I agree with you completely. However, I think the "avalon will be available for 2k/xp" is patently false.
Companies like like this all the time: promise a backward-compatible feature, but then claim that their old product isn't good enough to work with the new feature, thus giving all the people that never would have upgraded in the first place but were looking forward to the feature a much larger incentive to upgrade: they'd already invested in the idea of avalon, so they will buy it.
Where the exploit is doesn't define the severity, but what is possible through the exploit. For instance, there was a "kernel exploit" months back which was essentially a DoS problem in the kernel. Most people didn't even worry about it because it was so benign (in the grand scale) and so unlikely to happen that it wasn't worth the effort.
Look at the system-compromising exploits, on a grander scale. Most kernel exploits aren't possible without the assistance of a specific version of a daemon or application and most normal exploits in Linux don't lead to a root exploit - or "system compromising exploit". On the other hand, most Windows exploits - while "kernel" exploits or not, lead to a complete compromise in the system's integrity.
Besides, this entire issue is completely blindsided by the plethora of spyware that is on damned near every Windows system out there. IE itself is pretty much a massive security hole which can only be patched by its removal (or complete re-engineering).
I'd believe it, too. The only thing this office did was (large) 2D architectural stuff, so the demand for that kind of thing wasn't really there. Though, given the bloat of some of the CAD stuff out there (AutoCAD 98 - 2002 in this case), it's no small surprise that a good cpu and lots of ram was necessary...
Don't be rediculous. There's not an organization in the world that has a single 100,000 PC department, branch, or general "group of PCs under one IT dept's responsibility" in the world. If there is, it's a massive and likely painfully unmanageable group, as its likely to have around 500 servers and thousands of switches, routers, and relays.
No, companies split up their responsibilites by department. That's why you'll hear htings like "Marketing Department IT". Not only do they do incrimental changes within these departments, but as an organization as a whole. It's a given that you'll have a 20:80 ratio of new to old stuff on a given year (give or take, depending on your cyclical planning).
And on a final note: Tier-1 vendors don't dick their customers around like that. Tier-1 companies realize that customers won't be dicked around - they'll just go to another provider that won't shove a slippery one up their asses.
So maybe it's just a good business practice to not buy from Dell?
I'm fairly certain IBM doesn't do this. Their up-front cost might be marginally higher, but you reap the benefits through superior product, service, etc. - that, and their hardware doesn't suck. Did I mention IBM makes good hardware?
Sure, but that support contract only covers the hardware. If the user manages to screw up their system it's the company support staff that is going to have to re-install the OS, drivers and apps.
I don't see your point. If you're managing 300+ systems, you're, well, managing them. Chances are you sent the OEM provider a disk image which you designed yourself for specific use on those systems (well, maybe not with 300 systems, but with, say, 1000 systems -definately). Reimage the damned thing and you're back where you were before things broke. This is no different than buying the 'expensive' system.
But the point is, if you've already got an existing system install, and the "business optimized" (whatever) systems are not the same model as what you've currently got, what's it matter if you get 300 systems that are cheaper and powerful, or 300 that are older and cost more? Well, it does. It just shouldn't matter that there's "business optimized" PCs, as any level headeded person is going to get a support contract for those systems guaranteeing $years of support for that specific model.
I know this is how IBM does it, as I currently work for a large non-profit which relies largely on such contracts. The machines aren't by any means "top of the line" but they're not gimps and reasonably priced for the spec.
Complete and utter bullshit. PC parts are, by very nature, COTS equipment - as the military would say.
Now, it might be the case that there are some systems which system OEMs produce which have more intensively tested drivers and things of that sort, but when you buy from someone like Dell for a business, you're still getting raked over the coals.
I ran a small office's network (3-13 employees) from 1998 until just recently this past month. The first systems purchased were "business optimized" systems and were well under "top of the line" spec at a much higher cost, as the grandparent mentioned (poor SOB). They had to be replaced within a year due to being too slow to actually get work done on, and Windows 95 was no longer being supported by some of the software vendors the company used. So we went to "custom" PCs and managed to keep a fairly uniform hardware ecosystem while getting much more bang for the buck (CAD systems) and being able to pick our hardware specifics (3com network cards, ATI video, etc.,) so we wouldn't have an outdated piece of shit in 2 years that needed significantly new hardware to run the latest OS.
Now, while I can understand a large company's situation is definately a bit different, it's not true at all that you need to compromise on quality for price. As the grandparent post explained, "business optimized" isn't always going to be better in terms of hardware is 'standard'. Still, it doesn't seem like a wise choice from where I'm sitting, regardless of reasons for "business optimized" PCs - they're still Windows PCs, and the technology is (unfortunately) so quickly relegated to the dumpster that it's just asinine for such a consideration to be serious. Now, if you had something like Sun workstations or Linux machines, which can run for many a year with pretty much the same software without problems, I could understand...
Besides, anyone ordering as many as 300 systems sure as hell better get a support contract for them as well. Doing otherwise would be beyond foolish.
I swear I read this post on slashdot (and/or possibly elsewhere) a couple years ago the last time the topic of the kernel forking came up. Did you steal it from someone else, or was it your post originally?
Even if there is a major fork from a distro such as RedHat, I can only see two possible scenarios, or possibly a combination of two:
The distro that forks gets major ostricization and criticism from the community until they back out. Even still, it's unlikely that it would be a "longhorn" style fork, where they obsolete anything.
A distro forks and produces something quite superior to what is currently used. Within a couple weeks, those changes are all rolled into not only the main kernel tree, but various distro kernels as patches for short-term utilization of the new feature.
Remember, this is open source software. It's (legally) impossible for someone to pull a gates on the community and get away with it. There's no incentive to pull something which would lock other vendors out because by the bindings of the license they've agreed to (GPL), they have to disclose what they did. Not only that, but they would be majorly reamed by the OSS community and lose all semblance of trust.
That 'mainframe' process is still common practice with IBM. When everyone's dual processor systems got cheaper, their's stayed the same and they just made the dual proc a quad proc, wiht the 2nd pair disabled until you call up and let them know they're needed.
I'll agree with you, re: Tolkien, actually. After a little bit more thought on the matter, I came to the conclusion that indeed, much of Tolkien's writing was "boring", in that nothing really happened during that frame and all that occured was a lot of description.
That's part of what I like about Tolkien, personally: he described a lot of fictional things, creating them for the first time. Not only that, but he described things well, and I could imagine myself there. Yes, much of his work has plot, characters, and development, and reality probably is that the "story" could have been condensed down to a single book instead of 3. However, I feel that the difference between what he did write and a book which might have resulted from such condensement is like the difference between (say) Star Wars (where you've got an in-depth sprawling "world" for the story to take place in, but which is not specifically mentioned) and a no-name production which has been forgotten: the backdrop is set for the story, and motivations are made apparent. The problem Tolkien had was that he couldn't display everything on screen - he had to describe it all. (I wonder if he'd have wanted to make his work into movies, had he the opportunity...)
The joy of Tolkien, IMO, is not so much his work, but the imagination involved; a person could go on for days after reading his book with their head still in the world he created (I did/have/do(?)). And, being as his work is very firmly a "fantasy" genre, it works.
Those comments were likely written by the people I had an English writing "lab" with when I was still trying to decide what to major in. Most of my classmates were Junior or Senior class standing, but none of them could write to save their lives.
I dropped out of the class roughly 1/3rd of the way through it as I could no longer take the stupidity. Everyone in the class had to write a 5-10 page short story from (IIRC - this occured about 3 years ago) the first person perspective. I wrote this short story - or rather, this portion of the short story. I never got it done, as I'd started it at around 2am the night prior to it being due, and the class was at 8am.
Regardless of my story's completion, I handed 6 copies of my it out the next morning to the others in the class. In return, I got 6 copies of their respective stories: abusive nonsense the lot of them. There was one that barely resembled coherrent sentences. There was another that lacked any sort of point whatsoever - such as plot, meaningful characters, or use of words that didn't opitomize the typical Valleygirl vocabulary. I've managed to blot the rest from my mind through excessive drug abuse.
The next class resulted in half a dozen people comopletely misunderstanding everything about what was going on. Some didn't realize that the narrator was not "Ed" (the focus of the (partial) story, if you've not yet read it); some made asinine comments illustrating their inability to understand how one sentence follows the one proceeding it, eventually forming a single concept - something most of us know as a "paragraph". Now, granted, the story isn't air tight and isn't even complete, but if you've read it you'll realize that it's a decent enough story. These pedants couldn't see past the stars in their eyes of writting Xena: Princess Warrior scripts (which, if i recall correctly, one of the students mentioned as one of their main motivators into becoming an English major to write).
There was one guy that had good ideas, good characters, and good development, but his ability to form properly written text was a bit stunted. A shame. When he read his stories (as we all had to do) he seemed to accomidate for this negligence, though. But he also never attended class, so...
Good science fiction is just like good writing of any variety. It's an almost mathematical formula that will work regardless of the actual topic (well, almost), earning you a place in publishing history. In essence, an author is trying to tell a story.
Story = plot + characters + development;
If you can't make a person care a damn about any one of those things, your book or story is worthless. Good stories make people care about all three (LotR, for instance), while many just adhere to a single one or two (Dan Brown's writing somewhat springs to mind - he seems to focus most on development, with overall plot being minimalist and characters barely existing:P). Asimov seems to have captured both development and plot perfectly, with a bit of sparseness in the area of characters.
Of course, this is just one amateur author and avid reader's opinion.
I can tell you that as a former English major, most of what is produced by English majors is worthy of contempt and completely void of anything approaching literature. Mostly, it's not worthy for pulp.
I'm not saying this is true for your writing, of course. I may have just encountered a fairly vile cluster of English majors. They weren't even decent human beings, being fairly void of general intellect and interest. No, I'm sipmly saying that it's no small surprise that most writers are fortunate that their writing gets recognition because it is at best, indeed, a ream full of fire starting material. Nothing less than a lucky spell could provide them recognition.
Yes, same for me. I didn't want to even look at them at all, but my curiosity to see how well they photoshopped won me over. But now they're haunting me.
It was almost as if there was someone else in the room with me when I was looking at them. Somewhat like I imagine it would be like living in a house where you know someone else died previously, or the feeling you get when you're all alone in a big, old house.
*shiver* Now I just want to kill these rapists and child pornographers all the more.
Bullshit. This is like having to wear condoms, because the vasectomy you got didn't work despite a 100% gaurantee that it would. (Not osmething that would happen, but just go with the analogy).
So yes, it's responsible to use condoms in such a situation, but this in no way alieviates the person that was supposed to do their job properly, and didn't (Doctor|MS).
How is it responsible to take responsibility for others' faults? Were you the 3rd child in your family or something, and got blamed for everything the third child did? That's just crazy talk, man.
What you say makes sense to me. I've not really thought about it much before, but after reading your post I started thinking about how devices are used and referred to, what you said is pretty clear.
I wouldn't necessarily agree that it's all that terribly "broken" but I do agree that some more thought should probably be put into redesigning the different device chains, based on my (limited) knowledge of how they work. Maybe simply have a universal chain which any device can use, and then have modularity to it so any bus type could use it.
But I digress, that's not very monolithic, now is it.:P It does seem to be the best, most efficient way to do it, though. Things like CF/IDE/PCMCIA/USB interaction is all quite a bit kludged at the moment. The CF card -> USB -> cardbus/yenta_socket ordeal springs to mind in addition to your comment on SCSI IDE emulation...
I get the feeling that HURD is going to end up being to operating systems what Enlightenment is to window managers: every time it starts to look promising for non-development use, it's scrapped for a rewrite and all the functionality is detroyed due to redesign of the lower-level stuff.
Oh, quite the opposite. Some comic strips are designed from the ground up as advertising platforms. They don't have any humor value for anyone over 6, and yet still manage to at least make some money with the whorish accessorizing.
Maybe some people like it because it's cute (little old women with 12 cats?), but I can't see how anyone with a wit of humor would find it humorous.
Slashdot Mirror
I agree with you completely. However, I think the "avalon will be available for 2k/xp" is patently false.
Companies like like this all the time: promise a backward-compatible feature, but then claim that their old product isn't good enough to work with the new feature, thus giving all the people that never would have upgraded in the first place but were looking forward to the feature a much larger incentive to upgrade: they'd already invested in the idea of avalon, so they will buy it.
Wow, that's ignorant.
Where the exploit is doesn't define the severity, but what is possible through the exploit. For instance, there was a "kernel exploit" months back which was essentially a DoS problem in the kernel. Most people didn't even worry about it because it was so benign (in the grand scale) and so unlikely to happen that it wasn't worth the effort.
Look at the system-compromising exploits, on a grander scale. Most kernel exploits aren't possible without the assistance of a specific version of a daemon or application and most normal exploits in Linux don't lead to a root exploit - or "system compromising exploit". On the other hand, most Windows exploits - while "kernel" exploits or not, lead to a complete compromise in the system's integrity.
Besides, this entire issue is completely blindsided by the plethora of spyware that is on damned near every Windows system out there. IE itself is pretty much a massive security hole which can only be patched by its removal (or complete re-engineering).
I'd believe it, too. The only thing this office did was (large) 2D architectural stuff, so the demand for that kind of thing wasn't really there. Though, given the bloat of some of the CAD stuff out there (AutoCAD 98 - 2002 in this case), it's no small surprise that a good cpu and lots of ram was necessary...
Don't be rediculous. There's not an organization in the world that has a single 100,000 PC department, branch, or general "group of PCs under one IT dept's responsibility" in the world. If there is, it's a massive and likely painfully unmanageable group, as its likely to have around 500 servers and thousands of switches, routers, and relays.
No, companies split up their responsibilites by department. That's why you'll hear htings like "Marketing Department IT". Not only do they do incrimental changes within these departments, but as an organization as a whole. It's a given that you'll have a 20:80 ratio of new to old stuff on a given year (give or take, depending on your cyclical planning).
And on a final note: Tier-1 vendors don't dick their customers around like that. Tier-1 companies realize that customers won't be dicked around - they'll just go to another provider that won't shove a slippery one up their asses.
So maybe it's just a good business practice to not buy from Dell?
I'm fairly certain IBM doesn't do this. Their up-front cost might be marginally higher, but you reap the benefits through superior product, service, etc. - that, and their hardware doesn't suck. Did I mention IBM makes good hardware?
Sure, but that support contract only covers the hardware. If the user manages to screw up their system it's the company support staff that is going to have to re-install the OS, drivers and apps.
I don't see your point. If you're managing 300+ systems, you're, well, managing them. Chances are you sent the OEM provider a disk image which you designed yourself for specific use on those systems (well, maybe not with 300 systems, but with, say, 1000 systems -definately). Reimage the damned thing and you're back where you were before things broke. This is no different than buying the 'expensive' system.
But the point is, if you've already got an existing system install, and the "business optimized" (whatever) systems are not the same model as what you've currently got, what's it matter if you get 300 systems that are cheaper and powerful, or 300 that are older and cost more? Well, it does. It just shouldn't matter that there's "business optimized" PCs, as any level headeded person is going to get a support contract for those systems guaranteeing $years of support for that specific model.
I know this is how IBM does it, as I currently work for a large non-profit which relies largely on such contracts. The machines aren't by any means "top of the line" but they're not gimps and reasonably priced for the spec.
That seems almost counter-intuitive from where I sit. I'd think that a home user would be less likely to splurge than a small business.
Complete and utter bullshit. PC parts are, by very nature, COTS equipment - as the military would say.
Now, it might be the case that there are some systems which system OEMs produce which have more intensively tested drivers and things of that sort, but when you buy from someone like Dell for a business, you're still getting raked over the coals.
I ran a small office's network (3-13 employees) from 1998 until just recently this past month. The first systems purchased were "business optimized" systems and were well under "top of the line" spec at a much higher cost, as the grandparent mentioned (poor SOB). They had to be replaced within a year due to being too slow to actually get work done on, and Windows 95 was no longer being supported by some of the software vendors the company used. So we went to "custom" PCs and managed to keep a fairly uniform hardware ecosystem while getting much more bang for the buck (CAD systems) and being able to pick our hardware specifics (3com network cards, ATI video, etc.,) so we wouldn't have an outdated piece of shit in 2 years that needed significantly new hardware to run the latest OS.
Now, while I can understand a large company's situation is definately a bit different, it's not true at all that you need to compromise on quality for price. As the grandparent post explained, "business optimized" isn't always going to be better in terms of hardware is 'standard'. Still, it doesn't seem like a wise choice from where I'm sitting, regardless of reasons for "business optimized" PCs - they're still Windows PCs, and the technology is (unfortunately) so quickly relegated to the dumpster that it's just asinine for such a consideration to be serious. Now, if you had something like Sun workstations or Linux machines, which can run for many a year with pretty much the same software without problems, I could understand...
Besides, anyone ordering as many as 300 systems sure as hell better get a support contract for them as well. Doing otherwise would be beyond foolish.
Meanwhile, you can get the same memory from crucial.com for less than half that price, complete with a lifetime warranty...
I think that's probably the most appropriate and insightful comment on this thread.
Well done.
I swear I read this post on slashdot (and/or possibly elsewhere) a couple years ago the last time the topic of the kernel forking came up. Did you steal it from someone else, or was it your post originally?
The distro that forks gets major ostricization and criticism from the community until they back out. Even still, it's unlikely that it would be a "longhorn" style fork, where they obsolete anything.
A distro forks and produces something quite superior to what is currently used. Within a couple weeks, those changes are all rolled into not only the main kernel tree, but various distro kernels as patches for short-term utilization of the new feature.
Remember, this is open source software. It's (legally) impossible for someone to pull a gates on the community and get away with it. There's no incentive to pull something which would lock other vendors out because by the bindings of the license they've agreed to (GPL), they have to disclose what they did. Not only that, but they would be majorly reamed by the OSS community and lose all semblance of trust.
Because it reduces the cost of production while increasing the features (and thus the price they're able to charge for the things) within the vehicle.
That 'mainframe' process is still common practice with IBM. When everyone's dual processor systems got cheaper, their's stayed the same and they just made the dual proc a quad proc, wiht the 2nd pair disabled until you call up and let them know they're needed.
I'll agree with you, re: Tolkien, actually. After a little bit more thought on the matter, I came to the conclusion that indeed, much of Tolkien's writing was "boring", in that nothing really happened during that frame and all that occured was a lot of description.
That's part of what I like about Tolkien, personally: he described a lot of fictional things, creating them for the first time. Not only that, but he described things well, and I could imagine myself there. Yes, much of his work has plot, characters, and development, and reality probably is that the "story" could have been condensed down to a single book instead of 3. However, I feel that the difference between what he did write and a book which might have resulted from such condensement is like the difference between (say) Star Wars (where you've got an in-depth sprawling "world" for the story to take place in, but which is not specifically mentioned) and a no-name production which has been forgotten: the backdrop is set for the story, and motivations are made apparent. The problem Tolkien had was that he couldn't display everything on screen - he had to describe it all. (I wonder if he'd have wanted to make his work into movies, had he the opportunity...)
The joy of Tolkien, IMO, is not so much his work, but the imagination involved; a person could go on for days after reading his book with their head still in the world he created (I did/have/do(?)). And, being as his work is very firmly a "fantasy" genre, it works.
Those comments were likely written by the people I had an English writing "lab" with when I was still trying to decide what to major in. Most of my classmates were Junior or Senior class standing, but none of them could write to save their lives.
I dropped out of the class roughly 1/3rd of the way through it as I could no longer take the stupidity. Everyone in the class had to write a 5-10 page short story from (IIRC - this occured about 3 years ago) the first person perspective. I wrote this short story - or rather, this portion of the short story. I never got it done, as I'd started it at around 2am the night prior to it being due, and the class was at 8am.
Regardless of my story's completion, I handed 6 copies of my it out the next morning to the others in the class. In return, I got 6 copies of their respective stories: abusive nonsense the lot of them. There was one that barely resembled coherrent sentences. There was another that lacked any sort of point whatsoever - such as plot, meaningful characters, or use of words that didn't opitomize the typical Valleygirl vocabulary. I've managed to blot the rest from my mind through excessive drug abuse.
The next class resulted in half a dozen people comopletely misunderstanding everything about what was going on. Some didn't realize that the narrator was not "Ed" (the focus of the (partial) story, if you've not yet read it); some made asinine comments illustrating their inability to understand how one sentence follows the one proceeding it, eventually forming a single concept - something most of us know as a "paragraph". Now, granted, the story isn't air tight and isn't even complete, but if you've read it you'll realize that it's a decent enough story. These pedants couldn't see past the stars in their eyes of writting Xena: Princess Warrior scripts (which, if i recall correctly, one of the students mentioned as one of their main motivators into becoming an English major to write).
There was one guy that had good ideas, good characters, and good development, but his ability to form properly written text was a bit stunted. A shame. When he read his stories (as we all had to do) he seemed to accomidate for this negligence, though. But he also never attended class, so...
Good science fiction is just like good writing of any variety. It's an almost mathematical formula that will work regardless of the actual topic (well, almost), earning you a place in publishing history. In essence, an author is trying to tell a story.
:P). Asimov seems to have captured both development and plot perfectly, with a bit of sparseness in the area of characters.
Story = plot + characters + development;
If you can't make a person care a damn about any one of those things, your book or story is worthless. Good stories make people care about all three (LotR, for instance), while many just adhere to a single one or two (Dan Brown's writing somewhat springs to mind - he seems to focus most on development, with overall plot being minimalist and characters barely existing
Of course, this is just one amateur author and avid reader's opinion.
I can tell you that as a former English major, most of what is produced by English majors is worthy of contempt and completely void of anything approaching literature. Mostly, it's not worthy for pulp.
I'm not saying this is true for your writing, of course. I may have just encountered a fairly vile cluster of English majors. They weren't even decent human beings, being fairly void of general intellect and interest. No, I'm sipmly saying that it's no small surprise that most writers are fortunate that their writing gets recognition because it is at best, indeed, a ream full of fire starting material. Nothing less than a lucky spell could provide them recognition.
Dear sweet mother of God! That's so wretched!
I've seen better writing from bad high school students. Though, on the flip side, it's likely on par with most of the romance novels out there.
Oh, what suffering.
Yes, same for me. I didn't want to even look at them at all, but my curiosity to see how well they photoshopped won me over. But now they're haunting me.
It was almost as if there was someone else in the room with me when I was looking at them. Somewhat like I imagine it would be like living in a house where you know someone else died previously, or the feeling you get when you're all alone in a big, old house.
*shiver* Now I just want to kill these rapists and child pornographers all the more.
Bullshit. This is like having to wear condoms, because the vasectomy you got didn't work despite a 100% gaurantee that it would. (Not osmething that would happen, but just go with the analogy).
So yes, it's responsible to use condoms in such a situation, but this in no way alieviates the person that was supposed to do their job properly, and didn't (Doctor|MS).
How is it responsible to take responsibility for others' faults? Were you the 3rd child in your family or something, and got blamed for everything the third child did? That's just crazy talk, man.
(I'm not this post's grandparent, btw)
:P It does seem to be the best, most efficient way to do it, though. Things like CF/IDE/PCMCIA/USB interaction is all quite a bit kludged at the moment. The CF card -> USB -> cardbus/yenta_socket ordeal springs to mind in addition to your comment on SCSI IDE emulation...
What you say makes sense to me. I've not really thought about it much before, but after reading your post I started thinking about how devices are used and referred to, what you said is pretty clear.
I wouldn't necessarily agree that it's all that terribly "broken" but I do agree that some more thought should probably be put into redesigning the different device chains, based on my (limited) knowledge of how they work. Maybe simply have a universal chain which any device can use, and then have modularity to it so any bus type could use it.
But I digress, that's not very monolithic, now is it.
I get the feeling that HURD is going to end up being to operating systems what Enlightenment is to window managers: every time it starts to look promising for non-development use, it's scrapped for a rewrite and all the functionality is detroyed due to redesign of the lower-level stuff.
Oh, quite the opposite. Some comic strips are designed from the ground up as advertising platforms. They don't have any humor value for anyone over 6, and yet still manage to at least make some money with the whorish accessorizing.
Maybe some people like it because it's cute (little old women with 12 cats?), but I can't see how anyone with a wit of humor would find it humorous.