IE is probably bloated, even if users don't see it since code is shared with Windows, but the day microsoft would want to fix all these security bugs by making IE independent from Windows, this would probably be much clearer to the users.
And I agree that Mozilla is very memory hungry too. Under Linux, just after starting Mozilla (browser + calendar only, no mail/news module), my free (RAM + swap) memory decreases by 100 MB.
Installing the setmixer Debian package on the PowerBook can be dangerous for the ears too. I was listening to music with a headset when I installed it, and I won't do that again! After that, I'd had a headache for several hours.
Promptly fixed? I reported a security bug on June 12 with an example showing how one can execute arbitrary code (under Linux), and there are still no comments.
It doesn't need very specific requirements. Anyway, there is the same problem with messages sent to a wrong e-mail address (in particular, this is the case of many spam messages), and rejected messages should be bounced (at this time, one doesn't know if these messages are spam or not). Breaking the RFCs because some people don't know how to or don't want to configure their mailservers is a silly idea.
No need to change the mail server. You just need a special filter for mailer-daemons, in procmail or whatever, where you receive your mail. You may receive false challenge messages, but you'll never see them.
You should send the challenge message as a mailer-daemon (just like for other rejected messages). Then the receiver can check the message-id of the rejected message. If the message-id doesn't match some pattern used by the receiver, the mailer-daemon is discarded by the mail server (said otherwise, the message-id allows some form of authentication). I've been using this method for years and it works perfectly.
You could use challenge / response only for suspicious messages. At the same time, the author of the suspicious message would learn why his message was declared as probable spam, and therefore how to write a better message next time, fix his broken config and so on.
This reminds me the latest ad (in France) from Microsoft, which ends like (translated) "we're interested in her talent, we write software to allow her to express it". Apart from virus writers, I don't know whom this could be targeted to...
"The amount of times I've had apps go haywire and had to reset due to the OS using CMT is rather silly."
This is due to memory protection, not to CMT. But this is quite rare. When an application behaves very badly, I don't use it any longer.
Under Unix, though having pre-emptive MT, applications can also behave badly and completely freeze the machine; this is the case with Netscape for instance. You can also do a while(1) fork(); or something similar with lockfiles under NFS (both problems happened to me)...
"It's not monotasking, but CMT is far from optimal. I remember talk about Wimp2 when somebody implimented PMT, but it had a lot of trouble with the rest of the OS and applications being designed for CMT."
This isn't true. PMT isn't optimal either. Under Unix, a background task (even niced) can slow down the whole machine, and you can often get windows that are partly redrawn... Under RISC OS, tasks get interrupted at logical points, and the desktop generally has the priority, making it very responsive.
Slashdot Mirror
Well, if what you upload is something you have the rights to upload, then it will be completely legal (for you).
ASCII is flawed too, as "1" looks like "l" with many fonts, and it is too easy to be misled.
IE is probably bloated, even if users don't see it since code is shared with Windows, but the day microsoft would want to fix all these security bugs by making IE independent from Windows, this would probably be much clearer to the users.
And I agree that Mozilla is very memory hungry too. Under Linux, just after starting Mozilla (browser + calendar only, no mail/news module), my free (RAM + swap) memory decreases by 100 MB.
Installing the setmixer Debian package on the PowerBook can be dangerous for the ears too. I was listening to music with a headset when I installed it, and I won't do that again! After that, I'd had a headache for several hours.
Bug 246524
Promptly fixed? I reported a security bug on June 12 with an example showing how one can execute arbitrary code (under Linux), and there are still no comments.
It doesn't need very specific requirements. Anyway, there is the same problem with messages sent to a wrong e-mail address (in particular, this is the case of many spam messages), and rejected messages should be bounced (at this time, one doesn't know if these messages are spam or not). Breaking the RFCs because some people don't know how to or don't want to configure their mailservers is a silly idea.
No need to change the mail server. You just need a special filter for mailer-daemons, in procmail or whatever, where you receive your mail. You may receive false challenge messages, but you'll never see them.
You should send the challenge message as a mailer-daemon (just like for other rejected messages). Then the receiver can check the message-id of the rejected message. If the message-id doesn't match some pattern used by the receiver, the mailer-daemon is discarded by the mail server (said otherwise, the message-id allows some form of authentication). I've been using this method for years and it works perfectly.
You could use challenge / response only for suspicious messages. At the same time, the author of the suspicious message would learn why his message was declared as probable spam, and therefore how to write a better message next time, fix his broken config and so on.
This reminds me the latest ad (in France) from Microsoft, which ends like (translated) "we're interested in her talent, we write software to allow her to express it". Apart from virus writers, I don't know whom this could be targeted to...
"The amount of times I've had apps go haywire and had to reset due to the OS using CMT is rather silly."
This is due to memory protection, not to CMT. But this is quite rare. When an application behaves very badly, I don't use it any longer.
Under Unix, though having pre-emptive MT, applications can also behave badly and completely freeze the machine; this is the case with Netscape for instance. You can also do a while(1) fork(); or something similar with lockfiles under NFS (both problems happened to me)...
"It's not monotasking, but CMT is far from optimal. I remember talk about Wimp2 when somebody implimented PMT, but it had a lot of trouble with the rest of the OS and applications being designed for CMT."
This isn't true. PMT isn't optimal either. Under Unix, a background task (even niced) can slow down the whole machine, and you can often get windows that are partly redrawn... Under RISC OS, tasks get interrupted at logical points, and the desktop generally has the priority, making it very responsive.