Using/dev/random (or dev/urandom, for that matter) will give you an unknown and variable length: $ dd if=/dev/random bs=1024 count=5 | tr -dc [:alnum:] | wc -c 0+5 records in 0+5 records out 13 dd if=/dev/random bs=1024 count=5 | tr -dc [:alnum:] | wc -c 0+5 records in 0+5 records out 46 $
You'll want to loop until you have enough entropy./dev/urandom is likely going to give you a lot more chars per count parameter on most systems: $ dd if=/dev/urandom bs=1024 count=5 | tr -dc [:alnum:] | wc -c 5+0 records in 5+0 records out 1245 $ dd if=/dev/urandom bs=1024 count=5 | tr -dc [:alnum:] | wc -c 5+0 records in 5+0 records out 1278 $
Then just tail -c 63 for your 63 chars: $ dd if=/dev/urandom bs=1024 count=5 | tr -dc [:alnum:] | tail -c 63; echo 5+0 records in 5+0 records out avh2tglob7FQbgOZmMc8OdRjtw616R6CeOtrFro97pDXOa1W6S qTlsFAv8osERE $ dd if=/dev/urandom bs=1024 count=5 | tr -dc [:alnum:] | tail -c 63; echo 5+0 records in 5+0 records out COvjA1ubg11q0N5gPZVlO7VrQhLXg3ZXzKAnbuJ8Z5XzkHjEw6 i2Vv4OYGoGYe3 $
Cosmic rays can flip a bit, and have biological effects. Cosmic ray arrival rates are related to your altitude. Blacksburg VA may be only a thousand feet up, but Los Alamos is more like 8-9K--can't remember, exactly. But it's considerably higher than Santa Fe, which is about 7K.
Also, you may not be able to accept single-bit errors, even at the lower rate you'd experience at sea level. What's good enough for a Joe Gamer's PC in Denver (5K feet) may not be remotely good enough for mission-critical servers. Wall Street is at sea level, for the purposes of this discussion. But you can bet they don't always look at random bit-flips kindly.
As usual, how you spec a machine depends upon what you're planning to do with it. That even filters down into whatever spares you may stock for a server farm. Compare mfg. memory prices for large multi-CPU (think 8-64 CPUs, not some generic 4-way) servers running Solaris or HP-UX to memory available from the mass market vendors. Last time I did that, the mfg. memory was about 10X more expensive than something generic that would at least boot your system.
In that environment you need to be sure that the memory you're buying really is equivalent--not just that the machine will still boot. In that case, BTW, it turned out that there was equivalent memory available, without paying a huge vendor markup. It wasn't as cheap as the rock-bottom stuff (which would still have yielded a bootable machine) but it was cheap enough to justify buying a (much cheaper) machine just to stress test it before adding it to the ready spares bin for production systems. Sometimes a four hour support contract is still too slow, but you don't want to pay six figures for a hot backup system. Which also has to be under that expensive service contract.
If you have TB of memory to deal with, and mission-critical means minimize or eliminate flipped bits, the rules change considerably. It's a whole different world.
God only knows what gyrations the NSA must go through. Oh, wait, they just pay vendor rates, no matter the cost. My tax dollars at work...
"do you know how many ships and aircraft will divert off course"
No, and neither do you. Nor is anyone likely to have better than a rough estimate of two far more important numbers: passengers in the air, and the percentage whose lives would actually be threatened by a short duration (boost phase only) remapping.
"thats not a viable option"
That would depend upon the number of launches, their intended targets, the ability of traffic control systems to intervene (get on the radio to pilots, essentially), the ability of pilots to sense (visually, or prove via other nav aids) that a sudden skew in GPS nav data might be something to ignore, and probably lots of other variables.
Even one or two launches against large populations might easily make it worthwhile to remap, if the capability exists.
What review sites do you trust? I got disgusted with most of them a couple of years back--the last time I needed to spec PC hardware. I'm sure there must be some reputable ones out there, but I don't know who's currently the best.
The AC states that AMD has had an 80% market share for the last four years, immediately after asking someone if they were retarded. And the world ratchets up yet another notch on the weirdness scale.
This is indeed one of those cases where the entire chip industry is slowing. Not just CPUs, but RAM, the standard digital and analog 'building-block' parts used in all sorts of electronics, etc. In part, this is just a sign of the times--the global economy is slowing.
I'll call the current Intel/AMD price war a win for the consumer, as well. At least as long as it doesn't get serious enough to cripple AMD, which would damage competion between the two in the long run. Which I just don't think is in the cards. AMD recently inked a deal to build a new fab in New York, after all, at a cost of 3.2 billion dollars, and they say they're going to fund it out of operating cash flows.
It's supposed to be a 32nm fab, which is pretty interesting. Also, it's probably going to give them more capacity than they seem to need, based on current projects. Maybe they're planning some big wins, either with new CPUs, or lots more Dell, (or HP, or ?) business down the road?
I don't find it amazing, but part of a steady evolution.
Spacecraft autonomy software was a high-risk technology evaluated with Deep Space One, back in 1998-2001. http://nmp.nasa.gov/ds1/
The Autonomous Sciencecraft Experiment which flew aboard the EO-1 satellite mentioned in TFA http://ase.jpl.nasa.gov/ even used autonomy software related specifically to clouds. The AES delivered results in 2003-04, from looking at that link, though TFA would seem to imply that the effort is ongoing.
And of course autonomous operation software is a research focus in countless terrestrial projects.
I do see ongoing references to the seemingly never-ending Geek problem of bandwidth. I expected better from NASA, although I can't think why. They've had bandwidth problems for years.
- What will it take to create a scalable bandwidth solution? - Who's working on it, and what progress has been made? - Is it an international effort?
I'd like to see a lot more data coming down and being archived. There's a long history of discoveries have been made or confirmed through a review of old data. Another good use for recent data mining advances, as opposed to undermining our privacy, selling us more useless widgets, etc.
Autonomy at the tip of the spear must surely be regarded as a Good Thing (tm). I just hope it's not masking another, more fundamental, weakness in the system. It wouldn't surprise me to find that this is the case. Building infrastructure isn't sexy, and in these days of falling science budgets...
"It started to make me wonder if people outside the USA have a better picture of what's going on (even in our own country) since we are so "sheltered" from information."
I'm absolutely certain of it. The first three Web sites I hit for international news are:
Tremendously different slant on many issues than staying with US domestic news. Also some very clearcut cases of biased reporting in US media (not just the horrible Fox News). In fact, after a few years of the first two (Google News is comparatively new, of course) I can't *stand* domestic US 'news' services. But then I was never into pot-bellied pigs, either.
Except that on the version of BBC I get on Dish Network, it won't be available. There are 3-4 episodes of 'What Not to Wear' on tonight, though.
From the Linux Journal LJ Index, June 2006: Percentage of local authorities using Linux in the UK: 33 Percentage of local authorities using Linux in France: 71 Percentage of local authorities using Linux in Holland: 55 Percentage of local authorities using Linux in Germany: 68
Local authorities using Linux in the US is probably in single digits, which represents a horrible waste of tax dollars if I'm correct. I wonder if that's because mainstream press has still never mentioned it here. Or maybe BBC-USA just knows it's target audience only too well. Which is a scary thought, for all those that couldn't stand five minutes of 'What Not to Wear'.
Good on you! My experience (probably less extensive) has been the opposite, which I found profoundly depressing. Might I suggest that you write your experiences up for Linux Journal, or perhaps a medical or IT trade journal? I'd call this information that should have a wider audience than replying to a Slashdot post.
I'm not being facetious. If you've had good results, which would tend to make high quality rural medical care more available, I'd love to see the results get as wide an audience as possible.
If I ever had to pick a candidate for my being most hapilly dead wrong in a posting, this would pretty much be the one. It's an important issue.
United Sates. I try to remember to mention country of origin, which sounds like a lame excuse, but a search for previous posts should bear me out.
If a country isn't mentioned, it might be reasonable to assume a US origin.
This isn't http://slashdot.jp./ From the FAQ: Slashdot seems to be very U.S.-centric. Do you have any plans to be more international in your scope? Slashdot is U.S.-centric. We readily admit this, and really don't see it as a problem. Slashdot is run by Americans, after all, and the vast majority of our readership is in the U.S. We're certainly not opposed to doing more international stories, but we don't have any formal plans for making that happen. All we can really tell you is that if you're outside the U.S. and you have news, submit it, and if it looks interesting, we'll post it. It is worth noting that there is a Japanese Slashdot run by VA Japan. While we helped them a little in their early days, they essentially run their own content without any real involvement from us... none of us can read Kanji! There are currently no plans to do other language or nation specific Slashdot sites.
Answered by: CmdrTaco
Last Modified: 10/3/04
Yes. Major blunder on my part. I should probably avoid doing IM, mail, and Slashdot at the same time, on roughly the same topic, as I obviously cannot handle it.
Maybe because you're not that good at humor? Maybe because you fail at communication to the point that you think 'snarky' is a universally understood term, or adding a smiley conveys all?
I believe you're wrong. Check the ACM reference. I realize that really was a MS 'dirty little secret' for quite some time, but I believe they're completely MS-hosted now. If you have any evidence that they're still on a BSD, please post. A large number of Slashdotters would love to see it--myself very much included.
I'm a *ix guy, for reasons that would fill a book. But at the moment, *ix isn't a universal win. Take a small, rural (US), medical professional office as an example. Rural medicine is growing more problematic, every year. Half a dozen doctors, in a rural environment, are probably not going to be able to do Linux. They may not even know it exists, much less be familiar with better medical record security possibilities. Their office manager probably knows nothing but Windows, and the US, in it's infinite wisdom, is now busily stripping away recently passed protection of medical records requirements anyway.
Who gets that win, Microsoft, or an *ix?
Bear in mind that the local admin or office manager probably knows Microsoft because that's all the local educational system exposed him too--and that situation was funded with your tax dollars. Microsoft enjoys powerful advantages, not the least of which is the ability to fund development of operating systems, generic office apps, and vertical apps at a tremendous rate.
With their last quarterly report, they've finally been willing to take a hit on stock prices, and commit to serious R&D. This isn't all bad--maybe their security will finally become better than abysmal, for the average home user, who could use a break.
I think that *ix is still a better way forward. But it's not going to be a win in all cases. Nor should it be. In the case of that small rural medical center, I hope they do Win, as it currently gives them a better chance of remaining viable, and providing a local care option for people that need it badly. Religious wars on Slashdot are pretty pale in the face of whether there's care available when you've just lost lost a confrontation with farm, mining, etc., equipment.
*ix has to win on merits, in individual situations. Security, availability of admins, availability of vertical apps, etc. I think it will, in the case of Linux distros and perhaps the BSDs, (which are at the very least arguably better) but haven't had the press. But there's no place for religious wars, such as KDE v Gnome, or even Win v *ix.
Staying on top of things, doing your best to be brutally honest, and contributing, is the way forward. It's not about bashing something else but lifting what you believe in. I regret any MS bashing I've done. It took time away from recognizing that MS will not sleep much longer, and turning in patches, bug reports, and documentation.
Example. I totally bailed on Gnome when they trotted out the Spacial Abortion. But have you read KDE application help files? Do some brutal honesty: they suck! So what's the best use of my time? Should I mindlessly beat on Gnome, or get off my ass and make KDE better?
As a security guy, I was offended that lastb was broken in RH 7 through at least 9, and was not fixable by touching btmp. That makes *ix security reporting scripts less portable. It works in FC3 or better, but is still broken in SUSE 10. A year ago, in a client newsletter, I railed against lastb being broken in any Linux distro. That was beyond stupid. 2% of the effort should have been a newsletter mention, and 98% should have been submitting a fix.
If you love *ix, get up to date, find current problems, and get fixes in. Anything less is just wanking on Slashdot. You can argue that I'm currently doing exactly that, but I have 30 (mostly evening) hours in the next week blocked out for Linux fixes. A full day of that is writing docs, and I hate writing docs, BTW. So I'm putting my money where my mouth is.
"Until PCs do almost nothing and next to no data is locally owned, then McNealy's vision of the future still hasn't come true."
Good point. I think that's the trend, but let's hope we get a lot better at both network and host security, and understanding privacy concerns, before we get too much further down that road. Google wants to know far to much about me, companies delivering US IRS tax preparation apps want to legally sell your data, etc. Insert all the usual security/privacy (tin-foil hat?) arguments here.
I don't know that I'd say Sun created that wave, but neither were they a small part. Remember that for some time (at least a couple of years, IIRC) after the Mosaic browser, the killer app was still email. For all I know, there are still more mail than Web packets on the backbones. Anybody have any figures?
But it was definitely those relatively innexpensive Sun workstation class machines that powered much of DNS, mail, FTP, and gopher, in the days before the Web, and for at least a couple of years after the Web.
I have to call Sun a *major* contributor. To the extent that we're perhaps 3-5 years further along than we would have been without them, though there's absolutely no way to verify that SWAG.
Unfortunately, you're probably right. Wall Street penalized Sun stock for a long time because of R&D spending. After the just released MSFT quarter, the stock dropped 11%. One of the reasons the analysts gave was the increased R&D spend.
On one level, this sort of short-sighted thinking makes me want to throw things. It's not good for the industry, and it's not good for the country. But the market is what it is. Given the speed at which capital flows these days, I don't see it changing.
Which begs the question of where future R&D is going to come from. Universities increasingly want to lock up and license anything remotely marketable. Government funding is sliding.
Not a good situation, IMHO, and I'm fresh out of brilliant ideas. Support any state initiatives, and organizations such as ACM and USENIX, is about all I can suggest.
That turns out to not be the case. Any competent Unix admin can handle several systems. In large installations, it's common for a small admin group to handle hundreds of Unix and/or Linux systems. In the last survey I saw, the ratio of Windows admins to systems was 1:10. I'm not sure I completely buy into that, as I know of a couple of sites where it's more like 1:70 or 1:80, and I don't operate in the Windows world much. Maybe some surveys reflect the fact that there are more small installations running Win than *ix?
But *ix definitely lends itself to large ratios. If you think Google, Amazon, much of Wall Street, etc., would run it if this weren't the case, you are in error.
MS Hotmail would surely qualify as a large Win installation, wouldn't it? We're talking tens of thousands of machines.
Short answer is that less than 100 people admin it. A very high ratio.
The evidence shows either environment can run at high ratios, *if* you have the talent on hand, though *ix has a longer and better history in this area.
And finally, (opinion only) my take is that it's somewhat easier to find highly talented *ix admins. You may have to pay them a bit more, but the business case for that depends on, naturally, the business.
Or maybe you're referring to the single-machine case, though that isn't really that common? Sorry--you're still incorrect. Common or not, I know of several such, where one admin handles the system. Usually on top of Win duties as, completely contrary to your post, this isn't even remotely a full time job for a single person.
I've trained several such. As expected, some get it very quickly, some don't. The ones that didn't tended to be the ones that were hostile to the idea at a fundamental level. In some cases, that stance turned out to be 'career limiting', as they say. The ones that stepped up to the plate tended to be rewarded. They also added another valuable skill to their portfolio--surely this is no bad thing, in these uncertain times?
"Applications are all still hosted on the local machine with the exception of webmail clients."
What an amazing statement. I take it you don't do any remote banking, your workplace doesn't use one of the Web based CRM or system management apps, etc.
Slashdot Mirror
Using /dev/random (or dev/urandom, for that matter) will give you an unknown and variable length:
/dev/urandom is likely going to give you a lot more chars per count parameter on most systems:
S qTlsFAv8osERE6 i2Vv4OYGoGYe3
$ dd if=/dev/random bs=1024 count=5 | tr -dc [:alnum:] | wc -c
0+5 records in
0+5 records out
13
dd if=/dev/random bs=1024 count=5 | tr -dc [:alnum:] | wc -c
0+5 records in
0+5 records out
46
$
You'll want to loop until you have enough entropy.
$ dd if=/dev/urandom bs=1024 count=5 | tr -dc [:alnum:] | wc -c
5+0 records in
5+0 records out
1245
$ dd if=/dev/urandom bs=1024 count=5 | tr -dc [:alnum:] | wc -c
5+0 records in
5+0 records out
1278
$
Then just tail -c 63 for your 63 chars:
$ dd if=/dev/urandom bs=1024 count=5 | tr -dc [:alnum:] | tail -c 63; echo
5+0 records in
5+0 records out
avh2tglob7FQbgOZmMc8OdRjtw616R6CeOtrFro97pDXOa1W6
$ dd if=/dev/urandom bs=1024 count=5 | tr -dc [:alnum:] | tail -c 63; echo
5+0 records in
5+0 records out
COvjA1ubg11q0N5gPZVlO7VrQhLXg3ZXzKAnbuJ8Z5XzkHjEw
$
I'm betting on on them asking for the last four digits of my social security number. Wow, how did I get that cynical?
Cosmic rays can flip a bit, and have biological effects. Cosmic ray arrival rates are related to your altitude. Blacksburg VA may be only a thousand feet up, but Los Alamos is more like 8-9K--can't remember, exactly. But it's considerably higher than Santa Fe, which is about 7K.
Also, you may not be able to accept single-bit errors, even at the lower rate you'd experience at sea level. What's good enough for a Joe Gamer's PC in Denver (5K feet) may not be remotely good enough for mission-critical servers. Wall Street is at sea level, for the purposes of this discussion. But you can bet they don't always look at random bit-flips kindly.
As usual, how you spec a machine depends upon what you're planning to do with it. That even filters down into whatever spares you may stock for a server farm. Compare mfg. memory prices for large multi-CPU (think 8-64 CPUs, not some generic 4-way) servers running Solaris or HP-UX to memory available from the mass market vendors. Last time I did that, the mfg. memory was about 10X more expensive than something generic that would at least boot your system.
In that environment you need to be sure that the memory you're buying really is equivalent--not just that the machine will still boot. In that case, BTW, it turned out that there was equivalent memory available, without paying a huge vendor markup. It wasn't as cheap as the rock-bottom stuff (which would still have yielded a bootable machine) but it was cheap enough to justify buying a (much cheaper) machine just to stress test it before adding it to the ready spares bin for production systems. Sometimes a four hour support contract is still too slow, but you don't want to pay six figures for a hot backup system. Which also has to be under that expensive service contract.
If you have TB of memory to deal with, and mission-critical means minimize or eliminate flipped bits, the rules change considerably. It's a whole different world.
God only knows what gyrations the NSA must go through. Oh, wait, they just pay vendor rates, no matter the cost. My tax dollars at work...
"do you know how many ships and aircraft will divert off course"
No, and neither do you. Nor is anyone likely to have better than a rough estimate of two far more important numbers: passengers in the air, and the percentage whose lives would actually be threatened by a short duration (boost phase only) remapping.
"thats not a viable option"
That would depend upon the number of launches, their intended targets, the ability of traffic control systems to intervene (get on the radio to pilots, essentially), the ability of pilots to sense (visually, or prove via other nav aids) that a sudden skew in GPS nav data might be something to ignore, and probably lots of other variables.
Even one or two launches against large populations might easily make it worthwhile to remap, if the capability exists.
"Time for a proper democracy combined with literacy."
Hell, we could use that here in the US.
What does this have to do with any AMD parts? The reference you've provided compares single and dual core Intel Xeons!
What review sites do you trust? I got disgusted with most of them a couple of years back--the last time I needed to spec PC hardware. I'm sure there must be some reputable ones out there, but I don't know who's currently the best.
The AC states that AMD has had an 80% market share for the last four years, immediately after asking someone if they were retarded. And the world ratchets up yet another notch on the weirdness scale.
Intel has only recently seen it's stock price improve off of 3-year lows. Of course, AMD is also down quite a bit. 1 year graphs:
/ 27/39076/AMDtosetup32nmNewYorkfab.htm
http://finance.yahoo.com/q/bc?s=INTC&t=1y
http://finance.yahoo.com/q/bc?s=AMD&t=1y
This is indeed one of those cases where the entire chip industry is slowing. Not just CPUs, but RAM, the standard digital and analog 'building-block' parts used in all sorts of electronics, etc. In part, this is just a sign of the times--the global economy is slowing.
I'll call the current Intel/AMD price war a win for the consumer, as well. At least as long as it doesn't get serious enough to cripple AMD, which would damage competion between the two in the long run. Which I just don't think is in the cards. AMD recently inked a deal to build a new fab in New York, after all, at a cost of 3.2 billion dollars, and they say they're going to fund it out of operating cash flows.
http://www.electronicsweekly.com/Articles/2006/06
It's supposed to be a 32nm fab, which is pretty interesting. Also, it's probably going to give them more capacity than they seem to need, based on current projects. Maybe they're planning some big wins, either with new CPUs, or lots more Dell, (or HP, or ?) business down the road?
I have my private speculations...
Someone please mod parent up.
I don't find it amazing, but part of a steady evolution.
Spacecraft autonomy software was a high-risk technology evaluated with Deep Space One, back in 1998-2001.
http://nmp.nasa.gov/ds1/
The Autonomous Sciencecraft Experiment which flew aboard the EO-1 satellite mentioned in TFA
http://ase.jpl.nasa.gov/
even used autonomy software related specifically to clouds. The AES delivered results in 2003-04, from looking at that link, though TFA would seem to imply that the effort is ongoing.
And of course autonomous operation software is a research focus in countless terrestrial projects.
I do see ongoing references to the seemingly never-ending Geek problem of bandwidth. I expected better from NASA, although I can't think why. They've had bandwidth problems for years.
http://deepspace.jpl.nasa.gov/advmiss/#bandwidth
would seem to be a good information source. There's a lot to go through on that site, and none of it might answer my real questions:
- What will it take to create a scalable bandwidth solution?
- Who's working on it, and what progress has been made?
- Is it an international effort?
I'd like to see a lot more data coming down and being archived. There's a long history of discoveries have been made or confirmed through a review of old data. Another good use for recent data mining advances, as opposed to undermining our privacy, selling us more useless widgets, etc.
Autonomy at the tip of the spear must surely be regarded as a Good Thing (tm). I just hope it's not masking another, more fundamental, weakness in the system. It wouldn't surprise me to find that this is the case. Building infrastructure isn't sexy, and in these days of falling science budgets...
Amazing to see so many comments about hippies on Slashdot, as they're very nearly extinct. They haven't been *common* for twenty years.
"It started to make me wonder if people outside the USA have a better picture of what's going on (even in our own country) since we are so "sheltered" from information."
I'm absolutely certain of it. The first three Web sites I hit for international news are:
http://news.bbc.co.uk/
http://www.timesonline.co.uk/
http://news.google.com/
Tremendously different slant on many issues than staying with US domestic news. Also some very clearcut cases of biased reporting in US media (not just the horrible Fox News). In fact, after a few years of the first two (Google News is comparatively new, of course) I can't *stand* domestic US 'news' services. But then I was never into pot-bellied pigs, either.
Except that on the version of BBC I get on Dish Network, it won't be available. There are 3-4 episodes of 'What Not to Wear' on tonight, though.
From the Linux Journal LJ Index, June 2006:
Percentage of local authorities using Linux in the UK: 33
Percentage of local authorities using Linux in France: 71
Percentage of local authorities using Linux in Holland: 55
Percentage of local authorities using Linux in Germany: 68
Local authorities using Linux in the US is probably in single digits, which represents a horrible waste of tax dollars if I'm correct. I wonder if that's because mainstream press has still never mentioned it here. Or maybe BBC-USA just knows it's target audience only too well. Which is a scary thought, for all those that couldn't stand five minutes of 'What Not to Wear'.
Embedded systems in industrial controls, small network appliances, etc. There are millions of 486 & 586 class CPUs out there--and more every day.
Good on you! My experience (probably less extensive) has been the opposite, which I found profoundly depressing. Might I suggest that you write your experiences up for Linux Journal, or perhaps a medical or IT trade journal? I'd call this information that should have a wider audience than replying to a Slashdot post.
I'm not being facetious. If you've had good results, which would tend to make high quality rural medical care more available, I'd love to see the results get as wide an audience as possible.
If I ever had to pick a candidate for my being most hapilly dead wrong in a posting, this would pretty much be the one. It's an important issue.
United Sates. I try to remember to mention country of origin, which sounds like a lame excuse, but a search for previous posts should bear me out.
If a country isn't mentioned, it might be reasonable to assume a US origin.
This isn't http://slashdot.jp./ From the FAQ:
Slashdot seems to be very U.S.-centric. Do you have any plans to be more international in your scope?
Slashdot is U.S.-centric. We readily admit this, and really don't see it as a problem. Slashdot is run by Americans, after all, and the vast majority of our readership is in the U.S. We're certainly not opposed to doing more international stories, but we don't have any formal plans for making that happen. All we can really tell you is that if you're outside the U.S. and you have news, submit it, and if it looks interesting, we'll post it.
It is worth noting that there is a Japanese Slashdot run by VA Japan. While we helped them a little in their early days, they essentially run their own content without any real involvement from us... none of us can read Kanji! There are currently no plans to do other language or nation specific Slashdot sites.
Answered by: CmdrTaco
Last Modified: 10/3/04
Yes. Major blunder on my part. I should probably avoid doing IM, mail, and Slashdot at the same time, on roughly the same topic, as I obviously cannot handle it.
How embarrassing.
Maybe because you're not that good at humor? Maybe because you fail at communication to the point that you think 'snarky' is a universally understood term, or adding a smiley conveys all?
a rky
Have a look at http://www.urbandictionary.com/define.php?term=sn
There are several degrees of freedom in that term.
I believe you're wrong. Check the ACM reference. I realize that really was a MS 'dirty little secret' for quite some time, but I believe they're completely MS-hosted now. If you have any evidence that they're still on a BSD, please post. A large number of Slashdotters would love to see it--myself very much included.
I'm a *ix guy, for reasons that would fill a book. But at the moment, *ix isn't a universal win. Take a small, rural (US), medical professional office as an example. Rural medicine is growing more problematic, every year. Half a dozen doctors, in a rural environment, are probably not going to be able to do Linux. They may not even know it exists, much less be familiar with better medical record security possibilities. Their office manager probably knows nothing but Windows, and the US, in it's infinite wisdom, is now busily stripping away recently passed protection of medical records requirements anyway.
Who gets that win, Microsoft, or an *ix?
Bear in mind that the local admin or office manager probably knows Microsoft because that's all the local educational system exposed him too--and that situation was funded with your tax dollars. Microsoft enjoys powerful advantages, not the least of which is the ability to fund development of operating systems, generic office apps, and vertical apps at a tremendous rate.
With their last quarterly report, they've finally been willing to take a hit on stock prices, and commit to serious R&D. This isn't all bad--maybe their security will finally become better than abysmal, for the average home user, who could use a break.
I think that *ix is still a better way forward. But it's not going to be a win in all cases. Nor should it be. In the case of that small rural medical center, I hope they do Win, as it currently gives them a better chance of remaining viable, and providing a local care option for people that need it badly. Religious wars on Slashdot are pretty pale in the face of whether there's care available when you've just lost lost a confrontation with farm, mining, etc., equipment.
*ix has to win on merits, in individual situations. Security, availability of admins, availability of vertical apps, etc. I think it will, in the case of Linux distros and perhaps the BSDs, (which are at the very least arguably better) but haven't had the press. But there's no place for religious wars, such as KDE v Gnome, or even Win v *ix.
Staying on top of things, doing your best to be brutally honest, and contributing, is the way forward. It's not about bashing something else but lifting what you believe in. I regret any MS bashing I've done. It took time away from recognizing that MS will not sleep much longer, and turning in patches, bug reports, and documentation.
Example. I totally bailed on Gnome when they trotted out the Spacial Abortion. But have you read KDE application help files? Do some brutal honesty: they suck! So what's the best use of my time? Should I mindlessly beat on Gnome, or get off my ass and make KDE better?
As a security guy, I was offended that lastb was broken in RH 7 through at least 9, and was not fixable by touching btmp. That makes *ix security reporting scripts less portable. It works in FC3 or better, but is still broken in SUSE 10. A year ago, in a client newsletter, I railed against lastb being broken in any Linux distro. That was beyond stupid. 2% of the effort should have been a newsletter mention, and 98% should have been submitting a fix.
If you love *ix, get up to date, find current problems, and get fixes in. Anything less is just wanking on Slashdot. You can argue that I'm currently doing exactly that, but I have 30 (mostly evening) hours in the next week blocked out for Linux fixes. A full day of that is writing docs, and I hate writing docs, BTW. So I'm putting my money where my mouth is.
"Until PCs do almost nothing and next to no data is locally owned, then McNealy's vision of the future still hasn't come true."
Good point. I think that's the trend, but let's hope we get a lot better at both network and host security, and understanding privacy concerns, before we get too much further down that road. Google wants to know far to much about me, companies delivering US IRS tax preparation apps want to legally sell your data, etc. Insert all the usual security/privacy (tin-foil hat?) arguments here.
The way forward is both murky and interesting.
I don't know that I'd say Sun created that wave, but neither were they a small part. Remember that for some time (at least a couple of years, IIRC) after the Mosaic browser, the killer app was still email. For all I know, there are still more mail than Web packets on the backbones. Anybody have any figures?
But it was definitely those relatively innexpensive Sun workstation class machines that powered much of DNS, mail, FTP, and gopher, in the days before the Web, and for at least a couple of years after the Web.
I have to call Sun a *major* contributor. To the extent that we're perhaps 3-5 years further along than we would have been without them, though there's absolutely no way to verify that SWAG.
Unfortunately, you're probably right. Wall Street penalized Sun stock for a long time because of R&D spending. After the just released MSFT quarter, the stock dropped 11%. One of the reasons the analysts gave was the increased R&D spend.
On one level, this sort of short-sighted thinking makes me want to throw things. It's not good for the industry, and it's not good for the country. But the market is what it is. Given the speed at which capital flows these days, I don't see it changing.
Which begs the question of where future R&D is going to come from. Universities increasingly want to lock up and license anything remotely marketable. Government funding is sliding.
Not a good situation, IMHO, and I'm fresh out of brilliant ideas. Support any state initiatives, and organizations such as ACM and USENIX, is about all I can suggest.
That turns out to not be the case. Any competent Unix admin can handle several systems. In large installations, it's common for a small admin group to handle hundreds of Unix and/or Linux systems. In the last survey I saw, the ratio of Windows admins to systems was 1:10. I'm not sure I completely buy into that, as I know of a couple of sites where it's more like 1:70 or 1:80, and I don't operate in the Windows world much. Maybe some surveys reflect the fact that there are more small installations running Win than *ix?
t outs_unix/ for the situation there in 2002. Now it's all Win, and for a look at what it's currently like to admin, see http://www.acmqueue.com/modules.php?name=Content&p a=showpage&pid=353
But *ix definitely lends itself to large ratios. If you think Google, Amazon, much of Wall Street, etc., would run it if this weren't the case, you are in error.
MS Hotmail would surely qualify as a large Win installation, wouldn't it? We're talking tens of thousands of machines.
OK, now see http://www.theregister.co.uk/2002/11/21/ms_paper_
Short answer is that less than 100 people admin it. A very high ratio.
The evidence shows either environment can run at high ratios, *if* you have the talent on hand, though *ix has a longer and better history in this area.
And finally, (opinion only) my take is that it's somewhat easier to find highly talented *ix admins. You may have to pay them a bit more, but the business case for that depends on, naturally, the business.
Or maybe you're referring to the single-machine case, though that isn't really that common? Sorry--you're still incorrect. Common or not, I know of several such, where one admin handles the system. Usually on top of Win duties as, completely contrary to your post, this isn't even remotely a full time job for a single person.
I've trained several such. As expected, some get it very quickly, some don't. The ones that didn't tended to be the ones that were hostile to the idea at a fundamental level. In some cases, that stance turned out to be 'career limiting', as they say. The ones that stepped up to the plate tended to be rewarded. They also added another valuable skill to their portfolio--surely this is no bad thing, in these uncertain times?
"Applications are all still hosted on the local machine with the exception of webmail clients."
What an amazing statement. I take it you don't do any remote banking, your workplace doesn't use one of the Web based CRM or system management apps, etc.