I remember it from installing Linux on a DOS system some years back and markting the Linux partition active without having installed LILO on the boot block.
Case in point, you mention schools. Art departments have more nudity than your average issue of Hustler, and rightly so.
Hey, at my elementary school, there wasn't even an art department! There was just an art teacher who would come in once a week. She didn't bring any naked pictures, either. It was normally just paints and construction paper, and on occasion some popcorn seeds.
I have trouble seeing nudity in an art department going over in any US elementary school after that incident with Janet Jackson and the wardrobe malfunction.
The cynic in me suspects this is a move to ensure that the huge bribes^H^H^H^H^H^Hcampaign donations keep rolling in from Microsoft at least through the next US presidential election. The only real downside of this ruling for Microsoft appears to be the risk of a less-friendly attorney general taking office -- that is, through a Republican Party loss in the 2008 presidential elections.
Yeah, here at Sun there are nearly 40 thousand people that connect through 40-50 different proxy servers. That is a thousand to 1 in the opposite direction the article claims.
That's a good point. However, the majority of the sites being visited through the Sun proxy are probably also being visited by the same users from home as well. So, while Sun may not be adding much to the skew, it's quite a leap to say it skews it in the other direction 1000:1
For example, I have visited 3 or 4 of the same websites from the following locations last week:
1) the computer lab at the university
2) my dsl at home
3) two different cybercafés
While the cybercafés and my university are probably both using proxies with a single IP, that still means my hits show up from 4 IP's while I am only one visitor.
This really sounds like asking "Where are the boundaries (if any) to mammals" during the late Mesosoic era. The answer, as I see it, is they are best off out of reach of the dinosaurs busy racing to extinction.
First of all you can't have promotion of science or arts without willing participants. Anything else reeks of artistic or intellectual slavery.
If the participants are willing without a state-granted monopoly over their programs, it's not slavery. As a counterexample, staking off entire classes of "ideas" with patents reeks of privatized water in Bolivia where people were being prohibited from collecting rainwater, as it was deemed a violation the state-granted monolopy on water distribution.
Then there is the other problem of copyright when applied to software for which no source code is supplied: It's not advancing the arts in the same way the written word was, as what was actually written is not on offer. It is illegal also to break DRM, which further restricts that, especially applied to media with a lifespan which is shorter than the duration of copyright. So, the side of the exchange where the copyrighted material is released to the public in exchange for the temporary monopoly never actually happens.
Second IP is NOT the exclusive domain of corporations. An awful lot of slashdotters forget that simple fact in their zeal.
I wasn't implying that copyright was limited to being used by corporations. I was discussing the issue brought up by the parent comment, which was the relative scare value of RMS versus the standard big business stance. As I said, I think that a company is much better off using Free Software than the alternative.
Anyway, when it comes to "stealing IP", Microsoft, for example, isn't the one with the clean track record. Remember Stacker? Or their partners they keep for long enough to get the trade secrets and run? Comparatively, RMS seems ultra-low-risk to me.
Third the "theft"* you accuse corporations of is happening (not all mind you, just some). However any viable society has legitimate means for effecting change. The problem I have is that they're not being used, and just as bad people are making excuses for why they're not even going to try to use them.
Well, isn't that what we are doing here? Slashdot played a big role in getting Free Software and Open Source being seen in the mainstream. The problem is more cultural than it is political. Helping to screw heads on straight here can end up having benefits down the road.
Simply writing a letter to a senator isn't going to do it. You don't explain with reason in politics, you explain with numbers, whether it be a number on a check, or a number in a poll. Reason is what you use with people.
*I prefer to think of it as a violation of a social contract (and don't any of you smart-asses give me that "but I didn't sign anything nonsense")
Turnabout is fair play. It's the "IP Lobby" who insists on using the word "theft" for unauthorized copying.
Maybe you might consider this a trolling or a flame, but I think that it is quotes such as these that may end up bringing the most amount of trouble for the RMS crowd... I think the man is losing touch with reality, and approaching a point where zealotry is clowding his judgment to a dangerous level. How can we convince businesses that using the GPL and open source is a GOOOD THING if one of the main characters is in effect condoning IP theft if done for the 'right reasons'?
The only thing that has changed is RMS is being interviewed by Forbes now. If you had read his essay "Why software should not have owners", for instance, it would have been apparent that he has stayed true to his goals over the years.
Actually what I find disturbing is the "IP" proponents are proposing that DRM be sanctified as more precious than human life. Personally, I would much prefer someone like RMS who would support giving me the source to any programs that run my company as an assurance of never being left high and dry or strongarmed to someone who says they would have no problems with killing me if it would help the bottom line the next quarter.
Also, I think it bears mentioning that RMS actually is in favor of the right to write your programs and keep them completely private, not releasing them to anyone. That's one of the reasons he entered the fray of the big Apple license debate some years back. The license was requiring that any changes be sent back to Apple, whether or not the resulting source/binary was released to anyone.
Further, when it comes to discussing the ethical basis of copyright, I think it bears repeating that the reason we have copyright at all in the United States is to "promote the progress of science and the useful arts". I'm not sure if you noticed, but it has been a trend lately amongst many technology and entertainment companies to injure the progress of science and the useful arts freely as long as they think it will keep them in the money. That includes activities such as subverting international standards organizations and activities such as subverting national governments so that they retroactively extend copyright, effectively "stealing IP", to use your terms, from all of humanity (or at least all of that country's citizens).
Thanks for the info! I noticed that the packages from the other CD's are indeed on public mirrors, which does indeed make ISO downloads a moot point, and the official X.org is working again.:)
I had only seen the set of new Xorg packages that recently showed up with a priority of security. I took that as meaning I'd only get security fixes from those, and chosen the SoS security updates instead.
I've been using Mandrake since 2001, when I switched from Debian to get a version of X that would support my new video card. At the time, it was was flourishing, engaged the community surrounding it, and was hiring developers who were working on projects that were making crucial advances for Linux. One that comes to mind was the developer of a partition resizer that would work on NTFS back when when all the other distros were instructing their users to use Partition Magic.
Of course, all that great work had a price tag attached to it, so when Mandrake Club was announced, I was first in line to join. The idea back then was that it was a voluntary donation with no extra benefits other than supporting continued development.
Unfortunately, once the club started to take off, they started closing things off to the public one by one to drive membership numbers higher. Now it's to the point where standard members can't even download the full set of CD images for their $60 yearly membership fees.
Something seems to have really changed in a big way since the Connectiva merger, though. With the release of Mandriva 2006, they've been focusing on marketing deals like that with Skype. Then, there was the worldwide Mandriva party, where the locations weren't announced until the night before... until then, there was just a form to fill out for organizations to get corporate schwag.
Also, I was reading on the Mandriva forums earlier that the reason their cut of X.org doesn't work with my ATI Radeon 7500 is that they "chose the wrong X.org" and are staying with it due to an Intel marketing agreement. Luckily, seerofsouls.org has working RPM's, but needing to depend on a third party to provide core components of the distribution is not exactly ideal.
Anyway, it looks like their management has decided that it wants to be Red Hat or Novell. I wish them good luck with that. I've seen it mentioned that PCLinuxOS is trying to be what Mandrake was, so hopefully they will provide a good upgrade path from Mandriva so I can get off this sinking ship without getting my clothes too wet.
You decide which framework to use the same way you decide which ANYTHING to use! You research the choices, go through the available information on each, and quickly become overwhelmed with the details.
Then you get the one featured the most prominently in ads!
I think the most important possible thing is to take input liberally. It is likely that specific sysadmins will be more experienced than you in certain areas, or be able to figure out problems first. If so, you should thank the admin who gave you the input (where his boss can see), and implement it. The key being that you try not to be a hurdle, instead try to make things more effective by co-ordinating efforts, while at the same time winning allies by giving positive feedback to those who are interested in team effort.
Of course, if you are single-handedly responsible for running 4 platforms, you will be overloaded, which will negatively impact everyone if you are the only one given administrative rights. I would try to plan for that up-front by developing a procedure and getting management buy-in for overflow/emergency situations where admin privileges can be delegated where necessary. Eg., sticky problem on an admin's workstation that could take hours of troubleshooting? Instead of making him take downtime and wait for to be free, let him take a crack at it himself.
I have been in simliar situations, with another team managing common-use infrastructure machines, and a patchlevel/base software compliance tool scanning aspects of the system. In all cases you should make it clear that you are doing your job, and not be patronizing to your peers.
i.e., get a requirement from management? Refer to it as a requirement from management, and don't try to "sell" it, unless you really do think it's a good idea. Of course, if one of your fellow admins succeeds in convincing you otherwise while you are selling an idea you initially agreed with, don't pull rank, try to get things worked out to everyone's advantage.
In all cases, I think focusing on the positive teamwork will serve you better than getting drawn into disputes. Let management fight their own battles when someone is being difficult. After all, you're sysadmin, not nanny, and you will have to work with the others every day.:)
I guess this is a good example about why RMS suggests avoiding the term IP. As for the companies mentioned, I think the real reason they didn't go anywhere was because they were trying to serve a market that didn't exist.
Since most of the shops who adopted Linux ended up coming from Unix, they didn't need these companies to hold their hands through every command issued at the prompt, and since Linux runs on commodity hardware, the value add of "Linux-compatible hardware at exhorbitant prices" wasn't really much of a selling point.
If you look at the open source businesses which are still around, like Red Hat, Mandriva, and Sleepycat, they aren't doing it through "IP", they're doing it by selling services around something they, themselves produce. I think that's what it boils down to.
I suppose there is a loose relationship to their success with trademark and branding, but I think the most significant factor is that they are selling services as the suppliers, rather than as third parties. The emphasis both being in the position to really deliver something useful, and with having the right credentials as being the ones creating what they're servicing!
I just don't see that much of a market for generic "Linux support", "Linux hardware", "Linux consulting" as such. The "Linux" part is the easy part, after all, no matter what we deluded ourselves into thinking in 1999. I think business generally are looking at either getting a particular task done or just doing what they've always done, but adding Linux to the mix.
So rather than a Linux cabal guiding companies through the forbidden realms of Linux and Open Source, the money is to be had more in specialized areas which also happen to involve Linux, or more generalized areas, where Linux is just one of a list of services offered. I think it's pretty clear that businesses are more likely to use their own in-house talent, or their outsourced IT (eg., IBM) for base Operating System support, and standard whiteboxes rather than paying a premium for "wow, the vendor has Linux in its name!" stuff.
Having worked in large heterogenous environments, I can attest to the fact that the processes used to run Linux and the processes used to run Solaris aren't really that much different. Yeah, you run rpm instead of patchadd, and call your on-site whitebox support instead of the on-site Sun support when a component fails, but it doesn't require a multibillion dollar triumvirate to bridge that gap.
You might say that those companies failed because of IP - funded because they had "Linux" in their names, and spending millions of dollars to create huge branded behemoths. But given the era, I think it would be more accurate to say that they failed because of IPO.:)
I think a better way to phrase Mr. Rosenberg's message is: Don't try to sell a commodity as though you're the only one in the market who can provide it.
Ironically, the companies the article is warning are using the Red Hat/Mandriva supplier-based services model, while it brings up LinuxCare as a bad example. Sure, these companies might end up producing something simple enough that no one needs their services, is better done elsewhere, or simply addresses non-existent markets.. but at least they're walking a path more like that of Red Hat than that of Linuxcare.
Maybe to be more inclusive the article should read "Rapshody Targets non-windows users" instead of linux. Last I checked Safari was was not available for linux.
A music subscription service for Mac isn't newsworthy, though, with iTunes being widely available there.;) As for BSD, while it may be possible to use this service there, the realplayer port requires Linux emulation to be installed last I checked, so they wouldn't appear to be 'targetted'. Rather, they'd be jumping in front of the arrow, hoping to be grazed.
Here are a couple of interesting snippets from the GPLv3 process pdf that I didn't see in the other responses:
While the GPL is the most popular Free Software License, followed by
the LGPL, a significant set of free software is licensed under other terms
which are not compatible with version 2 of the GPL. Version 3 of the GPL
will provide compatibility with more non-GPL free licenses.
. . .
Version 3 of the GPL should reduce the difficulties of
internationalization. Version 3 should more fully approximate the otherwise
unsought ideal of the global copyright license.
When you have one of the world's largest collection of nuclear missiles effectively being controlled by what other comments have identified as a band of criminals, it sounds like a matter of global importance to me. After all, UN inspectors are sent into other countries with a history of corrupt elections. And wasn't that a similar argument behind the US invasion of Iraq?
- which tried to use bleeding-edge software on an old OS software platform
Answer: All the components used were available in the time-correct period of the study. For example, if they installed a component in the simulated September 2004 time period then that version was available in September 2004.
- which apparently didn't upgrade the system first if that's what they had in mind
Answer: Good point! The only configuration control issue was that the enterprise wouldn't upgrade the OS version until July 1, 2005. This is mainly based on our experience with companies that don't move to the latest OS version until it has had time to "bake" in the community. At that time, SLES 9 was hot off the compiler.
How about your experience with companies which generally don't like modify their operating systems to the point of breaking any operating system support contract that could reseaonbly be expected to exist?
Since you've chosen rapid deployment of a new version vendor application as a target, that implies that the application had been identified as a critical project, and therefore vendor support of the components would be a requirement - trumping any general guidelines as to operating system deployment schedules.
What you've ended up doing is setting up a scenario where the company is choosing what amounts to in-house distro with no external support instead the new version of a distro which would would have ongoing support from the vendor.
Since you've solicited feedback, I'd say this is a glaring error in the study. Would you create a study that involved replacing critical core OS componets from Windows Server 2003 and putting them onto Windows 2000? With your stated requirements for Linux, the only alternatives would appear to be giving up, saying that the application couldn't be upgraded, or documenting the vain efforts of the Windows administrators breaking Windows in the same way you required the Linux admins to break Linux.
I didn't comment on the two previous stories, but I think the concerns over reputation are valid.
As the researcher has stated, this study doesn't really say anything that can be applied in a general manner between the operating systems, yet it should have been obvious that Microsoft would treat any positive results as "proof" that Windows is more maintainable than Linux. And, that's exactly what they proceeded to do.
This is what smells funny to me:
We can never have a verifiable answer to the following questions:
1) How many studies did Microsoft commission from this researcher or his company?
2) Was there an extra fee demanded for rights to publish the results? (If so, there is an inherent incentive to select for cases which would favor Microsoft)
3) Might we imagine that Microsoft would be more likely to give him repeat business if favorable results were given?
Regarding the glibc issue, common business requirements are that one uses a certified application stack from the software vendor. Anything that would be 'certified' on a Frankenstein system with the system glibc torn out and replaced is a bit of a stretch. One would also expect that Novell would not support such a configuration - getting into that sort of situation is generally a big no-no in an enterprise environment. Speaking as a system administrator, the fact that something this took place removes any credibility this study may have had.
Of course, these studies can sometimes lead to positive change, but often at the expense of the reputation of researcher or company lending its efforts to the Microsoft marketing machine.
In the case of shoe-horning an application into an unsupported distro, however, the only thing it really says is: Don't create business processes that require pounding a square peg into a round hole.
We will never know the true story about what happened between the researcher and Microsoft, but, I personally have trouble accepting any Microsoft-sponsored research concerning the competition based simply on their long history of backroom deals aimed at gaining some sort of tactical advantage.
As for your 'slashbot' remark - Humankind in general has been judging people by the company they keep for quite some time now.
Reading between the lines, it sounds like Google has caught onto the standard tactics used by executives in their "Deal Making". That is, making whatever deal that gives them the best kickback, as long as they're supplied with a Powerpoint deck good enough to lend plausible deniability.
I suspect this is just sour grapes on the part of executives who were called on the carpet for this type of behavior, especially since that the guy interviewed in the article seemed shocked and amazed that he was asked to demonstrate his acumen prior to being hired.
Having seen my share of "Strategic Initiatives" (code for: we're going to rip out a working system to replace it with a multimillion dollar "solution" barely works, requiring 10x the hardware and manpower to operate, and no, we aren't open to feedback from the technical staff) over the course of my career only convines me further that this is what's going on. Incorporating sanity checks by technical types sounds more to me like removing a caste system (with the suits as Brahmins) rather than creating one.
And without a caste system, why should the suits be exempt from requirements of competence and integrity?
I the fact that Firefox is being promoted again says more than what the promotions might actually contain. There are often short-lived projects or products which ride a wave of hype, then die off, so any message would imply some sort of staying power.
Of course 'real people' telling you how great it is sort of implies that the Beautiful People will be saying it's Cool and Popular to use Firefox®! It's not enough to just have an iPod® these days. I mean OMG what if the in-crowd finds out that you use MSIE! You could be spreading a VIRUS! Or have WORMS!
I think those who are mostly concerned about the feature set aren't the ones waiting for real people in ads telling them how great it is.;)
Why not just use 2 different algorithms? Yes, it's possible. Or hell, use 3. Can some one tell me why not this isn't a standard practice? Even if one has a weakness, you still have the other to back it up
I noticed that NetBSD's source-based package management system, pkgsrc, already does this using SHA1 and RMD160 (apparently RIPEMD-160 is the official name for the digest). Here's what it looks like in the archive fetching phase of a package installation:
=> Checksum SHA1 OK for unzip-5.52/unzip552.tar.gz.
=> Checksum RMD160 OK for unzip-5.52/unzip552.tar.gz.
===> Extracting for unzip-5.52nb2
One might also imagine that colliding two different hash types at the same time would be much more difficult than only at a time, anyway.
The only reference I have heard made to the queen growing up in the United States was when adults would try to instill a sense of nonchalance in children who giggled at flatulence. "Even the Queen of England farts!", is what they would say.
It's a simple process, but apparently thre's a lot of sysadmins out there who can't be bothered to exercise due dilligence...hence, my accusation of complacency.
You sound like you are a bit out of your depth here. The topic at hand is corporations losing millions of dollars. At those corporations, the operations team sysadmins aren't the ones with the authority to build new servers. Neither are they the ones who perform application-level testing.
Often there are a multitude of teams and managers involved in the process, and the sysadmin only has permission to act after the application teams have tested their applications in the Development and QA environments all concerned parties have agreed to give permission to install a patch in production.
Sometimes they don't give that permission for no other reason than they have the power to say no and have the idea that "if it's working don't fix it".
If anything, I would say management is more responsible for patches not being deployed than sysadmins, because any large, well-run operation will have security update tracking to ensure compliance with relevant patch levels. Management is also who defines policy concerning the political priority of patching vs. development and release dates, staffing levels, etc, all of which have an impact.
They are also the ones who are in a position to negotiate with other teams and their management.
I've also seen the problem of management enforcing application vendor requirements to use, for instance, a specific point revision and "certified" build of Apache so as to not allow security updates.
Ultimately, management is who is responsible for making sure things are working properly and that the company is able to make money, too.
Word to the wise, in such environments, flying off the handle and blaming someone for not doing something they had no permission to do is generally considered bad form. Especially in the cases when we really do want to do the right thing, and put in great amounts effort, of which the actual patching is perhaps 5%, in order to have it get done.
Slashdot Mirror
Thanks for the tip!
See http://support.microsoft.com/default.aspx?scid=kb; en-us;321626
I remember it from installing Linux on a DOS system some years back and markting the Linux partition active without having installed LILO on the boot block.
Hey, at my elementary school, there wasn't even an art department! There was just an art teacher who would come in once a week. She didn't bring any naked pictures, either. It was normally just paints and construction paper, and on occasion some popcorn seeds.
I have trouble seeing nudity in an art department going over in any US elementary school after that incident with Janet Jackson and the wardrobe malfunction.
The cynic in me suspects this is a move to ensure that the huge bribes^H^H^H^H^H^Hcampaign donations keep rolling in from Microsoft at least through the next US presidential election. The only real downside of this ruling for Microsoft appears to be the risk of a less-friendly attorney general taking office -- that is, through a Republican Party loss in the 2008 presidential elections.
For example, I have visited 3 or 4 of the same websites from the following locations last week:
1) the computer lab at the university
2) my dsl at home
3) two different cybercafés
While the cybercafés and my university are probably both using proxies with a single IP, that still means my hits show up from 4 IP's while I am only one visitor.
This really sounds like asking "Where are the boundaries (if any) to mammals" during the late Mesosoic era. The answer, as I see it, is they are best off out of reach of the dinosaurs busy racing to extinction.
If the participants are willing without a state-granted monopoly over their programs, it's not slavery. As a counterexample, staking off entire classes of "ideas" with patents reeks of privatized water in Bolivia where people were being prohibited from collecting rainwater, as it was deemed a violation the state-granted monolopy on water distribution.
Then there is the other problem of copyright when applied to software for which no source code is supplied: It's not advancing the arts in the same way the written word was, as what was actually written is not on offer. It is illegal also to break DRM, which further restricts that, especially applied to media with a lifespan which is shorter than the duration of copyright. So, the side of the exchange where the copyrighted material is released to the public in exchange for the temporary monopoly never actually happens.
I wasn't implying that copyright was limited to being used by corporations. I was discussing the issue brought up by the parent comment, which was the relative scare value of RMS versus the standard big business stance. As I said, I think that a company is much better off using Free Software than the alternative.
Anyway, when it comes to "stealing IP", Microsoft, for example, isn't the one with the clean track record. Remember Stacker? Or their partners they keep for long enough to get the trade secrets and run? Comparatively, RMS seems ultra-low-risk to me.
Well, isn't that what we are doing here? Slashdot played a big role in getting Free Software and Open Source being seen in the mainstream. The problem is more cultural than it is political. Helping to screw heads on straight here can end up having benefits down the road.
Simply writing a letter to a senator isn't going to do it. You don't explain with reason in politics, you explain with numbers, whether it be a number on a check, or a number in a poll. Reason is what you use with people.
Turnabout is fair play. It's the "IP Lobby" who insists on using the word "theft" for unauthorized copying.Actually what I find disturbing is the "IP" proponents are proposing that DRM be sanctified as more precious than human life. Personally, I would much prefer someone like RMS who would support giving me the source to any programs that run my company as an assurance of never being left high and dry or strongarmed to someone who says they would have no problems with killing me if it would help the bottom line the next quarter.
Also, I think it bears mentioning that RMS actually is in favor of the right to write your programs and keep them completely private, not releasing them to anyone. That's one of the reasons he entered the fray of the big Apple license debate some years back. The license was requiring that any changes be sent back to Apple, whether or not the resulting source/binary was released to anyone.
Further, when it comes to discussing the ethical basis of copyright, I think it bears repeating that the reason we have copyright at all in the United States is to "promote the progress of science and the useful arts". I'm not sure if you noticed, but it has been a trend lately amongst many technology and entertainment companies to injure the progress of science and the useful arts freely as long as they think it will keep them in the money. That includes activities such as subverting international standards organizations and activities such as subverting national governments so that they retroactively extend copyright, effectively "stealing IP", to use your terms, from all of humanity (or at least all of that country's citizens).
Thanks for the info! I noticed that the packages from the other CD's are indeed on public mirrors, which does indeed make ISO downloads a moot point, and the official X.org is working again. :)
I had only seen the set of new Xorg packages that recently showed up with a priority of security. I took that as meaning I'd only get security fixes from those, and chosen the SoS security updates instead.
I've been using Mandrake since 2001, when I switched from Debian to get a version of X that would support my new video card. At the time, it was was flourishing, engaged the community surrounding it, and was hiring developers who were working on projects that were making crucial advances for Linux. One that comes to mind was the developer of a partition resizer that would work on NTFS back when when all the other distros were instructing their users to use Partition Magic.
Of course, all that great work had a price tag attached to it, so when Mandrake Club was announced, I was first in line to join. The idea back then was that it was a voluntary donation with no extra benefits other than supporting continued development.
Unfortunately, once the club started to take off, they started closing things off to the public one by one to drive membership numbers higher. Now it's to the point where standard members can't even download the full set of CD images for their $60 yearly membership fees.
Something seems to have really changed in a big way since the Connectiva merger, though. With the release of Mandriva 2006, they've been focusing on marketing deals like that with Skype. Then, there was the worldwide Mandriva party, where the locations weren't announced until the night before... until then, there was just a form to fill out for organizations to get corporate schwag.
Also, I was reading on the Mandriva forums earlier that the reason their cut of X.org doesn't work with my ATI Radeon 7500 is that they "chose the wrong X.org" and are staying with it due to an Intel marketing agreement. Luckily, seerofsouls.org has working RPM's, but needing to depend on a third party to provide core components of the distribution is not exactly ideal.
Anyway, it looks like their management has decided that it wants to be Red Hat or Novell. I wish them good luck with that. I've seen it mentioned that PCLinuxOS is trying to be what Mandrake was, so hopefully they will provide a good upgrade path from Mandriva so I can get off this sinking ship without getting my clothes too wet.
You decide which framework to use the same way you decide which ANYTHING to use! You research the choices, go through the available information on each, and quickly become overwhelmed with the details.
Then you get the one featured the most prominently in ads!
You do have Adblock disabled, right?
I think the most important possible thing is to take input liberally. It is likely that specific sysadmins will be more experienced than you in certain areas, or be able to figure out problems first. If so, you should thank the admin who gave you the input (where his boss can see), and implement it. The key being that you try not to be a hurdle, instead try to make things more effective by co-ordinating efforts, while at the same time winning allies by giving positive feedback to those who are interested in team effort.
:)
Of course, if you are single-handedly responsible for running 4 platforms, you will be overloaded, which will negatively impact everyone if you are the only one given administrative rights. I would try to plan for that up-front by developing a procedure and getting management buy-in for overflow/emergency situations where admin privileges can be delegated where necessary. Eg., sticky problem on an admin's workstation that could take hours of troubleshooting? Instead of making him take downtime and wait for to be free, let him take a crack at it himself.
I have been in simliar situations, with another team managing common-use infrastructure machines, and a patchlevel/base software compliance tool scanning aspects of the system. In all cases you should make it clear that you are doing your job, and not be patronizing to your peers.
i.e., get a requirement from management? Refer to it as a requirement from management, and don't try to "sell" it, unless you really do think it's a good idea. Of course, if one of your fellow admins succeeds in convincing you otherwise while you are selling an idea you initially agreed with, don't pull rank, try to get things worked out to everyone's advantage.
In all cases, I think focusing on the positive teamwork will serve you better than getting drawn into disputes. Let management fight their own battles when someone is being difficult. After all, you're sysadmin, not nanny, and you will have to work with the others every day.
I guess this is a good example about why RMS suggests avoiding the term IP. As for the companies mentioned, I think the real reason they didn't go anywhere was because they were trying to serve a market that didn't exist.
:)
Since most of the shops who adopted Linux ended up coming from Unix, they didn't need these companies to hold their hands through every command issued at the prompt, and since Linux runs on commodity hardware, the value add of "Linux-compatible hardware at exhorbitant prices" wasn't really much of a selling point.
If you look at the open source businesses which are still around, like Red Hat, Mandriva, and Sleepycat, they aren't doing it through "IP", they're doing it by selling services around something they, themselves produce. I think that's what it boils down to.
I suppose there is a loose relationship to their success with trademark and branding, but I think the most significant factor is that they are selling services as the suppliers, rather than as third parties. The emphasis both being in the position to really deliver something useful, and with having the right credentials as being the ones creating what they're servicing!
I just don't see that much of a market for generic "Linux support", "Linux hardware", "Linux consulting" as such. The "Linux" part is the easy part, after all, no matter what we deluded ourselves into thinking in 1999. I think business generally are looking at either getting a particular task done or just doing what they've always done, but adding Linux to the mix.
So rather than a Linux cabal guiding companies through the forbidden realms of Linux and Open Source, the money is to be had more in specialized areas which also happen to involve Linux, or more generalized areas, where Linux is just one of a list of services offered. I think it's pretty clear that businesses are more likely to use their own in-house talent, or their outsourced IT (eg., IBM) for base Operating System support, and standard whiteboxes rather than paying a premium for "wow, the vendor has Linux in its name!" stuff.
Having worked in large heterogenous environments, I can attest to the fact that the processes used to run Linux and the processes used to run Solaris aren't really that much different. Yeah, you run rpm instead of patchadd, and call your on-site whitebox support instead of the on-site Sun support when a component fails, but it doesn't require a multibillion dollar triumvirate to bridge that gap.
You might say that those companies failed because of IP - funded because they had "Linux" in their names, and spending millions of dollars to create huge branded behemoths. But given the era, I think it would be more accurate to say that they failed because of IPO.
I think a better way to phrase Mr. Rosenberg's message is: Don't try to sell a commodity as though you're the only one in the market who can provide it.
Ironically, the companies the article is warning are using the Red Hat/Mandriva supplier-based services model, while it brings up LinuxCare as a bad example. Sure, these companies might end up producing something simple enough that no one needs their services, is better done elsewhere, or simply addresses non-existent markets.. but at least they're walking a path more like that of Red Hat than that of Linuxcare.
Urhm, how was that Flamebait?
When you have one of the world's largest collection of nuclear missiles effectively being controlled by what other comments have identified as a band of criminals, it sounds like a matter of global importance to me. After all, UN inspectors are sent into other countries with a history of corrupt elections. And wasn't that a similar argument behind the US invasion of Iraq?
How about your experience with companies which generally don't like modify their operating systems to the point of breaking any operating system support contract that could reseaonbly be expected to exist?
Since you've chosen rapid deployment of a new version vendor application as a target, that implies that the application had been identified as a critical project, and therefore vendor support of the components would be a requirement - trumping any general guidelines as to operating system deployment schedules.
What you've ended up doing is setting up a scenario where the company is choosing what amounts to in-house distro with no external support instead the new version of a distro which would would have ongoing support from the vendor.
Since you've solicited feedback, I'd say this is a glaring error in the study. Would you create a study that involved replacing critical core OS componets from Windows Server 2003 and putting them onto Windows 2000? With your stated requirements for Linux, the only alternatives would appear to be giving up, saying that the application couldn't be upgraded, or documenting the vain efforts of the Windows administrators breaking Windows in the same way you required the Linux admins to break Linux.
I didn't comment on the two previous stories, but I think the concerns over reputation are valid.
As the researcher has stated, this study doesn't really say anything that can be applied in a general manner between the operating systems, yet it should have been obvious that Microsoft would treat any positive results as "proof" that Windows is more maintainable than Linux. And, that's exactly what they proceeded to do.
This is what smells funny to me:
We can never have a verifiable answer to the following questions:
1) How many studies did Microsoft commission from this researcher or his company?
2) Was there an extra fee demanded for rights to publish the results? (If so, there is an inherent incentive to select for cases which would favor Microsoft)
3) Might we imagine that Microsoft would be more likely to give him repeat business if favorable results were given?
Regarding the glibc issue, common business requirements are that one uses a certified application stack from the software vendor. Anything that would be 'certified' on a Frankenstein system with the system glibc torn out and replaced is a bit of a stretch. One would also expect that Novell would not support such a configuration - getting into that sort of situation is generally a big no-no in an enterprise environment. Speaking as a system administrator, the fact that something this took place removes any credibility this study may have had.
Of course, these studies can sometimes lead to positive change, but often at the expense of the reputation of researcher or company lending its efforts to the Microsoft marketing machine.
In the case of shoe-horning an application into an unsupported distro, however, the only thing it really says is: Don't create business processes that require pounding a square peg into a round hole.
We will never know the true story about what happened between the researcher and Microsoft, but, I personally have trouble accepting any Microsoft-sponsored research concerning the competition based simply on their long history of backroom deals aimed at gaining some sort of tactical advantage.
As for your 'slashbot' remark - Humankind in general has been judging people by the company they keep for quite some time now.
Reading between the lines, it sounds like Google has caught onto the standard tactics used by executives in their "Deal Making". That is, making whatever deal that gives them the best kickback, as long as they're supplied with a Powerpoint deck good enough to lend plausible deniability.
I suspect this is just sour grapes on the part of executives who were called on the carpet for this type of behavior, especially since that the guy interviewed in the article seemed shocked and amazed that he was asked to demonstrate his acumen prior to being hired.
Having seen my share of "Strategic Initiatives" (code for: we're going to rip out a working system to replace it with a multimillion dollar "solution" barely works, requiring 10x the hardware and manpower to operate, and no, we aren't open to feedback from the technical staff) over the course of my career only convines me further that this is what's going on. Incorporating sanity checks by technical types sounds more to me like removing a caste system (with the suits as Brahmins) rather than creating one.
And without a caste system, why should the suits be exempt from requirements of competence and integrity?
Of course 'real people' telling you how great it is sort of implies that the Beautiful People will be saying it's Cool and Popular to use Firefox®! It's not enough to just have an iPod® these days. I mean OMG what if the in-crowd finds out that you use MSIE! You could be spreading a VIRUS! Or have WORMS!
I think those who are mostly concerned about the feature set aren't the ones waiting for real people in ads telling them how great it is.
=> Checksum SHA1 OK for unzip-5.52/unzip552.tar.gz.
=> Checksum RMD160 OK for unzip-5.52/unzip552.tar.gz.
===> Extracting for unzip-5.52nb2
One might also imagine that colliding two different hash types at the same time would be much more difficult than only at a time, anyway.
I use vi!
Really, think about it. Ninjas travel light, with simple but effective weapons. It's the pirates who travel with entire ships wherever they go!
http://www.allwords.com/word-Queen's%20English.htm l
The only reference I have heard made to the queen growing up in the United States was when adults would try to instill a sense of nonchalance in children who giggled at flatulence. "Even the Queen of England farts!", is what they would say.
You sound like you are a bit out of your depth here. The topic at hand is corporations losing millions of dollars. At those corporations, the operations team sysadmins aren't the ones with the authority to build new servers. Neither are they the ones who perform application-level testing.
Often there are a multitude of teams and managers involved in the process, and the sysadmin only has permission to act after the application teams have tested their applications in the Development and QA environments all concerned parties have agreed to give permission to install a patch in production.
Sometimes they don't give that permission for no other reason than they have the power to say no and have the idea that "if it's working don't fix it".
If anything, I would say management is more responsible for patches not being deployed than sysadmins, because any large, well-run operation will have security update tracking to ensure compliance with relevant patch levels. Management is also who defines policy concerning the political priority of patching vs. development and release dates, staffing levels, etc, all of which have an impact.
They are also the ones who are in a position to negotiate with other teams and their management.
I've also seen the problem of management enforcing application vendor requirements to use, for instance, a specific point revision and "certified" build of Apache so as to not allow security updates.
Ultimately, management is who is responsible for making sure things are working properly and that the company is able to make money, too.
Word to the wise, in such environments, flying off the handle and blaming someone for not doing something they had no permission to do is generally considered bad form. Especially in the cases when we really do want to do the right thing, and put in great amounts effort, of which the actual patching is perhaps 5%, in order to have it get done.