
PS: This is not a pissing match between ZD and IDG, it's me alone calling attention to something that happened to come from InfoWorld. I have a lot of respect for Petreley and other writers at IDG.
Evan
It wasn't very difficult to make the transition from SCO to Linux -- it started for me mid-November, 1995.
I haven't been around the biz.* newsgroups (heck, newsgroups at all) for many years.
As for ZDNet, that's a longer story...