Today's programmers are not trained to write efficient code (i.e. massively parallel) using good tools, or even in making good technical decisions. The goals of "cheapest available coders" won, so now they will need to develop AI programs to generate this kind of code becuase today's group of lowest-cost "programmers" certainly cannot do it.
Or you could do what I did, give them back the sheet of paper, blank, and invite them to escort me immediately from their "secure" facility. And then lecture them on the religious concept of "being forgiven and fully cleansed of past sins, which are now forgotten".
Obviously, if your technical skills are not that good, this might be a bad idea.
As a condition for "forgiving" this debt, this school district should have to implement the Microsoft-approved IBM/Suse/Novell version of Linux (this is a great punishment, it is very ancient compared to modern Linux!).
It shouldn't be that hard to setup something simpler than the current mess. For example, mycompany.com might point to DNS squatters in the ICANN domain, but to an actual "mycompany" in the "new" domain. The new domain would feature low registration fees to those willing to put up a real website. Perhaps something like torrents could be used to spread the DNS updates to eliminate control by evil entities.
Since normal DNS servers would get ICANN entries, non-evil people can point to port 53 to use non-evil DNS servers (primary/secondary, perhaps using an evil ICANN-based DNS server as a tertiary fallback).
Search engines could choose to return an IP-based page hit for a non-ICANN "mycompany.com" search.
We could eliminate the ability of evil entities being able to "take down" a site using DNS measures (i.e Thailand, GWB, China, etc.).
Credit Union members are required to have a "common bond". I'm not sure if being a Linux/OSS user is sufficient. Also, the family of CU members are also usually eligible for membership.
This would be much better than the current program, since all of the profits of the "OSS Users Credit Union" could be designated for OSS projects, and a variety of credit cards could be issued with Tux Penguins, FreeBSD devils, etc.
Plus, this credit union would not be evil.
I volunteer to be the (paid) president of this credit union!
I want to get rid of all of SeLinux on Debian
on
SELinux by Example
·
· Score: 1
It seems to be installed as a "required" package in a minimal install (debootstrap).
When I do 'apt-get remove libselinux1' it wants to remove most of the packages on my system (xorg, etc.). Other posts seem to indicate they are not running some of it, but I want to run NONE of it.
Your Wikipedia link does not refute my assertion, it only says direct record company payments to radio stations are illegal, not the payments by third party intermediaries. Your link also shows three record companies settling with Elliot Spitzer in 2005 over this practice.
Meanwhile, the corporate Clear Channels pay just $550 Million for broadcasting the same songs
Ordinary radio stations are expected to play [only] the songs they've received incentive pay or broadcasting discounts to promote. For example, if a performer is giving a concert soon nearby, airplay will be purchased of that performer's songs to drive ticket sales.
There is software that "listens" to the radio station and verifies that the songs and commercials they've been paid to play a certain number of times are actually being played that many times. Usually it is fully automated, but occasionally a human with headphones is needed when the software isn't sure (kind of like voice recognition software).
I interviewed at a company that provided this software in 1995 (it ran under DOS).
starts polling finance.yahoo.com You can't trade profitably using delayed quotes (15-20 minutes delay), like at finance.yahoo.com. In some markets, you cannot even afford a two second delay. That's why there are expensive real time quote services.
The OEM version of Windows Dell includes may not be the version of Windows the end user wants -- this affects Windows users as well as Linux users.
If you want to run a dual boot Linux/Windows machine mostly in Linux, the Dell XP OEM will not run in a virtual machine like
VMWare. So you have to buy another generic XP Pro OEM copy, for perhaps USD 139.
In a business, re-imaging a mixture a Dell/Non-Dell machines requires a non-Dell OEM version
of XP (generic XP OEM works on Dells, but Dell XP only works on Dells). Re-imaging is the only efficient way to deploy/maintain lots of machines, since the estimate for "installation from scratch" is 10+ hours to install XP, Office, Updates, applications, leading to white papers advising how it is cheaper to "refresh hardware" than use valuable technician time to reinstall XP from scratch on each workstation.
The non-transferable OEM license associated with a particular Dell machine/COA implies a licensing model for Windows similar to Redhat, i.e. the OEM license is tied to support, which is not truthful at all since the license is for the purpose of operating the computer.
Looking through many competitive software products was too much effort for those in all companies needing to make an IT decision. So.. just choose whatever is from Microsoft! If your company is smaller than [threshold], you will not be bothered by BSA,..., -- our software is [effectively] free!
So now we have a huge number of small businesses running "Small Business Server" [pathetic] with pirated XP Pro installs. All of them are barely working, with server/workstations "virused up".
But try to talk to any decision makers at such companies "we like it the way it is -- we are in control!". [Even though we know everything about IT] we can't fix it -- you fix it [for free]. We will get our revenge against Bill G. for his crappy product (against you, the IT person).
The only answer is to let the businesses who have made the "Microsoft decisions" crash and burn, as they deserve.
Now with the WGA, "our recommended Linux solution [that is even worse than MSFT products] is from our great partners IBM, Novell, and Suse". It makes our [garbage] stuff look good!
There is the assumption that the PIN is transmitted "in the clear" across the internal ATM/Pinpad network. I only know about pinpads, but in pinpads the PIN is encrypted in hardware using a key that changes each time. I would assume ATMs are equally secure.
In some countries, I understand the entire transaction (PIN, amount, account number) are encrypted in hardware, so trying to replay the encrypted packet is a wasted effort.
Any lack of security is caused by the same forces at play everywhere, the desire to use the cheapest "almost competent" programmers.
In all fields "Point of Sale" related, I'm guessing 99.9% is now good enough rather than 99.9999%. I assume ATM's will follow.
1) Find large customer considering Linux/having Linux problem
2) Demo easy fix
3) Find IBM/Novell/Suse got there first
4) Easy fix is not welcome, they want a broken system (need high headcount of idiots, still looking for fix done the "IBM way" (lots of Java written in India).
5) Get angry at IBM & friends, recommend M$ as something that sucks less/less evil
6) M$ solution implemented, difficult, expensive, unreliable, but still better than IBM/Suse/Novell crap.
7) ?? Profit (for M$ only). No wonder M$ & Suse work so well together.
It's happened to me twice this year, on two completely different hardware platforms. IBM had a broken "customized/ancient" version of the kernel attached to drivers for their "secret hardware interface". It is much easier to download M$ stuff (passport, etc.) than the secret stuff through Novell for IBM hardware/joint ventures.
Customers don't care, they either want IBM or M$. At a status meeting: "IBM/M$ is looking into the issue why (whatever) is totally slow/broken.
And we wonder why Linux is not getting anywhere with such great "friends". Be sure to recommend Suse/OpenSuse to newbies, we must honor our commercial hosts who contribute rather dubious value to Linux. OS/2 looks good relative to the IBM incarnations of Linux -- I laugh at the stupid customers who bought this IBM offering. It was funny to hear IBM remind customers "we are no longer in the PC business" just after "15 years from now, which vendor will be here to support your Linux/Java/..."
M$ is quite willing to relegate Linux to those "customers" (government, etc.) that require a broken, slow system to justify "how can we fix this mess we selected [because we wanted a broken system]. So Suse/M$ work together well as a team.
Sunbird/Lightning for Linux x86 have had a critical bug -- cannot click on 'recipients' field. Retested with last nights builds, bug still there. The product is basically useless with this bug.
Similarly, current Xulrunner 1.9 nightly builds run very slow -- need to revert to the version from Sept. 2005. Xulrunner is relevant because to help with Sunbird you need to know XUL. Learning XUL, you cannot even find a "Hello World" program to get started, instead the tutorials at xulplanet.com try to wow you with all the features in XUL (if you could _just_ get your first "Hello World" program working!). I had to work backwards from a "pacman" game (the only program I could find that actually had the three components JavaScript, CSS, and XUL working), to produce my own "Hello World" program. You can do some amazing things with XUL at that point. Mozdev actually has a program "Exch" in its "XUL hall of fame" that the author couldn't get the CSS working on due to their convoluted, poorly documented, and constantly changing "Chrome://", overlays, skins, etc.
The various Mozilla components share lots of code, and I see things like "Firefox and Thunderbird (and presumably Sunbird/Lightning) will run as Xulrunner applications once Xulrunner reaches 1.0" (1.9 ?> 1.0). Seems like that plan was abandoned, and the current xulrunner nightly build is barely running.
To summarize, it seems the entire Mozilla development process needs review to see why major components apparently have no simple test cases/documentation and the community is being asked to do a bug hunt on a non-working product.
With the SCO lawsuit dying out, it looks like MSFT needs new ways to entangle FOSS. Hence, firing execs using pure FUD against Linux, and putting on a friendly face on "open interfaces" and placing more source code we're not allowed to use out there. There is a certain class of cretin that will incorporate "shared source" code into an OSS project.
The legal departments of large corporations are already terrified of FOSS (from the SCO lawsuit) and will require indemnification and eventually all OSS projects will have to go through rigorous audits to show they contain no code from tainted MSFT source releases. Thanks MSFT, for adding massive code auditing overhead to OSS development!
In the good old days, a new company (such as Vonage) would go public long before it was "discovered", allowing early investors to get rich (like Microsoft, for example). In today's world, major banks provide working capital and the objective is to delay the IPO as long as possible, so that only the banks and the founders make any real money. I'm predicting the disappointment seen today with Vonage is going to become the norm for technology IPO's.
If you want to know what the temperature was 1-3 hours ago, go to NOAA. Sometimes the weather stations stop responding for hours, even days. These same non-current temperatures are then picked up and reported by "Accuweather". I found this out when my AC died and I needed to know what the CURRENT temperature was so I could cool my house with outside air.
When the system is working "correctly", you can go to the NOAA website at 8:05am and get the temperature, barometric pressure, etc. at 6:53am.
Kweather is even worse. The 'update now' button almost never works, so even if a 1.5 hour ago temperature is available it will try to cling onto it for an extra hour or more "to reduce network abuse" I guess; when I looked at the marvelous C++ source code I was amazed that someone could code a "fetch FTP page, parse, display in a KDE button" in 500000+ lines of incomprehensible C++.
If you need the current weather conditions, buy an expensive weather station; I think most of the junk in the stores has a precision of plus/minus 5 degrees F.
Slashdot Mirror
Today's programmers are not trained to write efficient code (i.e. massively parallel) using good tools, or even in making good technical decisions. The goals of "cheapest available coders" won, so now they will need to develop AI programs to generate this kind of code becuase today's group of lowest-cost "programmers" certainly cannot do it.
Or you could do what I did, give them back the sheet of paper, blank, and invite them to escort me immediately from their "secure" facility. And then lecture them on the religious concept of "being forgiven and fully cleansed of past sins, which are now forgotten". Obviously, if your technical skills are not that good, this might be a bad idea.
As a condition for "forgiving" this debt, this school district should have to implement the Microsoft-approved IBM/Suse/Novell version of Linux (this is a great punishment, it is very ancient compared to modern Linux!).
It shouldn't be that hard to setup something simpler than the current mess. For example, mycompany.com might point to DNS squatters in the ICANN domain, but to an actual "mycompany" in the "new" domain. The new domain would feature low registration fees to those willing to put up a real website. Perhaps something like torrents could be used to spread the DNS updates to eliminate control by evil entities.
Since normal DNS servers would get ICANN entries, non-evil people can point to port 53 to use non-evil DNS servers (primary/secondary, perhaps using an evil ICANN-based DNS server as a tertiary fallback).
Search engines could choose to return an IP-based page hit for a non-ICANN "mycompany.com" search.
We could eliminate the ability of evil entities being able to "take down" a site using DNS measures (i.e Thailand, GWB, China, etc.).
Credit Union members are required to have a "common bond". I'm not sure if being a Linux/OSS user is sufficient. Also, the family of CU members are also usually eligible for membership.
This would be much better than the current program, since all of the profits of the "OSS Users Credit Union" could be designated for OSS projects, and a variety of credit cards could be issued with Tux Penguins, FreeBSD devils, etc.
Plus, this credit union would not be evil.
I volunteer to be the (paid) president of this credit union!
It seems to be installed as a "required" package in a minimal install (debootstrap). When I do 'apt-get remove libselinux1' it wants to remove most of the packages on my system (xorg, etc.). Other posts seem to indicate they are not running some of it, but I want to run NONE of it.
Your Wikipedia link does not refute my assertion, it only says direct record company payments to radio stations are illegal, not the payments by third party intermediaries. Your link also shows three record companies settling with Elliot Spitzer in 2005 over this practice.
Ordinary radio stations are expected to play [only] the songs they've received incentive pay or broadcasting discounts to promote. For example, if a performer is giving a concert soon nearby, airplay will be purchased of that performer's songs to drive ticket sales.
There is software that "listens" to the radio station and verifies that the songs and commercials they've been paid to play a certain number of times are actually being played that many times. Usually it is fully automated, but occasionally a human with headphones is needed when the software isn't sure (kind of like voice recognition software).
I interviewed at a company that provided this software in 1995 (it ran under DOS).
VMWare. So you have to buy another generic XP Pro OEM copy, for perhaps USD 139.
In a business, re-imaging a mixture a Dell/Non-Dell machines requires a non-Dell OEM version of XP (generic XP OEM works on Dells, but Dell XP only works on Dells). Re-imaging is the only efficient way to deploy/maintain lots of machines, since the estimate for "installation from scratch" is 10+ hours to install XP, Office, Updates, applications, leading to white papers advising how it is cheaper to "refresh hardware" than use valuable technician time to reinstall XP from scratch on each workstation.
The non-transferable OEM license associated with a particular Dell machine/COA implies a licensing model for Windows similar to Redhat, i.e. the OEM license is tied to support, which is not truthful at all since the license is for the purpose of operating the computer.
Looking through many competitive software products was too much effort for those in all companies needing to make an IT decision. So .. just choose whatever is from Microsoft! If your company is smaller than [threshold], you will not be bothered by BSA, ..., -- our software is [effectively] free!
So now we have a huge number of small businesses running "Small Business Server" [pathetic] with pirated XP Pro installs. All of them are barely working, with server/workstations "virused up".
But try to talk to any decision makers at such companies "we like it the way it is -- we are in control!". [Even though we know everything about IT] we can't fix it -- you fix it [for free]. We will get our revenge against Bill G. for his crappy product (against you, the IT person).
The only answer is to let the businesses who have made the "Microsoft decisions" crash and burn, as they deserve.
Now with the WGA, "our recommended Linux solution [that is even worse than MSFT products] is from our great partners IBM, Novell, and Suse". It makes our [garbage] stuff look good!
There is the assumption that the PIN is transmitted "in the clear" across the internal ATM/Pinpad network. I only know about pinpads, but in pinpads the PIN is encrypted in hardware using a key that changes each time. I would assume ATMs are equally secure. In some countries, I understand the entire transaction (PIN, amount, account number) are encrypted in hardware, so trying to replay the encrypted packet is a wasted effort. Any lack of security is caused by the same forces at play everywhere, the desire to use the cheapest "almost competent" programmers. In all fields "Point of Sale" related, I'm guessing 99.9% is now good enough rather than 99.9999%. I assume ATM's will follow.
1) Find large customer considering Linux/having Linux problem
2) Demo easy fix
3) Find IBM/Novell/Suse got there first
4) Easy fix is not welcome, they want a broken system (need high headcount of idiots, still looking for fix done the "IBM way" (lots of Java written in India).
5) Get angry at IBM & friends, recommend M$ as something that sucks less/less evil
6) M$ solution implemented, difficult, expensive, unreliable, but still better than IBM/Suse/Novell crap.
7) ?? Profit (for M$ only). No wonder M$ & Suse work so well together.
It's happened to me twice this year, on two completely different hardware platforms. IBM had a broken "customized/ancient" version of the kernel attached to drivers for their "secret hardware interface". It is much easier to download M$ stuff (passport, etc.) than the secret stuff through Novell for IBM hardware/joint ventures.
Customers don't care, they either want IBM or M$. At a status meeting: "IBM/M$ is looking into the issue why (whatever) is totally slow/broken.
And we wonder why Linux is not getting anywhere with such great "friends". Be sure to recommend Suse/OpenSuse to newbies, we must honor our commercial hosts who contribute rather dubious value to Linux. OS/2 looks good relative to the IBM incarnations of Linux -- I laugh at the stupid customers who bought this IBM offering. It was funny to hear IBM remind customers "we are no longer in the PC business" just after "15 years from now, which vendor will be here to support your Linux/Java/..."
M$ is quite willing to relegate Linux to those "customers" (government, etc.) that require a broken, slow system to justify "how can we fix this mess we selected [because we wanted a broken system]. So Suse/M$ work together well as a team.
Sunbird/Lightning for Linux x86 have had a critical bug -- cannot click on 'recipients' field. Retested with last nights builds, bug still there. The product is basically useless with this bug.
Similarly, current Xulrunner 1.9 nightly builds run very slow -- need to revert to the version from Sept. 2005. Xulrunner is relevant because to help with Sunbird you need to know XUL. Learning XUL, you cannot even find a "Hello World" program to get started, instead the tutorials at xulplanet.com try to wow you with all the features in XUL (if you could _just_ get your first "Hello World" program working!). I had to work backwards from a "pacman" game (the only program I could find that actually had the three components JavaScript, CSS, and XUL working), to produce my own "Hello World" program. You can do some amazing things with XUL at that point. Mozdev actually has a program "Exch" in its "XUL hall of fame" that the author couldn't get the CSS working on due to their convoluted, poorly documented, and constantly changing "Chrome://", overlays, skins, etc.
The various Mozilla components share lots of code, and I see things like "Firefox and Thunderbird (and presumably Sunbird/Lightning) will run as Xulrunner applications once Xulrunner reaches 1.0" (1.9 ?> 1.0). Seems like that plan was abandoned, and the current xulrunner nightly build is barely running.
To summarize, it seems the entire Mozilla development process needs review to see why major components apparently have no simple test cases/documentation and the community is being asked to do a bug hunt on a non-working product.
With the SCO lawsuit dying out, it looks like MSFT needs new ways to entangle FOSS. Hence, firing execs using pure FUD against Linux, and putting on a friendly face on "open interfaces" and placing more source code we're not allowed to use out there. There is a certain class of cretin that will incorporate "shared source" code into an OSS project. The legal departments of large corporations are already terrified of FOSS (from the SCO lawsuit) and will require indemnification and eventually all OSS projects will have to go through rigorous audits to show they contain no code from tainted MSFT source releases. Thanks MSFT, for adding massive code auditing overhead to OSS development!
In the good old days, a new company (such as Vonage) would go public long before it was "discovered", allowing early investors to get rich (like Microsoft, for example). In today's world, major banks provide working capital and the objective is to delay the IPO as long as possible, so that only the banks and the founders make any real money. I'm predicting the disappointment seen today with Vonage is going to become the norm for technology IPO's.
When the system is working "correctly", you can go to the NOAA website at 8:05am and get the temperature, barometric pressure, etc. at 6:53am.
Kweather is even worse. The 'update now' button almost never works, so even if a 1.5 hour ago temperature is available it will try to cling onto it for an extra hour or more "to reduce network abuse" I guess; when I looked at the marvelous C++ source code I was amazed that someone could code a "fetch FTP page, parse, display in a KDE button" in 500000+ lines of incomprehensible C++.
If you need the current weather conditions, buy an expensive weather station; I think most of the junk in the stores has a precision of plus/minus 5 degrees F.