Pirhana don't do the strip-the-flesh thing unless they're starving and provoked (which is how they got their reputation in the Western world: some Africans pulled a show to string Bwana Politician along, starving and confining pirhana for weeks, then driving a bleeding cow through them; nobody thought to question the show). There are also vegetarian pirhana which are very difficult to distinguish by eye from the omnivores.
The only real barrier to SAR pirhana is that they're too dumb. You'd basically need an aquatic mammal, and these (seals and dolphins) are being SAR-trained.
...try working with someone who describes the system on his machine as "Word" and complains about the (boilerplate) fax template from MS-Word not being present in OpenOffice as being one of the most important "failings" in the system. I kid you not.
From the sound of it, that's not far from the territory the GPP is in.
"Citizens are advised to buy and eat more beans to ensure continuity of supply in coming weeks. Authorities are considering airing an updated version of War of the Worlds during the first season of peak demand."
I know! (snaps fingers) It's because neither Cobb nor Badnarik have dodgy military service records to attack!
More seriously, I'm guessing it's because they don't rate either of the above as serious Presidential contenders, therefore there is no profit and great risk in trying to undermine their positions by personal attendance.
I personally find Badnarik a bit... I don't know... idealistic isn't quite the word. His ideas are mostly good, his background research seems to be good, his disclaimers are encouraging but I'm left with the nagging suspicion that he seriously underestimates the power of bureaucratic inertia and much other self-interested short-term thinking which keeps the USA in this current unhappy homeostasis. OTOH, possibly that's entirely appropriate for a candidate who doesn't genuinely expect to win this time around.
Linux CAN be made very secure against nasties on the net if the user is knowledgable.
[K] -> [System] -> [Configure your computer] -> root password -> [Security] -> [Firewall] -> tick what you want to expose (e.g. "Web Server") -> [OK] -> [X].
All easily enough discoverable by blundering around the menus. IPOF, it took me longer than it would a neophyte because my eye lit upon [Network & Internet] before the more obvious [Security] so I went bumbling around in there for a few seconds. Mandrake Linux, and the other major distros are all roughly as easy.
You can try the same thing in XP, and it's not many more clicks, not much more obscure, but the important thing about XP's equivalent tool is that it doesn't turn everything off.
Your point about hiding complexity but having it available immediately offstage is, however, a very good one.
No, they don't. I know a few small whiteboxers who bother to install all of the updates, and a few larger service companies, but your shiny new Dell/hp/AOpen box will be running the original version of Windows XP, unpatched.
You can often get all of the smaller patches seperately
No, you can't. You can get small patch collections separately, but you can't get down to the level of an individual patch, and nor can you (reasonably) alter those patches at all.
With Linux, you always have the choice of downloading the patch proper and rebuilding the thing yourself (typically rpm -bb package). Then if it causes you problems, you can delete or modify one element of a patch collection to make it happy again for you. You can also inspect the patch to see exactly what it does, rather than what the vendor says it does.
Of course you want to proactive before the damage
That's the GPP's point: Microsoft don't do the proactive thing nearly as well as the major FOSS packages. Part of this is that the MS code is not open to wide inspection, part of it is because Marketing rules the roost at Microsoft, and part of it is that William Henry "Trey" Gates III holds dear to his heart the principle of releasing new features in preference to fixing bugs - which attitude his company will eventually, inevitably propagate.
Have a look at the nature of the vulnerabilities, including those beyond IIS6 and Apache2. Typically, an Apache crack will get you limited access as user nobody or apache, but an IIS crack will get you carte blanche on the machine.
Would you rather break into a bank that had layer upon layer of security including internal hardwall partitioning, or one that only had thick external walls and a few alarms on those? Maybe ripping off the day's float would be not much different between banks, but what if getting into the vault is not significantly harder than getting the float at one bank?
It's not the admins (or at least, the difference in admin quality is not overwhelming enough to explain the differences in cracks). Roughly 80% of all email is transported by FOSS mailers, and yet where do practically all MTA attacks land? Shall we compare the difference in hardening between, say, PostFix or QMail and MS-Exchange?
Windows 2000 server runs IIS by default, and that's millions of target boxes that don't show in marketshare surveys.
Likewise, until recently, many Linux distros. Same story with PostgreSQL and MySQL.
This a testimony to the slackness of the people installing and running the boxes - they generally don't even know that they're running it.
Most Linux distros explicitly list services like that as they start up, but OTOH they very rarely have to start up. On the gripping hand ps uwax, service -s and the other what-is-my-machine-doing commands (and GUI tools) don't exactly hide the processes involved.
Ten-second job here, formatted for A4 and written in English.
Available in PS, PDF and SXW - so you can redo your own Letter or odd-sized verion. Uses font from LarabieFonts. Change "honour" to "honor" to get American instead of English.
Please don't let users mess with systems! Install drivers? Yeeargh! How many times d'you think they'll install the right driver? Nothing like have a user update a working ATI driver to a version which is broken on your hardware, is there?
WRT killing tasks, have you not seen stuff like GNOME System Monitor? All GUI and shiny, and the worst they can do on Linux is shoot one of their own tasks in the head. If your X is prone to locking up, either fix the #### thing or leave them Ctrl-Alt-BackSpace to play with.
Installing software is easy, you just open RPMdrake, type the root password (which I wouldn't give the user in a pink fit), select the packages you want and click "Install". If you do want the user installing packages (ie, you like extra work and don't mind frustration when they uninstall XFree86 and all dependencies because they can't see a use for it) then run RPMdrake SUID. SuSE, Debian, Red Hat all have similar pointy-clicky things.
...past three screens of a six-screen MS-Windows-XP setup, all shiny new boxes with (could be wrong, read the number at long range) 512MB of RAM and 19" LCD screens; things I noticed compared with a Linux-based Internet cafe I set up recently were:
of the three screens I saw, one had a frozen screen-saver (marquee), one was in use, one had a Norton anti-virus all-is-well report that the attendant couldn't make go away (clicking on yes, no or the little X blinked the dialog, and it was back pretty much instantly; she was trying the Task Manager as I walked out of sight); and
they had no CD drives, USB or firewire sockets exposed (main box was sealed away somewhere)
Despite the hardware costing (I guess) about double per station, the working machine still had odd hiccups and pauses in what it was doing (the guy using it was checking the RAM, presumably to see if it had enough for what he was doing).
I felt much better about my little Linux system after that.
20 x Mandrake 9.1 computers, no special effort, "ghosting" done once only with a NetCat-and-dd one-liner, zero maintenance. Really should spend an hour URPMI'ing them all before 10.1 gets released.
Ford Australia once had a bug in a run of their cars (Falcon EA, I think? Can never remember those model numbers) in which pressing hard on the brake pedal broke the mount loose form the firewall due to incorrect welding. Result: no brakes, just when you need them most.
I use OOo to rescue broken MSO documents with great success.
MSO saves documents out as a straight memory dump of its RAM OLE structure. If there is a mistake made in this process, when MSO loads the doc in it dumps the file straight into memory, does a few fixups as usual and flies straight into the deck.
Because OOo treats all documents as alien (ie it fully parses them on the way in), it will often recover all of a slightly damaged document which totals MSO.
As for MS-Access, it's a dead-end. Neat for little glue projects that aren't too complicated, but to turn a larger project into anything but a rolling disaster, you really have to know what you're doing.
Use a purpose-built DB admin tool to design your structures and queries, then plug OOo into that through ODBC and it'll work fine, but stay aware that an office suite is just one way of getting to a serious data system, it's a very frail thing to base a database system around. That's a lesson that many macro cowboys should have beaten into them at an early age.
Now this is something you're going to have to link to, or provide some other proof for. In any case, the article isn't technically censored; there's nothing preventing the people from publishing it elsewhere. They made an editorial decision not to publish it, it's up to them. If the paper really is good science, then they made an error in not publishing it, and in the long run it'll work to their detriment.
You didn't follow any of the links already supplied in the grandparent post, did you? Is your mind so made up that you don't want to be confused by any facts?
Follow the bouncing ball very slowly and carefully.
The article got peer reviewed by three highly competent and independent reviewers
The article has roughly a hundred references, the author's not pulling "facts" out of his butt
The article got published
NCSE and friends promptly browbeat the journal into promising never to publish another from an ID or Creationist author
The only reasons given were the author's profession, none of them had anything to do with the science behind the article
Future articles submitted to that journal will be censored for religious reasons
This is exactly the kind of self-censorship you say doesn't exist, and this is also exactly the kind of railroading of science that you report others complaining about.
The difference between ID and Creationism is that Creationism nominates a specific Designer, and ID does not. ID is precisely as religious as Atheism.
Atheism is a positive belief that there is no Creator, or to put it another way, the essence of Atheism is that we created ourselves. It could also be stated that Atheism means that we were created by accident.
All three of these statements are religious statements, even (or especially) the last one.
It is trivial to show that self-creation by accident is mathematically well beyond impossible. There's not nearly enough time (1E17 seconds) and materials (1E81 atoms) available under even the most stupidly optimistic of circumstances to achieve the required result. Bring on the self-structuring molecules, bring them all together and interact them at incredible rates in amazing quantities, do what you please, it still falls utterly flat. If you continue to believe that we are here by accident in the face of these observations, then it is clear that your Atheism is a religious belief, not a scientific one.
Note the absence of any appeal to philosophy, marginal dictionary definitions or any arguable point in the above line of reasoning. This is only one high-yield line of reasoning among many. Go roll the numbers out for yourself, from your own sources. Just be careful not to apply any un-proven theoretical magic: go and look at what Urey and Miller actually achieved, and how far backwards Miller went after that rather than assuming lightning plus goo plus time equals proteins. Pluck figures from the air for the smallest living creature. Even the most laughably simplistic designs are hundreds of orders of magnitude past the pale. Knock yourself out. The final answer is always zero (or within minus a hundred orders of magnitude, anyway).
Darwin was only able to fantasise about his little warm pond by thinking of cells as little structureless blobs of jelly instead of the fantastically intricate machines they actually are. Modern philosophers are similarly only able to hold their beliefs by carefully overlooking vast storehouses of observations in various scientific fields which speak loudly and clearly against evolution.
NCSE simply acted to retain their ignorance and thereby defend their faith, that's all.
Slashdot Mirror
Pirhana don't do the strip-the-flesh thing unless they're starving and provoked (which is how they got their reputation in the Western world: some Africans pulled a show to string Bwana Politician along, starving and confining pirhana for weeks, then driving a bleeding cow through them; nobody thought to question the show). There are also vegetarian pirhana which are very difficult to distinguish by eye from the omnivores.
The only real barrier to SAR pirhana is that they're too dumb. You'd basically need an aquatic mammal, and these (seals and dolphins) are being SAR-trained.
Perhaps you can pitch in and make it outstanding at doing Java?
...try working with someone who describes the system on his machine as "Word" and complains about the (boilerplate) fax template from MS-Word not being present in OpenOffice as being one of the most important "failings" in the system. I kid you not.
From the sound of it, that's not far from the territory the GPP is in.
IW4M.
...take on a whole new meaning.
"Citizens are advised to buy and eat more beans to ensure continuity of supply in coming weeks. Authorities are considering airing an updated version of War of the Worlds during the first season of peak demand."
RedHat as a distro doesn't exist, the closest you'll find is Red Hat.
Serious, there is a legal difference, but you just can't put a space into a hostname.
...and you insert .nyud.net:8090 at the end of the hostname.
...search engines and automated take-down notices.
...now why would that be? (-:
I know! (snaps fingers) It's because neither Cobb nor Badnarik have dodgy military service records to attack!
More seriously, I'm guessing it's because they don't rate either of the above as serious Presidential contenders, therefore there is no profit and great risk in trying to undermine their positions by personal attendance.
I personally find Badnarik a bit... I don't know... idealistic isn't quite the word. His ideas are mostly good, his background research seems to be good, his disclaimers are encouraging but I'm left with the nagging suspicion that he seriously underestimates the power of bureaucratic inertia and much other self-interested short-term thinking which keeps the USA in this current unhappy homeostasis. OTOH, possibly that's entirely appropriate for a candidate who doesn't genuinely expect to win this time around.
I now have a nice ironic screenshot of that headline with an MS "what will they call you?" ad above it.
All easily enough discoverable by blundering around the menus. IPOF, it took me longer than it would a neophyte because my eye lit upon [Network & Internet] before the more obvious [Security] so I went bumbling around in there for a few seconds. Mandrake Linux, and the other major distros are all roughly as easy.
You can try the same thing in XP, and it's not many more clicks, not much more obscure, but the important thing about XP's equivalent tool is that it doesn't turn everything off.
Your point about hiding complexity but having it available immediately offstage is, however, a very good one.
No, you can't. You can get small patch collections separately, but you can't get down to the level of an individual patch, and nor can you (reasonably) alter those patches at all.
With Linux, you always have the choice of downloading the patch proper and rebuilding the thing yourself (typically rpm -bb package). Then if it causes you problems, you can delete or modify one element of a patch collection to make it happy again for you. You can also inspect the patch to see exactly what it does, rather than what the vendor says it does.
That's the GPP's point: Microsoft don't do the proactive thing nearly as well as the major FOSS packages. Part of this is that the MS code is not open to wide inspection, part of it is because Marketing rules the roost at Microsoft, and part of it is that William Henry "Trey" Gates III holds dear to his heart the principle of releasing new features in preference to fixing bugs - which attitude his company will eventually, inevitably propagate.
Have a look at the nature of the vulnerabilities, including those beyond IIS6 and Apache2. Typically, an Apache crack will get you limited access as user nobody or apache, but an IIS crack will get you carte blanche on the machine.
Would you rather break into a bank that had layer upon layer of security including internal hardwall partitioning, or one that only had thick external walls and a few alarms on those? Maybe ripping off the day's float would be not much different between banks, but what if getting into the vault is not significantly harder than getting the float at one bank?
It's not the admins (or at least, the difference in admin quality is not overwhelming enough to explain the differences in cracks). Roughly 80% of all email is transported by FOSS mailers, and yet where do practically all MTA attacks land? Shall we compare the difference in hardening between, say, PostFix or QMail and MS-Exchange?
This a testimony to the slackness of the people installing and running the boxes - they generally don't even know that they're running it.
Most Linux distros explicitly list services like that as they start up, but OTOH they very rarely have to start up. On the gripping hand ps uwax, service -s and the other what-is-my-machine-doing commands (and GUI tools) don't exactly hide the processes involved.
Ten-second job here, formatted for A4 and written in English.
Available in PS, PDF and SXW - so you can redo your own Letter or odd-sized verion. Uses font from LarabieFonts. Change "honour" to "honor" to get American instead of English.
Please don't let users mess with systems! Install drivers? Yeeargh! How many times d'you think they'll install the right driver? Nothing like have a user update a working ATI driver to a version which is broken on your hardware, is there?
WRT killing tasks, have you not seen stuff like GNOME System Monitor? All GUI and shiny, and the worst they can do on Linux is shoot one of their own tasks in the head. If your X is prone to locking up, either fix the #### thing or leave them Ctrl-Alt-BackSpace to play with.
Installing software is easy, you just open RPMdrake, type the root password (which I wouldn't give the user in a pink fit), select the packages you want and click "Install". If you do want the user installing packages (ie, you like extra work and don't mind frustration when they uninstall XFree86 and all dependencies because they can't see a use for it) then run RPMdrake SUID. SuSE, Debian, Red Hat all have similar pointy-clicky things.
- of the three screens I saw, one had a frozen screen-saver (marquee), one was in use, one had a Norton anti-virus all-is-well report that the attendant couldn't make go away (clicking on yes, no or the little X blinked the dialog, and it was back pretty much instantly; she was trying the Task Manager as I walked out of sight); and
- they had no CD drives, USB or firewire sockets exposed (main box was sealed away somewhere)
- Despite the hardware costing (I guess) about double per station, the working machine still had odd hiccups and pauses in what it was doing (the guy using it was checking the RAM, presumably to see if it had enough for what he was doing).
I felt much better about my little Linux system after that.20 x Mandrake 9.1 computers, no special effort, "ghosting" done once only with a NetCat-and-dd one-liner, zero maintenance. Really should spend an hour URPMI'ing them all before 10.1 gets released.
Tron beat StarWars by a considerable margin, I'll have you young whipper-snappers know.
Dang, I used to have a DSL link.
Ford Australia once had a bug in a run of their cars (Falcon EA, I think? Can never remember those model numbers) in which pressing hard on the brake pedal broke the mount loose form the firewall due to incorrect welding. Result: no brakes, just when you need them most.
...for me, anyway.
I use OOo to rescue broken MSO documents with great success.
MSO saves documents out as a straight memory dump of its RAM OLE structure. If there is a mistake made in this process, when MSO loads the doc in it dumps the file straight into memory, does a few fixups as usual and flies straight into the deck.
Because OOo treats all documents as alien (ie it fully parses them on the way in), it will often recover all of a slightly damaged document which totals MSO.
As for MS-Access, it's a dead-end. Neat for little glue projects that aren't too complicated, but to turn a larger project into anything but a rolling disaster, you really have to know what you're doing.
Use a purpose-built DB admin tool to design your structures and queries, then plug OOo into that through ODBC and it'll work fine, but stay aware that an office suite is just one way of getting to a serious data system, it's a very frail thing to base a database system around. That's a lesson that many macro cowboys should have beaten into them at an early age.
The original IBM mainframes worked like this.
Also, earlier this year hp released a 4-way Linux machine which does the same trick with twice as many users - and considerably smaller licence fees.
Follow the bouncing ball very slowly and carefully.
- The article got peer reviewed by three highly competent and independent reviewers
- The article has roughly a hundred references, the author's not pulling "facts" out of his butt
- The article got published
- NCSE and friends promptly browbeat the journal into promising never to publish another from an ID or Creationist author
- The only reasons given were the author's profession, none of them had anything to do with the science behind the article
- Future articles submitted to that journal will be censored for religious reasons
This is exactly the kind of self-censorship you say doesn't exist, and this is also exactly the kind of railroading of science that you report others complaining about.The difference between ID and Creationism is that Creationism nominates a specific Designer, and ID does not. ID is precisely as religious as Atheism.
Atheism is a positive belief that there is no Creator, or to put it another way, the essence of Atheism is that we created ourselves. It could also be stated that Atheism means that we were created by accident.
All three of these statements are religious statements, even (or especially) the last one.
It is trivial to show that self-creation by accident is mathematically well beyond impossible. There's not nearly enough time (1E17 seconds) and materials (1E81 atoms) available under even the most stupidly optimistic of circumstances to achieve the required result. Bring on the self-structuring molecules, bring them all together and interact them at incredible rates in amazing quantities, do what you please, it still falls utterly flat. If you continue to believe that we are here by accident in the face of these observations, then it is clear that your Atheism is a religious belief, not a scientific one.
Note the absence of any appeal to philosophy, marginal dictionary definitions or any arguable point in the above line of reasoning. This is only one high-yield line of reasoning among many. Go roll the numbers out for yourself, from your own sources. Just be careful not to apply any un-proven theoretical magic: go and look at what Urey and Miller actually achieved, and how far backwards Miller went after that rather than assuming lightning plus goo plus time equals proteins. Pluck figures from the air for the smallest living creature. Even the most laughably simplistic designs are hundreds of orders of magnitude past the pale. Knock yourself out. The final answer is always zero (or within minus a hundred orders of magnitude, anyway).
Darwin was only able to fantasise about his little warm pond by thinking of cells as little structureless blobs of jelly instead of the fantastically intricate machines they actually are. Modern philosophers are similarly only able to hold their beliefs by carefully overlooking vast storehouses of observations in various scientific fields which speak loudly and clearly against evolution.
NCSE simply acted to retain their ignorance and thereby defend their faith, that's all.