However, here's something I see fairly often. A cop approaches a fairly quiet intersection--as he approaches, he turns on his lights and blips his siren, then blows on through a stop sign or red light--and then turns it off again, as he cruises down the street, quite clearly going to no particular event or issue. (In fact, I've actually seen them 'siren' their way through a red light and pull into a doughnut shop--the ultimate cliche!)
As for pacing, the only problem I have is the distance--five feet?! Regardless of training, that's not a safe distance at anything over 40km/h.
Fundamentally, cops need (and get) exemption from most laws when they're on a call--unfortunately, this gives some of them the feeling that they're completely beyond all traffic laws, and have no responsibility to behave appropriately. And around here at least, they don't get reprimanded unless they hit something/someone.
"OS instability is almost always a case of drivers, and in any case can be expected shortly after a major release."
Not really. All you need to do is make the OS backwards compatible, and suddenly you have a plethora of stable, tested, known drivers at your fingertips. This is how Solaris, HP-UX, AIX, etc. have done it. In many instances, this is how MacOS has behaved over the years (they've made a few 'big jumps' but mostly not). Why can't Windows do this? Why can't Linux? The reason for the former is business, and the reason for the latter is absence of business, as far as I can tell.
The same could be said about pretty much any of the products on the list, but the nod should go to the one that had the greatest impact. Tivo isn't available in many countries, but where other PVRs are the norm, Tivo is still the one that started the revolution, and deserves to be recognised as such.
First of all, understand that I'm looking at this from an outsider's point of view. That said, this is an excellent bill--it provides accountability and a barrier to ballot stuffing, the primary barriers to responsible electronic voting.
The question I have is why not paper ballots?
Much of the rest of the world (Yes, including the first world) uses paper ballots that are tallied by humans. Electronic ballots can only be secure from abuse by having a per-ballot paper trail, so what advantage does the electronic ballot provide at all?
Honestly, I'm curious about why electronic ballots are a good idea at all, given the present state of the art.
Is Epson archival ink actually archival now? A few years ago I was looking at it, and despite their claims, there was a lot of rigorous evidence to the contrary. Have they fixed the problem yet?
The problem is that HP hasn't yet recovered from the blight of Carly Fiorina, and is no longer capable of worthwhile R&D. Still, you're right--either this is a complete scam, or it's a significantly different (and cheaper!) technology than what HP has been using in the Indigo.
Hard to believe that a company without the resources of HP or Xeros can do this, but it'd be neat if they can.
Point taken. I had considered mentioning that type as a separate case. Those are the ones I like to call unemployed, eventually. There are a few other sub-cases as well, but bringing them all up would have led to a treatise on the qualities of a good IT worker, which was far from my original intent.
I read the article, and I've also been peripherally involved with NetQoS' products. Although the premise is fairly straightforward and mostly correct, he makes some insane extrapolations.
Good network engineers, sysadmins, infrastructure support folks, and so forth, don't avoid improving their environments. They usually don't have time to do so, because any down-time from disasters is considered wasteful. In the rare event of time to work on stuff, they're generally so burnt out they don't have time. After nonstop hours (or days!) of fixing emergencies, they often barely have enough energy to slump into their chairs, let alone improve the landscape. Basically, they don't have the time or energy to reduce their workload, except when opportunity presents itself.
Now bad network engineers (etc.) have another problem, and that problem is called tunnel vision. They're incapable of seeing anything other than the immediate task in front of them, so even when the opportunity comes up to truly solve a problem, they duct-tape the broken symptom for the umpteenth time, and end up creating even MORE work for themselves. (And for the rest of their team, not to mention giving users an unrealistic expectation of service.)
In come the productivity enhancing solutions. "Our product will reduce these six disparate reactive monitoring tasks you do now into a single proactive tool." There's a good chance that it will actually do what it says, but only after a test phase, approval, design, rollout (including installing clients on all 400 of your servers), and then tuning. For a medium-to-large scale environment, I'd throw out a rough guess of 9 months, consuming an average of 1/3 of an engineer's time. Given that you're looking at a group of probably 4 people for that environment, that's not insignificant. Still, the company takes a look at it--they bring in a box to build a limited-scope test, and look at it for a few weeks. Those weeks turn into a month and change, and the group realises that the tuning will take a LOT of time afterwards (because extensive tuning isn't part of the proposed rollout scope or timeframe), and ultimately decides to say no.
The vendor's conclusion: These guys would rather put out fires than solve problems.
Not to say that the connection between D&D and IT is invalid, but the firefighting/systemic improvement argument is total crap.
For some years now, a fair number of companies have been trying to figure out a way of using GPLed programs legitimately in their profit-making operations. Some have been successful, some not so much. (And of course, some companies try to get away with using the software withouth regard for the terms of the GPL, but that's another matter.)
It's not easy--the nature of the GPL runs pretty much counter to how businesses generally work. Still, there's a developing history of success with it.
If widely used, the GPL3 will change that. It is explicitly and methodically being written to be as anti-business as possible. Software released under the GPL3 will be avoided by all companies for any purpose. FOSS will become a hobbiest ghetto, with no commercial potential even for commercially useful software.
Of course, this isn't likely to happen. Realistically, MS will continue to fan the flames of confusion over the legal fuzziness of FOSS, slowing down the spread of good software. Rational developers will ignore the GPL3 and continue to write under more reasonable licenses, either GPL2 or CDDL or the like. ESR, RMS, and (strangely enough) Bruce Perens will end up carrying on a verbal pissing match against a vast throng of nobody at all.
The GPL3 is nothing more than the expression of an anti-business diatribe by a handful of increasingly irrelevant nutters. Nobody cares.
Hello, has anyone watched TV lately? It is full of sex. Sex in the bedroom, sex on the couches, naked models painted like oversexed fruit (Honestly, I can't make this stuff up!), and so forth.
What you don't see is nipples and genitalia, because regardless of the context, you know, that'd be BAD!
So you can watch couples faking orgasm between the sheets, but it's apparently not SEX, because any thing so naughty as an ass crack is blurred out. Even better is when they do that for a completely non-sexual context, because: nipples = sex genitals = sex buttocks = sex sex != sex
"The problem is that IT departments are lazy and don't want to learn anything new."
I'm SOOOOOO glad to heard you clear things up for us! And here I thought that I was actually hardworking and open to change if it offered some benefit for the company.
Here's a tip for you: Your anecdotal evidence is (a) quite likely wrong and (b) most certainly not generally true. Some third-rate IT workers are scared of losing their jobs if the technology gets too reliable. GOOD IT workers know that anything which makes their job easier will potentially lead to more time for forward-looking projects, and also know that there will never be a lack of work.
You are a complete idiot, and should not be allowed to touch any technology more advanced than a sea sponge.
OK, I see something like eight articles that have been put up by kdawson, whoever he (or she) is. This one is pretty typical: While spouting off about how the article got the physics wrong (arguable at best), the "editor" failed to notice that the article in question is over six years old!!!
Pathetic, really. It's like a return to the days of Jon Katz.
Speaking as an outsider (although I lived in the US a decade ago) I can definitely say that Bush will leave happily at the end of his term. Partly because he's too much of an idiot for even his masters to stomach (or control), but mostly because a majority of the US still believes they live in a democracy, and the people in charge are going to milk that as long as they can. It just makes good financial sense--why impose martial law when you can just keep stealing elections and get grudging support from the populace? (i.e. "Well he's an idiot, but our country voted him in so I guess we have to agree with him.")
The Democrats will likely win the next election as long as they have enough brains to put Hilary in, although the Republicans might be able to play the 'female president uncertainty' card and bring the totals close enough to rig another one. I doubt that'll happen though, because they can give up four years of officially calling the shots, in order to let the Democrats take the heat for the unholy mess the country is in now, and then step in for three consecutive terms. After that, (and maybe a fourth one--hard to say), then if there's anything left to seize control of, they'll do it--martial law will be VOTED on, and declared.
That is, of course, unless the rest of the world manages to extricate themselves and let the US collapse without taking everyone else down with them. At that point, who knows what'll happen?
I think the idea is that their voting machines have been so badly designed and implemented that the secret rulers are doubting their ability to effectively subvery democracy.
You're right of course. The first thing that came out of my mouth when I read it was, "no shit sherlock!"
BUT, let's not forget that Microsoft is a legal monopoly, and as such are subject to anti-trust acts. This may be something that contravenes allowable behaviour under such laws.
Thanks for your note. I'm going to reply to it rationally (more or less), hoping that you'll see it and take my comments as constructive criticism rather than just ranting.
First of all, I may have misstated Symantec's support policy, but that is verbatim what I was told by a support engineer. (I even have the email to prove it.) Maybe some internal training is in order?
"...you've downplayed the fact that two years is a long time in the world of security software."
Did I mention the "Veritas products" part of the equation? Two years is barely time to get one's feet wet with most (former) Veritas products. There's also support from other vendors to consider: Sun didn't actually provide support for Volume Manager 4.0 until after 4.1 was released. Already we're into the two year window, and we haven't even started a cluster OS upgrade!
And that's just volume manager. A full-blown enterprise Netbackup installation is a MAJOR event. Here's one scenario I dealt with recently:
NBU 5.0 gets released. After six months of waiting for it to become stable enough to actually use, the company started the implementation. This involved $980k of new hardware (and they already had the tape library and infrastructure in place). The planning, architecture, implementation, cutover, and validation took a total of roughly eight months. That's 14 months after initial release, and we've just gone live with the product. At that point, after over a million dollars of gear and time and effort, I am NOT planning on a major version upgrade in ten months or twenty or thirty. I want a MINIMUM of three years of full support after that point, and five is much more reasonable. We shouldn't be forced to upgrade our software until we've outgrown our infrastructure, which is about a 3.5-4.0 year turnaround for most big companies.
Furthermore, service packs or not, the very WEEK that NBU6.0 was released, we were told we couldn't get any more NBU 5.x client licenses. That's it, no more, thanks for coming out. Suddenly, regardless of bug support, we're left without any legal means of growing without upgrading to 6.0.
That is, in a word, crap.
This isn't a $100 anti-virus package for a PC that's going to be chucked in two years, this is software that runs enterprise installations. We don't spend $5k per client machine for a product that's obsolete almost as soon as we install it.
Now you can say that Symantec doesn't operate like this, that my details are all incorrect. That may be, but that is what we were told by our local sales guys, our regional managers, and the other end of Symantec's international support group.
So on the one hand, we have the model you describe, which is crap. On the other hand, you have the reality that I've described which is rancid festering crap. Add to this the fact that VxVM5.0 Basic (the freebie package) simply doesn't work, and you've got a company that is either too incompetent to survive, or trying very hard to destroy the Veritas products/division they bought a few years ago.
As for you liking your job, that's great. I really am genuinely happy every time I hear about someone enjoying their work, because we spend a lot of time at it. (random aside: Until recently I hated my job although I love my work, so I quit--now I'm working for a better company for less pay, and loving it.) Unfortunately, that doesn't change the fact that your company doesn't even have the vaguest understanding of what enterprise computing is really about.
OK, there's no doubt that Turner is pretty incompetent for not fixing this hole with a patch that's been out for most of a year.
But at the same time, I have to ask how incompetent a company that writes security software can be, when their own code is written so as to allow this type of exploit.
Furthermore, I've had quite a bit of experience with Symantec over the past few years. I've been using Veritas products for a decade and change (Netbackup and Volume Manager primarily), and know them very well. Once in a while, I'll come up against a bug and phone Veritas for support/workarounds/whatever. For years they weren't top notch, but they were decent and consistent.
Since Symantec took over, support has fallen through the bottom of the toilet. Their help desk is driven by 'time-to-close,' and actual technical experts are no longer brought in for difficult cases. Bug reports are not even accepted anymore! (Well they'll _take_ the bug report, but won't give you a bug ID to track it with.)
Furthermore, they've started to crank up the version release numbers so that they can promise support for two versions, but only support products for two years from initial release. TWO YEARS FROM RELEASE!!! That's completely unacceptable even in the home PC marketplace, let alone in an enterprise environment, where a product rollout may take over a year.
So yesterday I went to install the newly-free version of Storage Foundation, because I needed to migrate some data from an old system (flawlessly running vxvm 3.5) to a new one, where we'd then move it to ZFS and be done with Veritas for good. The installer put 40 packages on my newly built Solaris 10 system (11/06 release), but failed to actually install the volume manager! After screwing around with it for a while, I gave up and went to uninstall it. The uninstaller hung in kernel space, and for twelve hours did nothing but couldn't be killed.
I don't care about any axes that people have to grind. Symantec is an incompetent company, and DESERVES all of those people holding grudges against them. I'll be glad to see them die horribly.
You've made a good point, but now you're hanging onto it as if it is The Truth, without doubt.
The game is, in fact, a game. In many ways it's a simulation of real life, but not in all ways. Since there are no physical constraints, there are no 'essential' rules. Gravity? Optional. Shipping costs? There's nothing to ship, therefore no costs required.
Having a cost to move things in space and time DOES have economic impact. So does NOT having a cost to move things in time and space. You're deleting one real life modeling variable in your simulation. That makes it less authentic as a simulation, but may in fact lead to such things as the exploration of pure skill-based economies.
Think of it as experiment: What could things be like if we didn't have that constraint? It's worth playing with, even if just for fun.
As much as this stuff is cool, it's going to be insanely expensive to restore data from these guys.
Data integrity and uptime are served by RAID5. If it's not good enough, then it should be backed with mirroring (RAID5+0) or some form of dual-parity RAID (RAID-DP from NetApp, etc.).
But data gets lost or corrupted, even without disk failures. Backups are the place where data recovery is done. DO YOUR BACKUPS!
The very nature of publication rights these days means that your analogy is flawed.
If a paper company gave paper to a newspaper company, told them to fill it up with whatever content they saw fit, and THEN published it without editorial input, then it'd be a fair comparison. Unfortunately, that only happens on the internet. The medium and the publication are synonymous, so the responsibility for the message isn't quite so clear-cut as it has been in the past.
If you don't understand that this is a significant and subtle shift, I'm not sure I really want you to start a business.
Slashdot Mirror
Interesting and informative post--thanks.
However, here's something I see fairly often. A cop approaches a fairly quiet intersection--as he approaches, he turns on his lights and blips his siren, then blows on through a stop sign or red light--and then turns it off again, as he cruises down the street, quite clearly going to no particular event or issue. (In fact, I've actually seen them 'siren' their way through a red light and pull into a doughnut shop--the ultimate cliche!)
As for pacing, the only problem I have is the distance--five feet?! Regardless of training, that's not a safe distance at anything over 40km/h.
Fundamentally, cops need (and get) exemption from most laws when they're on a call--unfortunately, this gives some of them the feeling that they're completely beyond all traffic laws, and have no responsibility to behave appropriately. And around here at least, they don't get reprimanded unless they hit something/someone.
"OS instability is almost always a case of drivers, and in any case can be expected shortly after a major release."
Not really. All you need to do is make the OS backwards compatible, and suddenly you have a plethora of stable, tested, known drivers at your fingertips. This is how Solaris, HP-UX, AIX, etc. have done it. In many instances, this is how MacOS has behaved over the years (they've made a few 'big jumps' but mostly not). Why can't Windows do this? Why can't Linux? The reason for the former is business, and the reason for the latter is absence of business, as far as I can tell.
Damn, that's the best idewa I've heard all day! I'd still like to relocate them out of the control of the US government, though.
The same could be said about pretty much any of the products on the list, but the nod should go to the one that had the greatest impact. Tivo isn't available in many countries, but where other PVRs are the norm, Tivo is still the one that started the revolution, and deserves to be recognised as such.
First of all, understand that I'm looking at this from an outsider's point of view. That said, this is an excellent bill--it provides accountability and a barrier to ballot stuffing, the primary barriers to responsible electronic voting.
The question I have is why not paper ballots?
Much of the rest of the world (Yes, including the first world) uses paper ballots that are tallied by humans. Electronic ballots can only be secure from abuse by having a per-ballot paper trail, so what advantage does the electronic ballot provide at all?
Honestly, I'm curious about why electronic ballots are a good idea at all, given the present state of the art.
Is Epson archival ink actually archival now? A few years ago I was looking at it, and despite their claims, there was a lot of rigorous evidence to the contrary. Have they fixed the problem yet?
The problem is that HP hasn't yet recovered from the blight of Carly Fiorina, and is no longer capable of worthwhile R&D. Still, you're right--either this is a complete scam, or it's a significantly different (and cheaper!) technology than what HP has been using in the Indigo.
Hard to believe that a company without the resources of HP or Xeros can do this, but it'd be neat if they can.
Point taken. I had considered mentioning that type as a separate case. Those are the ones I like to call unemployed, eventually. There are a few other sub-cases as well, but bringing them all up would have led to a treatise on the qualities of a good IT worker, which was far from my original intent.
I read the article, and I've also been peripherally involved with NetQoS' products. Although the premise is fairly straightforward and mostly correct, he makes some insane extrapolations.
Good network engineers, sysadmins, infrastructure support folks, and so forth, don't avoid improving their environments. They usually don't have time to do so, because any down-time from disasters is considered wasteful. In the rare event of time to work on stuff, they're generally so burnt out they don't have time. After nonstop hours (or days!) of fixing emergencies, they often barely have enough energy to slump into their chairs, let alone improve the landscape. Basically, they don't have the time or energy to reduce their workload, except when opportunity presents itself.
Now bad network engineers (etc.) have another problem, and that problem is called tunnel vision. They're incapable of seeing anything other than the immediate task in front of them, so even when the opportunity comes up to truly solve a problem, they duct-tape the broken symptom for the umpteenth time, and end up creating even MORE work for themselves. (And for the rest of their team, not to mention giving users an unrealistic expectation of service.)
In come the productivity enhancing solutions. "Our product will reduce these six disparate reactive monitoring tasks you do now into a single proactive tool." There's a good chance that it will actually do what it says, but only after a test phase, approval, design, rollout (including installing clients on all 400 of your servers), and then tuning. For a medium-to-large scale environment, I'd throw out a rough guess of 9 months, consuming an average of 1/3 of an engineer's time. Given that you're looking at a group of probably 4 people for that environment, that's not insignificant. Still, the company takes a look at it--they bring in a box to build a limited-scope test, and look at it for a few weeks. Those weeks turn into a month and change, and the group realises that the tuning will take a LOT of time afterwards (because extensive tuning isn't part of the proposed rollout scope or timeframe), and ultimately decides to say no.
The vendor's conclusion: These guys would rather put out fires than solve problems.
Not to say that the connection between D&D and IT is invalid, but the firefighting/systemic improvement argument is total crap.
For some years now, a fair number of companies have been trying to figure out a way of using GPLed programs legitimately in their profit-making operations. Some have been successful, some not so much. (And of course, some companies try to get away with using the software withouth regard for the terms of the GPL, but that's another matter.)
It's not easy--the nature of the GPL runs pretty much counter to how businesses generally work. Still, there's a developing history of success with it.
If widely used, the GPL3 will change that. It is explicitly and methodically being written to be as anti-business as possible. Software released under the GPL3 will be avoided by all companies for any purpose. FOSS will become a hobbiest ghetto, with no commercial potential even for commercially useful software.
Of course, this isn't likely to happen. Realistically, MS will continue to fan the flames of confusion over the legal fuzziness of FOSS, slowing down the spread of good software. Rational developers will ignore the GPL3 and continue to write under more reasonable licenses, either GPL2 or CDDL or the like. ESR, RMS, and (strangely enough) Bruce Perens will end up carrying on a verbal pissing match against a vast throng of nobody at all.
The GPL3 is nothing more than the expression of an anti-business diatribe by a handful of increasingly irrelevant nutters. Nobody cares.
Hello, has anyone watched TV lately? It is full of sex. Sex in the bedroom, sex on the couches, naked models painted like oversexed fruit (Honestly, I can't make this stuff up!), and so forth.
What you don't see is nipples and genitalia, because regardless of the context, you know, that'd be BAD!
So you can watch couples faking orgasm between the sheets, but it's apparently not SEX, because any thing so naughty as an ass crack is blurred out. Even better is when they do that for a completely non-sexual context, because:
nipples = sex
genitals = sex
buttocks = sex
sex != sex
Screwed up country.
What genre is that? The "complete crap that didn't die when it should have" genre?
Every fighting game out there has sucked, because the genre sucks. It's like the gaming equivalent of rap music.
"The problem is that IT departments are lazy and don't want to learn anything new."
I'm SOOOOOO glad to heard you clear things up for us! And here I thought that I was actually hardworking and open to change if it offered some benefit for the company.
Here's a tip for you: Your anecdotal evidence is (a) quite likely wrong and (b) most certainly not generally true. Some third-rate IT workers are scared of losing their jobs if the technology gets too reliable. GOOD IT workers know that anything which makes their job easier will potentially lead to more time for forward-looking projects, and also know that there will never be a lack of work.
You are a complete idiot, and should not be allowed to touch any technology more advanced than a sea sponge.
x64 is an industry-common term for AMD's 64-bit processor architecture extensions. Nobody, including Microsoft, refers to Intel architecture as x64.
My thoughts exactly. Mind you, the ACPI standard still hasn't actually stabilised--the last update to it was in October (after seven years!).
Still, I would assume backwards compatibility. No scratch that--I would expect it, but wouldn't count on it.
OK, I see something like eight articles that have been put up by kdawson, whoever he (or she) is. This one is pretty typical: While spouting off about how the article got the physics wrong (arguable at best), the "editor" failed to notice that the article in question is over six years old!!!
Pathetic, really. It's like a return to the days of Jon Katz.
"Don't you care that Diebold's trying to destroy all mankind? Huh?"
"But they're so BAD at it."
(With apologies to Invader Zim.)
Speaking as an outsider (although I lived in the US a decade ago) I can definitely say that Bush will leave happily at the end of his term. Partly because he's too much of an idiot for even his masters to stomach (or control), but mostly because a majority of the US still believes they live in a democracy, and the people in charge are going to milk that as long as they can. It just makes good financial sense--why impose martial law when you can just keep stealing elections and get grudging support from the populace? (i.e. "Well he's an idiot, but our country voted him in so I guess we have to agree with him.")
The Democrats will likely win the next election as long as they have enough brains to put Hilary in, although the Republicans might be able to play the 'female president uncertainty' card and bring the totals close enough to rig another one. I doubt that'll happen though, because they can give up four years of officially calling the shots, in order to let the Democrats take the heat for the unholy mess the country is in now, and then step in for three consecutive terms. After that, (and maybe a fourth one--hard to say), then if there's anything left to seize control of, they'll do it--martial law will be VOTED on, and declared.
That is, of course, unless the rest of the world manages to extricate themselves and let the US collapse without taking everyone else down with them. At that point, who knows what'll happen?
I think the idea is that their voting machines have been so badly designed and implemented that the secret rulers are doubting their ability to effectively subvery democracy.
You're right of course. The first thing that came out of my mouth when I read it was, "no shit sherlock!"
BUT, let's not forget that Microsoft is a legal monopoly, and as such are subject to anti-trust acts. This may be something that contravenes allowable behaviour under such laws.
Thanks for your note. I'm going to reply to it rationally (more or less), hoping that you'll see it and take my comments as constructive criticism rather than just ranting.
First of all, I may have misstated Symantec's support policy, but that is verbatim what I was told by a support engineer. (I even have the email to prove it.) Maybe some internal training is in order?
"...you've downplayed the fact that two years is a long time in the world of security software."
Did I mention the "Veritas products" part of the equation? Two years is barely time to get one's feet wet with most (former) Veritas products. There's also support from other vendors to consider: Sun didn't actually provide support for Volume Manager 4.0 until after 4.1 was released. Already we're into the two year window, and we haven't even started a cluster OS upgrade!
And that's just volume manager. A full-blown enterprise Netbackup installation is a MAJOR event. Here's one scenario I dealt with recently:
NBU 5.0 gets released. After six months of waiting for it to become stable enough to actually use, the company started the implementation. This involved $980k of new hardware (and they already had the tape library and infrastructure in place). The planning, architecture, implementation, cutover, and validation took a total of roughly eight months. That's 14 months after initial release, and we've just gone live with the product. At that point, after over a million dollars of gear and time and effort, I am NOT planning on a major version upgrade in ten months or twenty or thirty. I want a MINIMUM of three years of full support after that point, and five is much more reasonable. We shouldn't be forced to upgrade our software until we've outgrown our infrastructure, which is about a 3.5-4.0 year turnaround for most big companies.
Furthermore, service packs or not, the very WEEK that NBU6.0 was released, we were told we couldn't get any more NBU 5.x client licenses. That's it, no more, thanks for coming out. Suddenly, regardless of bug support, we're left without any legal means of growing without upgrading to 6.0.
That is, in a word, crap.
This isn't a $100 anti-virus package for a PC that's going to be chucked in two years, this is software that runs enterprise installations. We don't spend $5k per client machine for a product that's obsolete almost as soon as we install it.
Now you can say that Symantec doesn't operate like this, that my details are all incorrect. That may be, but that is what we were told by our local sales guys, our regional managers, and the other end of Symantec's international support group.
So on the one hand, we have the model you describe, which is crap. On the other hand, you have the reality that I've described which is rancid festering crap. Add to this the fact that VxVM5.0 Basic (the freebie package) simply doesn't work, and you've got a company that is either too incompetent to survive, or trying very hard to destroy the Veritas products/division they bought a few years ago.
As for you liking your job, that's great. I really am genuinely happy every time I hear about someone enjoying their work, because we spend a lot of time at it. (random aside: Until recently I hated my job although I love my work, so I quit--now I'm working for a better company for less pay, and loving it.) Unfortunately, that doesn't change the fact that your company doesn't even have the vaguest understanding of what enterprise computing is really about.
OK, there's no doubt that Turner is pretty incompetent for not fixing this hole with a patch that's been out for most of a year.
But at the same time, I have to ask how incompetent a company that writes security software can be, when their own code is written so as to allow this type of exploit.
Furthermore, I've had quite a bit of experience with Symantec over the past few years. I've been using Veritas products for a decade and change (Netbackup and Volume Manager primarily), and know them very well. Once in a while, I'll come up against a bug and phone Veritas for support/workarounds/whatever. For years they weren't top notch, but they were decent and consistent.
Since Symantec took over, support has fallen through the bottom of the toilet. Their help desk is driven by 'time-to-close,' and actual technical experts are no longer brought in for difficult cases. Bug reports are not even accepted anymore! (Well they'll _take_ the bug report, but won't give you a bug ID to track it with.)
Furthermore, they've started to crank up the version release numbers so that they can promise support for two versions, but only support products for two years from initial release. TWO YEARS FROM RELEASE!!! That's completely unacceptable even in the home PC marketplace, let alone in an enterprise environment, where a product rollout may take over a year.
So yesterday I went to install the newly-free version of Storage Foundation, because I needed to migrate some data from an old system (flawlessly running vxvm 3.5) to a new one, where we'd then move it to ZFS and be done with Veritas for good. The installer put 40 packages on my newly built Solaris 10 system (11/06 release), but failed to actually install the volume manager! After screwing around with it for a while, I gave up and went to uninstall it. The uninstaller hung in kernel space, and for twelve hours did nothing but couldn't be killed.
I don't care about any axes that people have to grind. Symantec is an incompetent company, and DESERVES all of those people holding grudges against them. I'll be glad to see them die horribly.
You've made a good point, but now you're hanging onto it as if it is The Truth, without doubt.
The game is, in fact, a game. In many ways it's a simulation of real life, but not in all ways. Since there are no physical constraints, there are no 'essential' rules. Gravity? Optional. Shipping costs? There's nothing to ship, therefore no costs required.
Having a cost to move things in space and time DOES have economic impact. So does NOT having a cost to move things in time and space. You're deleting one real life modeling variable in your simulation. That makes it less authentic as a simulation, but may in fact lead to such things as the exploration of pure skill-based economies.
Think of it as experiment: What could things be like if we didn't have that constraint? It's worth playing with, even if just for
fun.
As much as this stuff is cool, it's going to be insanely expensive to restore data from these guys.
Data integrity and uptime are served by RAID5. If it's not good enough, then it should be backed with mirroring (RAID5+0) or some form of dual-parity RAID (RAID-DP from NetApp, etc.).
But data gets lost or corrupted, even without disk failures. Backups are the place where data recovery is done. DO YOUR BACKUPS!
The very nature of publication rights these days means that your analogy is flawed.
If a paper company gave paper to a newspaper company, told them to fill it up with whatever content they saw fit, and THEN published it without editorial input, then it'd be a fair comparison. Unfortunately, that only happens on the internet. The medium and the publication are synonymous, so the responsibility for the message isn't quite so clear-cut as it has been in the past.
If you don't understand that this is a significant and subtle shift, I'm not sure I really want you to start a business.