Aside from the fact that every game today looks exactly like the games we had 7-10 years ago, they're just too expensive. And PC games today have that annoyance of hearing the CD spin in the drive all the time because due to copy protection it refuses to run without it, then eventually failing to run if it gets too scratched. Then, 9/10ths of the games released in the last 4 years state that they require better hardware than came with my cheap 2 year old Dell. And finally, I have to be sure that they'll run on Linux before I buy them.
This morning I came to my computer to find that Google Talk had popped up five identical dialogs asking if I wanted to send my logs to my gmail account.
The less code there is, the fewer bugs you should have. This sounds like at least two or three programs, a GUI/controller and the nodes. Try to avoid threading, or any situation where there might be unexpected timing related interactions. And you make it clear that C++ is unavoidable, but does it all have to be C++? Or just the parts that need to be fast? The GUI could be managed. The node could be split into a managed program that starts an unmanaged program with a set of parameters and reads the results. Even if both parts were unmanaged, you'll still have a system where a single crash wouldn't crash the node entirely, but likely just the worker half of the node, which can be restarted by the monitoring half. And you mitigate a lot of memory leaks if your worker processes get restarted from time to time.
As far as how to handle crash recovery, whether the program should fight tooth and nail to overcome the error, ignore (but log) the error, crash and restart upon error, or crash until it's restarted manually, it really depends on the severity of possible errors, and the consequences of failure vs incorrect behaviour. Is this a program that you'll be unable to patch in a timely fashion if it breaks? Obviously you'll want to run a lot of stress tests before the thing goes live, and consider how it'll respond to bad data.
I have a credit card processing proxy (about 3 pages of javascript) in place to protect us from the outright dangerous credit card handling in our ERP system. The proxy just sits in the middle and blocks the unwanted transactions, returning fake replies to make the ERP system happy. Apparently our use of the ERP system is a little atypical, like sometimes shipping orders and charging cards on the same day the order was taken. Because its job is so important, if there is an error, I'd rather have it crash than take a chance that it might do something really bad. Most errors crash the proxy outright. The proxy is run by another script that mostly just sends me a message when it crashes. I like to live just a little dangerously, so the proxy gets restarted once after its first crash. But if it crashes a second time, I must restart it manually.
On the other hand, I sometimes write code that ignores most unexpected errors, so long as it's worse to completely fail than to partly fail, and it's not doing anything expecially important or dangerous. For important things, it's often better to fail than to not fail when it should.
I have a distant relative who occasionally does support for a company that has an old FreeBSD server. He was excited to learn that I'd be able to fix it, because he's only used Windows. It was misconfigured, causing it to break many years later when their dhcp server gave it a new IP address. If if BSD died, BSD servers will be around for many more years.
The big problem with Java and.NET is that they require runtimes which are not installed by default on Windows or most other operating systems, and both runtimes have strict redistribution restrictions, so they're not the platforms of choice for developers who like to write standalone applications that can be installed by ordinary consumers, but they're fine for business software. Microsoft is smothering its baby here. Additionally,.NET has nothing to offer for those wanting to write cross platform software, while Java does. Vista might fix the distribution part of the problem, but Microsoft has no desire to assist cross platform open source implementations of.NET, and threatens them with several of their.NET related patents.
Application developers want the ease of VB, the speed and power of C++, the cross platformness of Java, and the ability to produce standalone executables, free of large dependencies that that aren't preinstalled on most systems. I'n not much of a pascal programmer anymore, but Delphi/Kylix would rule the world (of rapid standalone consumer application development) if it wasn't so expensive and proprietary, and if Kylix wasn't abandoned. Maybe there's still a chance if the Lazarus project matures to production-ready.
The whole dead end technology thing is a bit of a non-argument though. There's nothing wrong with learning a new language or platform every five years, and nothing preventing the maintenance of legacy systems written with older technologies. There are a lot of complex, expensive ERP systems that still use COBOL, and are still finding new users.
It's good to know more than one language. A well rounded MIS+ developer might want to learn Java, C#, VB.NET & VBA, SQL (including DDL), several scripting languages (including web related), and Windows Server and Linux administration. And know how to use reporting programs like Crystal Reports. I'm a lone developer at a small business though, so my job responsibilities are a little broad. It really depends on where you expect to work. One language plus SQL might be all you need. Unfortunatly, at some companies that language might be VB.
You're encouraged (by success) to exploit the system. There's no reason why you should lose potential income for having a conscience, while the government rewards everyone else for being greedy. You don't want to stop just yourself from patenting software. You want to stop EVERYONE from patenting software. The system will survive just fine without your participation. You're abstinence just rewards all the other greedy software patenters out there. Someone will just take your place, thus ensuring your personal loss will exceed the public gain. Someone's getting rich whether you like it or not. I'm sure you'd rather it be you than some creep who actually supports software patents and is going to lobby to keep them.
That's not a great comparison. Several AV companies do give away stuff for free. Many have free removal tools for every major virus, free online scans, free standalone virus scanners, and sometimes free antivirus suites (AVG for example). And Microsoft doesn't lose sales by giving away virus removal tools. And Microsoft is the only company in a position to push an updated removal tool onto unsuspecting infected users.
How much could pushing an early update cost? How often before it's more than a drop in the bucket? They want users to think Microsoft cares about security right? They've learned nothing from all the bad press they got just a few weeks ago for initially deciding not to patch the WMF vulnerability. When there's an passive threat, like undisclosed vulnerability, then sure, it's good to get all the patches out in one set. It reduces work and risk. But immediate threats need immediate attention. Security problems don't wait for patch tuesday. Microsoft is just being lazy, and like every time before it's going to backfire. Not every virus makes headlines, and it's very rare for a virus to delete documents, especially on all network shares. Most viruses are profit or ego inspired, and it'll probably continue like that. When a highly destructive Windows virus comes around, and Microsoft can stop it with a small, inexpensive, low risk update to their Malicious Software Removal Tool, they shouldn't give up the opportunity. At the very least it's good PR.
Instead, it'll look bad for them because of news reports that they turned down an easy opportunity to mostly kill a seriously damaging and widespread virus, in favor of keeping with their normal monthly release schedule, knowing that by then all the damage that could happen will have already happened.
Because it deletes files on the "hidden" administrative file shares (enabled by default on Windows, and writeable by any domain admin), there's a strong chance that some companies running poorly secured Windows domains will lose every document on every computer by just having a single infected computer on their network on Friday.
Unlike most viruses, worms, and other malware, this one is going to cause some people some serious damage, and Microsoft is the position of being able to stop it on every system running auto-update.
It's like if you see someone robbing your neighbor's house, and not just stealing a few items, but unloading the whole thing into a truck. Not only that, but they're going door to door, unloading every house on your block but yours. And you're their landlord. Sure, they've paid you, but not to watch out for their stuff. So do you call the police? Or grab a beer and enjoy the show? There's no direct, immediate profit in it after all.
I used to get occasional crashes starting with the Deer Park builds before it was called 1.5, but they seemed to go away completely after I installed the flashblock extension and disabled java. Lots of flash ads across several tabs is a recipe for disaster. There's still a Linux-only tab dragging bug (drag doesn't end when button is released) that annoys me in 1.5, but I feel it beats the a Linux-only performance bug that annoyed me in 1.0.
Sure, XP is the now the most popular Windows, 4 years after its release, but it was a flop compared to the previous releases (http://www.pcworld.com/news/article/0,aid,76724,0 0.asp). Businesses are happy with Windows 2000. It took XP two years to overcome 2000 in the W3Schools browser statistics, and 2000 is still pretty strong at 13%. XP is most popular today because it comes preinstalled on new systems, and all the older systems are gradually dying of hardware failures or obsoletion.
A lot of people won't adopt a new standard until it offers a clear benefit over what they're already using. So far, most of what you get from XHTML is a bulkier page with more restrictions that no longer displays in 1% of browsers. XHTML 2 will work in even fewer browsers. XHTML 1.0 has been around for 6 years, and hardly anyone uses it. IE tries to parse it as HTML. Browsers parsing XHTML often don't fail gracefully. Sometimes the slightest typo results in a blank page. Google doesn't even use DOCTYPE tags, just plain HTML, and it works in every browser.
When fighting the kind of malware that installs itself to dozens of executables and dlls, to revive itself later, you can usually isolate most of that crap by searching by creation date, first making sure that explorer shows hidden and system files, and that the search doesn't exclude them.
You may need to disable system restore to remove some malware, or else Windows will automatically reinfect itself when it sees the files are missing. Reenable it before installing any new/updated drivers, as that seems to be when I need it most often.
Just in case, before you delete a bunch of stuff and reboot, check HKLM\Software\Microsoft\Windows NT\CurrentVersion\Winlogon\Userinit to ensure that it's not pointing to the malware, but to userinit.exe, wherever that is. Messing with userinit can render a system so that you can't log in, even in safe mode. XP SP2 might have fixed this, as I've seen some newer systems survive a broken userinit, or completely ignore it.
Also, empty out your host file (usually c:\windows\system32\drivers\etc\hosts on XP) to prevent browser hijacks.
If you suspect a rootkit, try a detector like rootkitrevealer. It won't remove it, but it might find it. Last resort: take your hard disk and slave it on another system, and remove the infected files.
Stinger is a good standalone virus scanner, and a small download
For future reference: Stop using IE and Outlook Express. Stop downloading free screensavers and other freebies, unless you get them directly from the author's website, and you trust them completely. I've seen places take my own shareware screensavers, bundle them with spyware, and redistribute them without permission or any regard for legality or morality.
We are always disappointed when software companies force us to publish details of vulnerabilities before making a fix available.
As bad as it is to publish unpatched vulnerabilities, it's worse if a company chooses to ignore security altogether. Ignoring security and suppressing vulnerability reports demands that vulnerabilities be published. People generally won't publish vulnerabilities if they see that the company it taking them seriously.
Slashdot Mirror
Aside from the fact that every game today looks exactly like the games we had 7-10 years ago, they're just too expensive. And PC games today have that annoyance of hearing the CD spin in the drive all the time because due to copy protection it refuses to run without it, then eventually failing to run if it gets too scratched. Then, 9/10ths of the games released in the last 4 years state that they require better hardware than came with my cheap 2 year old Dell. And finally, I have to be sure that they'll run on Linux before I buy them.
That's your signal to run for your life.
This morning I came to my computer to find that Google Talk had popped up five identical dialogs asking if I wanted to send my logs to my gmail account.
The less code there is, the fewer bugs you should have. This sounds like at least two or three programs, a GUI/controller and the nodes. Try to avoid threading, or any situation where there might be unexpected timing related interactions. And you make it clear that C++ is unavoidable, but does it all have to be C++? Or just the parts that need to be fast? The GUI could be managed. The node could be split into a managed program that starts an unmanaged program with a set of parameters and reads the results. Even if both parts were unmanaged, you'll still have a system where a single crash wouldn't crash the node entirely, but likely just the worker half of the node, which can be restarted by the monitoring half. And you mitigate a lot of memory leaks if your worker processes get restarted from time to time.
As far as how to handle crash recovery, whether the program should fight tooth and nail to overcome the error, ignore (but log) the error, crash and restart upon error, or crash until it's restarted manually, it really depends on the severity of possible errors, and the consequences of failure vs incorrect behaviour. Is this a program that you'll be unable to patch in a timely fashion if it breaks? Obviously you'll want to run a lot of stress tests before the thing goes live, and consider how it'll respond to bad data.
I have a credit card processing proxy (about 3 pages of javascript) in place to protect us from the outright dangerous credit card handling in our ERP system. The proxy just sits in the middle and blocks the unwanted transactions, returning fake replies to make the ERP system happy. Apparently our use of the ERP system is a little atypical, like sometimes shipping orders and charging cards on the same day the order was taken. Because its job is so important, if there is an error, I'd rather have it crash than take a chance that it might do something really bad. Most errors crash the proxy outright. The proxy is run by another script that mostly just sends me a message when it crashes. I like to live just a little dangerously, so the proxy gets restarted once after its first crash. But if it crashes a second time, I must restart it manually.
On the other hand, I sometimes write code that ignores most unexpected errors, so long as it's worse to completely fail than to partly fail, and it's not doing anything expecially important or dangerous. For important things, it's often better to fail than to not fail when it should.
I have a distant relative who occasionally does support for a company that has an old FreeBSD server. He was excited to learn that I'd be able to fix it, because he's only used Windows. It was misconfigured, causing it to break many years later when their dhcp server gave it a new IP address. If if BSD died, BSD servers will be around for many more years.
.NET is that they require runtimes which are not installed by default on Windows or most other operating systems, and both runtimes have strict redistribution restrictions, so they're not the platforms of choice for developers who like to write standalone applications that can be installed by ordinary consumers, but they're fine for business software. Microsoft is smothering its baby here. Additionally, .NET has nothing to offer for those wanting to write cross platform software, while Java does. Vista might fix the distribution part of the problem, but Microsoft has no desire to assist cross platform open source implementations of .NET, and threatens them with several of their .NET related patents.
The big problem with Java and
Application developers want the ease of VB, the speed and power of C++, the cross platformness of Java, and the ability to produce standalone executables, free of large dependencies that that aren't preinstalled on most systems. I'n not much of a pascal programmer anymore, but Delphi/Kylix would rule the world (of rapid standalone consumer application development) if it wasn't so expensive and proprietary, and if Kylix wasn't abandoned. Maybe there's still a chance if the Lazarus project matures to production-ready.
The whole dead end technology thing is a bit of a non-argument though. There's nothing wrong with learning a new language or platform every five years, and nothing preventing the maintenance of legacy systems written with older technologies. There are a lot of complex, expensive ERP systems that still use COBOL, and are still finding new users.
It's good to know more than one language. A well rounded MIS+ developer might want to learn Java, C#, VB.NET & VBA, SQL (including DDL), several scripting languages (including web related), and Windows Server and Linux administration. And know how to use reporting programs like Crystal Reports. I'm a lone developer at a small business though, so my job responsibilities are a little broad. It really depends on where you expect to work. One language plus SQL might be all you need. Unfortunatly, at some companies that language might be VB.
The 1.5.0.1 release is official now.
Everyone knows that passwords belong on sticky notes.
You're encouraged (by success) to exploit the system. There's no reason why you should lose potential income for having a conscience, while the government rewards everyone else for being greedy. You don't want to stop just yourself from patenting software. You want to stop EVERYONE from patenting software. The system will survive just fine without your participation. You're abstinence just rewards all the other greedy software patenters out there. Someone will just take your place, thus ensuring your personal loss will exceed the public gain. Someone's getting rich whether you like it or not. I'm sure you'd rather it be you than some creep who actually supports software patents and is going to lobby to keep them.
If they rebuild it, people will repurchase it.
Use something else to deliver your payload.
That's not a great comparison. Several AV companies do give away stuff for free. Many have free removal tools for every major virus, free online scans, free standalone virus scanners, and sometimes free antivirus suites (AVG for example). And Microsoft doesn't lose sales by giving away virus removal tools. And Microsoft is the only company in a position to push an updated removal tool onto unsuspecting infected users.
How much could pushing an early update cost? How often before it's more than a drop in the bucket? They want users to think Microsoft cares about security right? They've learned nothing from all the bad press they got just a few weeks ago for initially deciding not to patch the WMF vulnerability. When there's an passive threat, like undisclosed vulnerability, then sure, it's good to get all the patches out in one set. It reduces work and risk. But immediate threats need immediate attention. Security problems don't wait for patch tuesday. Microsoft is just being lazy, and like every time before it's going to backfire. Not every virus makes headlines, and it's very rare for a virus to delete documents, especially on all network shares. Most viruses are profit or ego inspired, and it'll probably continue like that. When a highly destructive Windows virus comes around, and Microsoft can stop it with a small, inexpensive, low risk update to their Malicious Software Removal Tool, they shouldn't give up the opportunity. At the very least it's good PR.
Instead, it'll look bad for them because of news reports that they turned down an easy opportunity to mostly kill a seriously damaging and widespread virus, in favor of keeping with their normal monthly release schedule, knowing that by then all the damage that could happen will have already happened.
Because it deletes files on the "hidden" administrative file shares (enabled by default on Windows, and writeable by any domain admin), there's a strong chance that some companies running poorly secured Windows domains will lose every document on every computer by just having a single infected computer on their network on Friday.
Unlike most viruses, worms, and other malware, this one is going to cause some people some serious damage, and Microsoft is the position of being able to stop it on every system running auto-update.
It's like if you see someone robbing your neighbor's house, and not just stealing a few items, but unloading the whole thing into a truck. Not only that, but they're going door to door, unloading every house on your block but yours. And you're their landlord. Sure, they've paid you, but not to watch out for their stuff. So do you call the police? Or grab a beer and enjoy the show? There's no direct, immediate profit in it after all.
You can try the 1.5.0.1 nightlies, which are supposed to contain a lot of crash fixes and other major bugfixes. http://ftp.mozilla.org/pub/mozilla.org/firefox/nig htly/latest-mozilla1.8.0/
I used to get occasional crashes starting with the Deer Park builds before it was called 1.5, but they seemed to go away completely after I installed the flashblock extension and disabled java. Lots of flash ads across several tabs is a recipe for disaster. There's still a Linux-only tab dragging bug (drag doesn't end when button is released) that annoys me in 1.5, but I feel it beats the a Linux-only performance bug that annoyed me in 1.0.
In a few years, they'll discover that 1 in 200 obese people have this terrible obesity virus, compared to 1 in 1000 non-obese people.
Now, if Windows Vista runs slow all the time, it'll be because of viruses/malware, not because of virus protection.
FTA: 156.33.*.*
Maybe this'll come in handy someday. Can't imagine what I'd use it for though.
Sure, XP is the now the most popular Windows, 4 years after its release, but it was a flop compared to the previous releases (http://www.pcworld.com/news/article/0,aid,76724,0 0.asp). Businesses are happy with Windows 2000. It took XP two years to overcome 2000 in the W3Schools browser statistics, and 2000 is still pretty strong at 13%. XP is most popular today because it comes preinstalled on new systems, and all the older systems are gradually dying of hardware failures or obsoletion.
A lot of people won't adopt a new standard until it offers a clear benefit over what they're already using. So far, most of what you get from XHTML is a bulkier page with more restrictions that no longer displays in 1% of browsers. XHTML 2 will work in even fewer browsers. XHTML 1.0 has been around for 6 years, and hardly anyone uses it. IE tries to parse it as HTML. Browsers parsing XHTML often don't fail gracefully. Sometimes the slightest typo results in a blank page. Google doesn't even use DOCTYPE tags, just plain HTML, and it works in every browser.
Code viewed under an agreement that impairs the viewer's right to reimplement the protocols doesn't count.
With respect to any offers by Microsoft to share their super secret Windows code at an affordable price, I'll believe it when I see it.
When fighting the kind of malware that installs itself to dozens of executables and dlls, to revive itself later, you can usually isolate most of that crap by searching by creation date, first making sure that explorer shows hidden and system files, and that the search doesn't exclude them.
You may need to disable system restore to remove some malware, or else Windows will automatically reinfect itself when it sees the files are missing. Reenable it before installing any new/updated drivers, as that seems to be when I need it most often.
Just in case, before you delete a bunch of stuff and reboot, check HKLM\Software\Microsoft\Windows NT\CurrentVersion\Winlogon\Userinit to ensure that it's not pointing to the malware, but to userinit.exe, wherever that is. Messing with userinit can render a system so that you can't log in, even in safe mode. XP SP2 might have fixed this, as I've seen some newer systems survive a broken userinit, or completely ignore it.
Also, empty out your host file (usually c:\windows\system32\drivers\etc\hosts on XP) to prevent browser hijacks.
If you suspect a rootkit, try a detector like rootkitrevealer. It won't remove it, but it might find it. Last resort: take your hard disk and slave it on another system, and remove the infected files.
Stinger is a good standalone virus scanner, and a small download
For future reference: Stop using IE and Outlook Express. Stop downloading free screensavers and other freebies, unless you get them directly from the author's website, and you trust them completely. I've seen places take my own shareware screensavers, bundle them with spyware, and redistribute them without permission or any regard for legality or morality.
We are always disappointed when software companies force us to publish details of vulnerabilities before making a fix available.
As bad as it is to publish unpatched vulnerabilities, it's worse if a company chooses to ignore security altogether. Ignoring security and suppressing vulnerability reports demands that vulnerabilities be published. People generally won't publish vulnerabilities if they see that the company it taking them seriously.
Those space shuttles are pretty expensive IIRC. They're not going to save much money by putting them at risk.