JSF is wholly unsuited to our (Australias) strategic requirements.
If anyone in canberra would pull their heads out of eachothers arses long enough to listen to the experts they'd keep the F111's flying for a few more years and buy F22's instead for when the time comes to retire the F18's and F111's
Size difference is not surprising when comparing a long range interceptor (F14) with a fighter (F18).
The tomcats primary purpose was as a long range interceptor/air superiority fighter (similar role to the F15 and the soviets MiG 25). Its job was to protect the fleet by destroying incoming supersonic bombers before they reached their launch range. It had to have legs, be fast, be able to track and launch at multiple targets at extreme range. It is a big powerful brute, but not that nimble.
The F18 fell out of the design requirements for the F16 (indeed the USAF took the YF16 and the navy the YF17) as a nimble cheap fighter aircraft. Both performed within spec (light, manoeverable, nimble) with IIRC the navy choosing the F18 due to it having dual engines.
I think you'll find the F/A 18 when devoid of bombs would more than out dogfight the F14/F15 in a furball, that was what it was designed for.
Of course the F/A 18 would first have to close range to the interceptors, while doing so it is vulnerable, falling into the envelope of what the interceptors were designed for (destroying targets at maximum standoff distance).
So what was my point? the F/A designation doesn't mean it is not as capable a fighter as it should be, it is just as capable as the F16 in both roles. Comparing the F18/F16 to the F14/F15 is comparing apples and oranges, they have completely different roles.
Wandering around away from my display (armed with booty to trade, mugs for penguins etc) I came across 2 middle aged IT geeks checking out some glorious powdercoated, properly cooled, neatly wired and well laid out rack equipment on display.
As they were tinkering with the offerings one was heard to pronounce "what a great rack, wouldn't you love one in your home".
At this point the poor unsuspecting geek was set upon by one of the very well endowed skimpily clad models hired to parade around and lure in the punters, who promptly slapped him across the face and berated the poor confused fellow (who had that mix of deer in the headlights and WHA!! look on his face) for being a "misogynist pig" etc etc.
LFS is now just a bunch of cut-and-pasting from the site into a box. And anyone with a clue can make one in about 2hrs.
Yup, thats the general idea, though we'd hope people would at least read the book too and take a bit more effort to learn how the toolchain hangs together (that is the real lesson).
You'd be surprised how many folks cant even C'n'P properly though...
We're putting control and choice into users hands where there was none before. It's a fact of life that ISPs are doing this. They should be working with us, just like users are. We look at this as giving you a dashboard and all the knobs and buttons you need to manage DNS. DNS is the root (no pun intended) of a ton of applications and services so why wouldn't you want to manage it just like you would a firewall or anti-spam service?
<sarcasm> Of course DNS needs more user-friendly helpful value adding to it. I mean, I dont know how have we coped over the last 16 years without these wonderful value added services, I mean it is only DNS, the cornerstone of the internet, why shouldn't we mess with its default behaviour?
What could possibly go wrong? </sarcasm>
Obviously the whole concept of "dont fix what isn't broken" or "the law of unintended consequences" has never crossed the minds of the folks implementing these "value added" services.
If it isn't in the IETF draft IT DOES NOT BELONG!!!
The DNS system performs all the functions intended of it, once you start fucking with the default behaviour things break. DNS isnt the mechanism to use for providing this crap, it should be done CLIENT SIDE, and especially for the use case promoted here, CLIENT SIDE IN THE BROWSER ONLY. Of course that means folks would have to opt-in and install a browser plugin, not quite the money spinner as abusing the
DNS service to redirect to advertising, but hey...
The whole point of NetBSD is portability. If it weren't for portability, NetBSD might as well not exist. But the problem is Linux has taken over as the portability leader and has a huge margin.
Yes and no. Most people who make this comment should add the caveat "for 2.2.x/2.4.x series kernels with glibc 2.2.x/2.3.x w linuxthreads".
If your hardware is new and relatively common, linux kernel will most likely have a current working port.
Move off the path well travelled and you are in for a lot of strife.
My favorite 2 examples to bring up, mainly due to the massive number of units still in existence are
mac-m68k - last worked kernel 2.2 series, broken 2.4 and 2.6 (ADB broken, serial console broken)
sparc32 sun4m - currently SMP support broken in 2.6 kernels for supersparc, hypersparc just broken.
At least sun4m looks like it will get fixed, dont hold your breath for the m68k macs.
Dont get me wrong, I am a linux zealot, but linux' rate of change pretty much ensures that older architectures will drop out of the tree due to lack of maintenance, be it in kernel or glibc.
NetBSD still runs, and you can be pretty assured will continue to run, on most hardware you point it at (including your old vaxen;-) sure beats VMS ).
Voting is compulsory in Australia for all citizens over 18 years of age.
Failure to show up and get your name crossed off the list on election day results in a fine.
Note the only compulsory part is showing up, whether the voter actually filled out the ballot correctly or not (incorrect ballot forms, otherwise known as "donkey votes", do not count) is irrelevant.
SNMP == "Security is Not My Problem";-)
Seriously though, what you use is entirely dependant on what it is you exactly want to monitor.
It is trivial to write up simple net-snmp based pollers to push into RRDTool for graphing (my preferred method for
generating traffic stats, after polling for which interfaces are administratively and operationally up, saves on having
to configure what interfaces to monitor as you do with MRTG). Same data can also be pushed into whatever you use for historical
logs.
Looks like I struck a nerve with a dipshit AC out there:-)
FWIW, tard, I respect people who are themselves without the airs and pretentions of being better than everbody else, which is a rare commodity especially with those thrust into (or seeking) celebrity.
As for caricature, you obviously never met the man. He may not have been as overly hyperactive as he was on the box, but he was damned close.
Next time try pulling your head out of your arse, I am no ocker but was always amused by the disdain wankers like you always held the man in for doing no more than being himself and getting paid arseloads of cash for doing it (as opposed to mortgaging up to the hilt to buy a BMW to sit around Double Bay pretending they are someone they are not).
It always does feel better for the try-hard urban pretentious wannabe sophisticate to run down the genuine down-to-earth ocker out there, get out of Sydney (you are from Sydney right) and go and meet some of them.
Depends on the industry.
Mission critical 5 9's required? NOT going to go on tinker toy wintel boxes.
Needs to be robust? Uptime is important? Again, not going to go on wintel.
Needs to be secure.... this goes without saying.
Needs to run highly multithreaded apps... thats going to unix (most likely solaris), not wintel.
Need a big kick-arse database server? running SAP, Oracle Financials or the like?... you would be insane to put it on wintel.
Sure for web services (sitting behind a hardened reverse proxy) wintel can be found, and for SMB file servers or other desktop support services, sure, but "actively" moving towards wintel... more like only if they have no other choice.
See folks like their unix boxes, they install them, put their software on them, then shove them into the back of the
server room and ignore them for the next 3 years until either the lease runs out, the box needs critical security fixes (most often no reboot required unless it is kernel related) or some (fully redundant) piece of hardware needs replacement. They bring guaranteed peace of mind.
Note: I am somewhat biased, true. I provide consultancy to large corporates, sometimes on behalf of Tier 1 hardware vendors. I just calls it as I sees it (lets just say work is very busy at the moment)
/me raises a stubby of XXXX to Steve.
An untimely, but fitting, end for one of my fellow Queenslanders... going out doing what he loved best.
He may have irritated the hell out of the chardonnay/latte sipping set in this country (with their "cultural cringe" against anyone who is unashamedly ocker), but he was the genuine article.
As you saw him on TV he was in real life, self deprecating, unpretentious, genuine.
To those losing their sense of humour about this, I'd suspect he too would have appreciated the irony and would be laughing along with us in the best tradition of non-politically correct black australian humour.
Plus, Kerberos gets you SSO and the ability to secure NFS, which using LDAP doesn't.
You'll get SSO for kerborized services sure, but for those services that aren't...
A combination of both is best, LDAP as a naming service backend (authorisation), kerberos for authentication.
Even better if (using OpenLDAP) you backend the LDAP server for user passwords via saslauthd to kerberos so at least for non-kerborized services that understand ldap, or even for those that dont using pam w pam_ldap, you at least get "Single Password" and a centralised authentication store.
Performance tanked in a linux/solaris environment? You obviously did something wrong.
Here's a question, is your passwd backend in/etc/nsswitch.conf setup to run in nis compat mode? Using netgroups?
If so there is your problem. If you must use netgroups, use pam_access under linux and for solaris either port pam_access (I can put this up somewhere for interested folk) or use pam_netgroups (available from OpenSolaris, just fix the malloc() without free() in it).
No point loading up the naming service layer for getpw* calls (and the ldap backend due to the large increase in lookups) when you really only need to evaluate during account checks during login.
Looks like another weapon to use in the typical divorce procedings debacle
Stage 1: Accuse soon to be ex-hubby, along with insinuations of child beating.
Stage 2: ???
Stage 3: Profit
You would sure hope there is some recourse available to those unfairly smeared. But as is the case with a lot of things, after your good name is smeared by such an accusation, it wont matter if it was true or not, it isn't even a case of guilty until proven innocent (just guilty).
Cool, so you wont mind folks redirecting all the wonderful new mountains of spam to your server which now gets through because forged bogus sending domains now resolve. There is a reason you dont fuck with the naming service...
Best sales performance I ever saw was when a company I was working for were replacing the old fleet of laptops.
IBM sales guy comes in, states the specs then demonstrated the robustness by putting it on the floor and standing on it.
It impressed us enough to ask the other vendors to perform a similar demonstration, funnily enough the acer and compaq folks declined;-)
Yeah, just dont put Oracle on the box, their core licensing policy suggests Larry Ellison needs a new yacht (the $200M one he currently has just isn't good enough)
Re:Yes, Debian supports multiple gcc versions.
on
The Future of NetBSD
·
· Score: 1
For easy cross-compiling I'd want this in/usr/bin:
somearch-gcc, somearch-ld, somearch-gas, somearch-cpp, somearch-g++
Then in ~/somearch/ I put symlinks to those, using the plain (gcc, ld, etc.) names. This lets me select the crosscompilers with my PATH environment variable.
You dont need to do that for binutils, it effectively gets done already (ie under/usr/target-triple/bin except they are hardlinks to/usr/bin/target-triple-ld etc) and that is in the gcc drivers binary search path (gcc tends not to search PATH to find ld etc).
Cross-gcc will install to/usr/bin as target-triple-gcc.
As for invoking gcc, you set the driver specific options in CC not CFLAGS
(ie CC="gcc [-V XX] [-b target-triple] [-m32|m64]" or CC="[target-triple-]gcc[-version] [-m32|-m64]")
except for with some old packages using older braindead libtool which barfs on spaces.
Yeah, compiling it all down yourself can be a PITA (especially with multilib systems), the point I was trying to make though is that GNU binutils/gcc allows multiple installations to the same prefix out of the box if you use the correct configure options... I am surprised a lot of the distros dont do this though, more kudos to debian.
The thing is, even with rules in place, mailing-lists always end up being a place for megaphone diplomacy, he who shouts loudest and longest (even if they have no clue) wins.
The lists end up being political flamefests so anyone of actual consequence (ie: folks that do the work) will just depart the list to use IM/IRC/private email so as to avoid the bullshit and get on with work.
Maybe to avoid this projects should use Slash instead of mailing lists, at least the smack-tards could be moderated out of existence;-)
I thought the origins of the comment was from Tom Wolfes "the electric kool-aid acid test", concerning Ken Kesey and his crew in the 60's setting up parties with LSD laced Kool-Aid...
Slashdot Mirror
JSF is wholly unsuited to our (Australias) strategic requirements.
If anyone in canberra would pull their heads out of eachothers arses long enough to listen to the experts they'd keep the F111's flying for a few more years and buy F22's instead for when the time comes to retire the F18's and F111's
Some interesting articles and papers with comparisons are available at http://www.ausairpower.net/jsf.html
Size difference is not surprising when comparing a long range interceptor (F14) with a fighter (F18).
The tomcats primary purpose was as a long range interceptor/air superiority fighter (similar role to the F15 and the soviets MiG 25). Its job was to protect the fleet by destroying incoming supersonic bombers before they reached their launch range. It had to have legs, be fast, be able to track and launch at multiple targets at extreme range. It is a big powerful brute, but not that nimble.
The F18 fell out of the design requirements for the F16 (indeed the USAF took the YF16 and the navy the YF17) as a nimble cheap fighter aircraft. Both performed within spec (light, manoeverable, nimble) with IIRC the navy choosing the F18 due to it having dual engines.
I think you'll find the F/A 18 when devoid of bombs would more than out dogfight the F14/F15 in a furball, that was what it was designed for.
Of course the F/A 18 would first have to close range to the interceptors, while doing so it is vulnerable, falling into the envelope of what the interceptors were designed for (destroying targets at maximum standoff distance).
So what was my point? the F/A designation doesn't mean it is not as capable a fighter as it should be, it is just as capable as the F16 in both roles. Comparing the F18/F16 to the F14/F15 is comparing apples and oranges, they have completely different roles.
Heh, best story I have comes from Comdex/Interop.
Wandering around away from my display (armed with booty to trade, mugs for penguins etc) I came across 2 middle aged IT geeks checking out some glorious powdercoated, properly cooled, neatly wired and well laid out rack equipment on display.
As they were tinkering with the offerings one was heard to pronounce "what a great rack, wouldn't you love one in your home".
At this point the poor unsuspecting geek was set upon by one of the very well endowed skimpily clad models hired to parade around and lure in the punters, who promptly slapped him across the face and berated the poor confused fellow (who had that mix of deer in the headlights and WHA!! look on his face) for being a "misogynist pig" etc etc.
Took 2 hours for my sides to stop hurting...
You'd be surprised how many folks cant even C'n'P properly though...
Of course DNS needs more user-friendly helpful value adding to it. I mean, I dont know how have we coped over the last 16 years without these wonderful value added services, I mean it is only DNS, the cornerstone of the internet, why shouldn't we mess with its default behaviour?
What could possibly go wrong?
</sarcasm>
Obviously the whole concept of "dont fix what isn't broken" or "the law of unintended consequences" has never crossed the minds of the folks implementing these "value added" services.
If it isn't in the IETF draft IT DOES NOT BELONG!!!
The DNS system performs all the functions intended of it, once you start fucking with the default behaviour things break. DNS isnt the mechanism to use for providing this crap, it should be done CLIENT SIDE, and especially for the use case promoted here, CLIENT SIDE IN THE BROWSER ONLY. Of course that means folks would have to opt-in and install a browser plugin, not quite the money spinner as abusing the DNS service to redirect to advertising, but hey...
If your hardware is new and relatively common, linux kernel will most likely have a current working port.
Move off the path well travelled and you are in for a lot of strife.
My favorite 2 examples to bring up, mainly due to the massive number of units still in existence are
- mac-m68k - last worked kernel 2.2 series, broken 2.4 and 2.6 (ADB broken, serial console broken)
- sparc32 sun4m - currently SMP support broken in 2.6 kernels for supersparc, hypersparc just broken.
At least sun4m looks like it will get fixed, dont hold your breath for the m68k macs.Dont get me wrong, I am a linux zealot, but linux' rate of change pretty much ensures that older architectures will drop out of the tree due to lack of maintenance, be it in kernel or glibc.
NetBSD still runs, and you can be pretty assured will continue to run, on most hardware you point it at (including your old vaxen
Failure to show up and get your name crossed off the list on election day results in a fine.
Note the only compulsory part is showing up, whether the voter actually filled out the ballot correctly or not (incorrect ballot forms, otherwise known as "donkey votes", do not count) is irrelevant.
SNMP == "Security is Not My Problem" ;-)
Seriously though, what you use is entirely dependant on what it is you exactly want to monitor.
It is trivial to write up simple net-snmp based pollers to push into RRDTool for graphing (my preferred method for generating traffic stats, after polling for which interfaces are administratively and operationally up, saves on having to configure what interfaces to monitor as you do with MRTG). Same data can also be pushed into whatever you use for historical logs.
If you dont want to roll your own just use Nagios
if they pointed it at slashdot...
"ass-hat" and "tard" could take on a whole new meaning
Looks like I struck a nerve with a dipshit AC out there :-)
FWIW, tard, I respect people who are themselves without the airs and pretentions of being better than everbody else, which is a rare commodity especially with those thrust into (or seeking) celebrity.
As for caricature, you obviously never met the man. He may not have been as overly hyperactive as he was on the box, but he was damned close.
Next time try pulling your head out of your arse, I am no ocker but was always amused by the disdain wankers like you always held the man in for doing no more than being himself and getting paid arseloads of cash for doing it (as opposed to mortgaging up to the hilt to buy a BMW to sit around Double Bay pretending they are someone they are not).
It always does feel better for the try-hard urban pretentious wannabe sophisticate to run down the genuine down-to-earth ocker out there, get out of Sydney (you are from Sydney right) and go and meet some of them.
Ass-hat.
Depends on the industry.
Mission critical 5 9's required? NOT going to go on tinker toy wintel boxes.
Needs to be robust? Uptime is important? Again, not going to go on wintel.
Needs to be secure.... this goes without saying.
Needs to run highly multithreaded apps... thats going to unix (most likely solaris), not wintel.
Need a big kick-arse database server? running SAP, Oracle Financials or the like?... you would be insane to put it on wintel.
Sure for web services (sitting behind a hardened reverse proxy) wintel can be found, and for SMB file servers or other desktop support services, sure, but "actively" moving towards wintel... more like only if they have no other choice.
See folks like their unix boxes, they install them, put their software on them, then shove them into the back of the server room and ignore them for the next 3 years until either the lease runs out, the box needs critical security fixes (most often no reboot required unless it is kernel related) or some (fully redundant) piece of hardware needs replacement. They bring guaranteed peace of mind.
Note: I am somewhat biased, true. I provide consultancy to large corporates, sometimes on behalf of Tier 1 hardware vendors. I just calls it as I sees it (lets just say work is very busy at the moment)
/me raises a stubby of XXXX to Steve.
An untimely, but fitting, end for one of my fellow Queenslanders... going out doing what he loved best.
He may have irritated the hell out of the chardonnay/latte sipping set in this country (with their "cultural cringe" against anyone who is unashamedly ocker), but he was the genuine article.
As you saw him on TV he was in real life, self deprecating, unpretentious, genuine.
To those losing their sense of humour about this, I'd suspect he too would have appreciated the irony and would be laughing along with us in the best tradition of non-politically correct black australian humour.
See ya mate, you will be missed.
the internet connection applies for YOU
Indeed, and with JES 6 there will be no limit on the amount of master replicas allowed to replicate.
Should be out end of this year.
You'll get SSO for kerborized services sure, but for those services that aren't...
A combination of both is best, LDAP as a naming service backend (authorisation), kerberos for authentication.
Even better if (using OpenLDAP) you backend the LDAP server for user passwords via saslauthd to kerberos so at least for non-kerborized services that understand ldap, or even for those that dont using pam w pam_ldap, you at least get "Single Password" and a centralised authentication store.
Performance tanked in a linux/solaris environment? You obviously did something wrong.
/etc/nsswitch.conf setup to run in nis compat mode? Using netgroups?
Here's a question, is your passwd backend in
If so there is your problem. If you must use netgroups, use pam_access under linux and for solaris either port pam_access (I can put this up somewhere for interested folk) or use pam_netgroups (available from OpenSolaris, just fix the malloc() without free() in it).
No point loading up the naming service layer for getpw* calls (and the ldap backend due to the large increase in lookups) when you really only need to evaluate during account checks during login.
Looks like another weapon to use in the typical divorce procedings debacle
Stage 1: Accuse soon to be ex-hubby, along with insinuations of child beating.
Stage 2: ???
Stage 3: Profit
You would sure hope there is some recourse available to those unfairly smeared. But as is the case with a lot of things, after your good name is smeared by such an accusation, it wont matter if it was true or not, it isn't even a case of guilty until proven innocent (just guilty).
Cool, so you wont mind folks redirecting all the wonderful new mountains of spam to your server which now gets through because forged bogus sending domains now resolve. There is a reason you dont fuck with the naming service...
For sure.
;-)
Best sales performance I ever saw was when a company I was working for were replacing the old fleet of laptops.
IBM sales guy comes in, states the specs then demonstrated the robustness by putting it on the floor and standing on it.
It impressed us enough to ask the other vendors to perform a similar demonstration, funnily enough the acer and compaq folks declined
gah s/elevator/escalator/ :-/
Never fear, if it doesn't work they can take the space elevator or the space stairs...
Yeah, just dont put Oracle on the box, their core licensing policy suggests Larry Ellison needs a new yacht (the $200M one he currently has just isn't good enough)
You dont need to do that for binutils, it effectively gets done already (ie under
Cross-gcc will install to
As for invoking gcc, you set the driver specific options in CC not CFLAGS
(ie CC="gcc [-V XX] [-b target-triple] [-m32|m64]" or CC="[target-triple-]gcc[-version] [-m32|-m64]")
except for with some old packages using older braindead libtool which barfs on spaces.
Yeah, compiling it all down yourself can be a PITA (especially with multilib systems), the point I was trying to make though is that GNU binutils/gcc allows multiple installations to the same prefix out of the box if you use the correct configure options... I am surprised a lot of the distros dont do this though, more kudos to debian.
The thing is, even with rules in place, mailing-lists always end up being a place for megaphone diplomacy, he who shouts loudest and longest (even if they have no clue) wins.
;-)
The lists end up being political flamefests so anyone of actual consequence (ie: folks that do the work) will just depart the list to use IM/IRC/private email so as to avoid the bullshit and get on with work.
Maybe to avoid this projects should use Slash instead of mailing lists, at least the smack-tards could be moderated out of existence
I thought the origins of the comment was from Tom Wolfes "the electric kool-aid acid test", concerning Ken Kesey and his crew in the 60's setting up parties with LSD laced Kool-Aid...