Actually, current drives DO warn you when they're failing.
i have 30 gig unit here that used to be on my aunt's box. i replaced it because... SMART told me it was failing.
i attached a new unit on the box, mirrored the disk and took the bad one out.
SMART is an old technology already, is present in all IDE units and all motherboards i've seen in the last 5 or 6 years, but many people ignores it. trust me, worked once for me and my aunt, so download a SMART monitor and put it running along with your lm_sensors daemon.
Actually, the xfree radeon driver doesn't even have 3d support for the Radeon 9800 Pro, and the ATI binary driver doesn't work with WINE(x). They even said in the readme that it's winex's problem! Now I'm wondering why Nvidia's driver just works.
strange... I have a radeon 8500 and i can play counter strike and black&white inside winex with BOTH drivers.
i'm at the offie now, but as soon as i'm home, i can mail you a set of modified binary drivers (indicated by transgaming's tech support) that DO work with Winex AND have ST3C so i can play native UT2003.
i also patched my kernel with beta DRI kernel drivers bundled with XFree 4.3 source that alredy have support for r200 GPUs.
if you have a r250 or r300, wait till i get home so i can mail you those binaries. they work ok with X 4.3 and linux 2.4.18 (tested in an upgraded conectiva). for debian it needs alien and some tweaking.
if lycoris has perpetual right to distribute linux kernel under the GPL, all linus need to do is download a kernel tarball from lycoris, and is coding as usual
It annoys me that Linux is inexpensive, because I don't get the satisfaction of bragging to my neighbors about how much I paid for it.
send a $699 check to SCO and come back here to brag about it
It annoys me that I can upgrade Linux without having to purchase state-of-the-art expensive PC hardware, so my hardware is always outdated.
install Winex and star gaming. since Winex sometimes cause a 30% loss in framerate, it'll be a good reason to buy a Radeon 9800 pro
It annoys me that I don't have to reboot frequently, so I never know how fresh the bytes of code are in memory.
linux 2.6 pre1
It annoys me that I can easily solve my Linux problems in the Google groups section, instead of getting to speak with a real live tech support person, who might be a really cute blonde chick.
pay for it (also answers #1) and call redhat's tech support instead.
It annoys me that I don't have a mascot like Clippy the paperclip in the vi text editor.
vigor. www.userfriendly.org for back reference on it. then apt-get install vigor if you have debian.
if i got it right, the only way for SCO code to show up in linux is if it came from IBM ? dude, mail linus and ask him a list of patches and additions made to 2.3 and 2.4 series by caldera/SCO employees. this was already discussed here, and when things go to court IBM will CERTAINLY play this card.
and it still shows nothing on screen if i pass vga=normal during boot, and it took me several atempts before I relized that regular ps/2 keyboard can be left out or compiled as a module. well, this kind of changes were expected.
after i managed to get it working (booting, with keyboard, framebuffer console, et. all) surprise... no DRM on X.
happens that for some reason X doesn't detect working agp when a Radeon 8500LE in inserted in my kt266 based mobo. even with agpgart and radeon modules loaded.
so here's a few sugestions:
leave ps/2 kboard selected by default for x86 architectures, same for a way to display the console on text mode vga and check this radeon issue.
except those minor stuff, the new kernel is great. really fast for regular use.
"The SCO Group's suit against IBM could be a way to make SCO a more attractive takeover target"
"In Gartner's opinion, SCO's claim that IBM misappropriated trade secrets from AIX will be difficult to prove, because an enterprise OS consists of many components, including high-availability features, diagnostics, security, kernel hardening, scheduling and queue management."
it's in brasil's constitution since 1988. it's called "reply right" and states that everyone has the right of reply _proportional to the damage_
this means that if someone badmouths me, i have the right to defend myself.
is usual to see this right being used by politicians during campaigns. ans since the arcticle is generic, it applies to _ALL KINDS_ of media. doesn't matter if is writen, spoken or digital. you badmouth me in your blog and your blog is in brasil, i'll get a reply right. if you don't publish my reply i'll get a warrant to shovel the reply down your server's throat. simple, efficient and it's there since before internet went public.
maybe because the market found that that was a company name SCO after all and that it was a penny stock company playing with the big guys.
now that investors already multiplied their money the stocks are plumeting back to what they were. my guess is that sooner than later SCO'll be a penny stock company again.
don't know about other countries, but here in Sao Paulo (brasil) works like this:
every DSL card installed in a telco central accepts 10 lines. suposing all of them are 256 kbps down/128 kbps up, they'll do traffic shaping allowing 256 kbps down/128 kbps up for the WHOLE CARD. this way if all 10 users have kazaa running using all the bandwith available, each one will be using 25/12.8 kbps... crappy modem speed. and it's written in the contract they send to you when you sign that only 10% of the nominal speed is guaranteed, so unless it stops completelly you'll have to accept it.
they charge 90 brasilian bucks (30 US bucks) for the service, which means they make 900 from 10 users. since a 256 kbps frame-relay cost here about 900 bucks a month... they're at least not loosing money.
other bandwidth providers probably do the same thing.
is just like dial-up providers that have more users than modems and the summed bandwidth off all modems are biger than their connection to the internet.
remember keanu reeve's johnny mnemonic ? a cyberpunk data courrier carrying the cure for a brain disease caused by excessive exposure to information ?
science fiction appart, we're more and more exposed to wireless communications and other forms of electro-magnetic interference since the radio was invented and no one knows exactly what kind of consequences this exposure has over humans (if some one knows, tell me. all I heard until now is noise. ppl from the industry says one thing, environmentalist another...).
what if all this electro-magnetic noise around us proves to be harmfull ? remember that early in the 20th century doctors used to prescribe cigaretes to relieve stress, and it took a century or more before everyone, including the tobacco industry, to agree smoking was not exactly healthy. how many time it'll take before we're sure about radio signals all over us ?
original starwars was an unbelievably good trilogy, i agree. trminator isn't complete yet (i guess T3 is comming this year, right ?) so we can't judge. Aliens wasn't a trilogy since the whole series has four flics, and the third sucks, while the fourth stinks. so take only the first 2 movies in account when discussing the Aliens series.
from my own experience, SMTP is good for trusted environments, where don't have to worry with ppl abusing the system, this means a pre-commercial internet when the only users were govt, big companies and academia.
now the situation is completelly diferent, and SPAM exists because is easy to abuse the SMTP standard and forge messages.
some flaws inherent to SMTP:
- it knows nothing about users.
There's nothing in RFC822 (or even RFC2822) that says a mail transport agent (MTA. think on sendmail, exchange...) must check if the user sending the message really exists. things like authentication exists only to ensure the server isn't being used by unauthorized people but it leaves the task of filling the "From:" header to the mail user application (MUA, think eudora, outlook, pine, mutt...). the correct would be the MTA to authenticate the user and fill the "From:" header itself leaving only the "Reply-to:" as an optional to the MUA to fill.
- There's no way for the receiving MTA to check if the message is valid.
One way to implement this is to make the "Message-Id:" mandatory and formated in a standard way, and implement in the messaging protocal a way to check it against the sending domain's MX (Mail eXchange). it'll work like this: mydomain.com.br receives a message with "Message-Id: 1234567890@your-domain.com". then my MTA would do a DNS lookup for your-domain.com's MX and ask it if it realy sent mesage "1234567890@your-domain.com". if the answer is "no" my MTA would drop the message.
- The envelope can be easilly forged.
The envelope is the part of message that contains information such as the sender, the subject, from where the message came, and so on.
Each MTA fills the envelope it's own way (IMHO Qmail is lousy on this). the envolope should should follow a rigid standard, containing information of sender, originating IP, user and maybe even a digital signature of both the user and the server issued by some acredited institution.
even messages generated by some proccess running on the same server (such as mailforms or webmail applications that write messages directly to the mail queue) would be subject to a rigid header checking/rewriting by the MTA to ensure it's valid.
- Messages from the obsolete (current) version of the protocol would be tagged accordingly leaving to the recipient the decision of accepting or refusing them.
how would this few additions to the mail protocol help stop SPAM ? well, it wont. but it'll help prevent forged messages or spoofed headers and make it easy to filter messages while they'll still on the server.
making it hard for spammers to hide themselves behind fake From's or Received's would reduce the efficiency of their junk, leaving only honest, respectuff mail-marketers (if there is such a thing) on the business.
but then, this is just my humble opinion. I still have to work this out, but i think it'd be a nice subject for a colege term paper, and i still have 3 years to prepare it.
who used to promote thin clients with just enough to boot a barebones system and that would load everything else from the network, probably from a subscription based service ?
well, if it is, he was wrong then and is probably wrong now. my guess is that software industry still have a long road (10, maybe 15 years more) ahead before most of the IT industry moves to a service-centric model built around open-source code.
is M$ fault. their software doesn't use ipaq's power managment the way it should, which results in 24 hour batery time. with familia 0.6 I get 4X the battery time.
seriously now, ppl. is M$ software reliable enough to such operation ?
heck, the software they supplied to ISS caused serious problems to astronauts and cosmonauts before, and a movie theatre isn't as close as sensitive as a space station. I'd hate go to a movie and leave frustrated because the projector BSoDed, even if I receive my money back.
before I wiped WinCE in my ipaq to install linuux I noticed several anoying bugs, one of them the shor lived battery. owners of top-of-the-line BMWs also are being annoyed by bugs in the M$ software that runs on the car. M$ code simply cannot be trusted to drive some things.
some webservers already allow you to resume broken http downloads. some web browsers are too brain damaged to do it right. in this case use a download manager or copy the URL into as a wget parameter and let it handle it.
Slashdot Mirror
Actually, current drives DO warn you when they're failing.
i have 30 gig unit here that used to be on my aunt's box. i replaced it because... SMART told me it was failing.
i attached a new unit on the box, mirrored the disk and took the bad one out.
SMART is an old technology already, is present in all IDE units and all motherboards i've seen in the last 5 or 6 years, but many people ignores it. trust me, worked once for me and my aunt, so download a SMART monitor and put it running along with your lm_sensors daemon.
Actually, the xfree radeon driver doesn't even have 3d support for the Radeon 9800 Pro, and the ATI binary driver doesn't work with WINE(x). They even said in the readme that it's winex's problem! Now I'm wondering why Nvidia's driver just works.
strange... I have a radeon 8500 and i can play counter strike and black&white inside winex with BOTH drivers.
i'm at the offie now, but as soon as i'm home, i can mail you a set of modified binary drivers (indicated by transgaming's tech support) that DO work with Winex AND have ST3C so i can play native UT2003.
i also patched my kernel with beta DRI kernel drivers bundled with XFree 4.3 source that alredy have support for r200 GPUs.
if you have a r250 or r300, wait till i get home so i can mail you those binaries. they work ok with X 4.3 and linux 2.4.18 (tested in an upgraded conectiva). for debian it needs alien and some tweaking.
if lycoris has perpetual right to distribute linux kernel under the GPL, all linus need to do is download a kernel tarball from lycoris, and is coding as usual
It annoys me that Linux is inexpensive, because I don't get the satisfaction of bragging to my neighbors about how much I paid for it.
send a $699 check to SCO and come back here to brag about it
It annoys me that I can upgrade Linux without having to purchase state-of-the-art expensive PC hardware, so my hardware is always outdated.
install Winex and star gaming. since Winex sometimes cause a 30% loss in framerate, it'll be a good reason to buy a Radeon 9800 pro
It annoys me that I don't have to reboot frequently, so I never know how fresh the bytes of code are in memory.
linux 2.6 pre1
It annoys me that I can easily solve my Linux problems in the Google groups section, instead of getting to speak with a real live tech support person, who might be a really cute blonde chick.
pay for it (also answers #1) and call redhat's tech support instead.
It annoys me that I don't have a mascot like Clippy the paperclip in the vi text editor.
vigor. www.userfriendly.org for back reference on it. then apt-get install vigor if you have debian.
if i got it right, the only way for SCO code to show up in linux is if it came from IBM ? dude, mail linus and ask him a list of patches and additions made to 2.3 and 2.4 series by caldera/SCO employees. this was already discussed here, and when things go to court IBM will CERTAINLY play this card.
Works now. thanks for the tip.
DRI...
i said "DRM" before because it's what X spills out on it's log file...
I'll try that. thanks.
and someone pls mod rakarnik up...
and it still shows nothing on screen if i pass vga=normal during boot, and it took me several atempts before I relized that regular ps/2 keyboard can be left out or compiled as a module. well, this kind of changes were expected.
after i managed to get it working (booting, with keyboard, framebuffer console, et. all) surprise... no DRM on X.
happens that for some reason X doesn't detect working agp when a Radeon 8500LE in inserted in my kt266 based mobo. even with agpgart and radeon modules loaded.
so here's a few sugestions:
leave ps/2 kboard selected by default for x86 architectures, same for a way to display the console on text mode vga and check this radeon issue.
except those minor stuff, the new kernel is great. really fast for regular use.
all those nifty PC with gaming power comparable to the xbox will need an OS in order to run.
even if you go with gnu/linux, it'll nedd wineX (which costs money) or the machine will have limited gaming capabilities.
in SCO's (actually caldera's) site they have a copy of a gartner arcticle that contains parts that are a litle detrimental to SCO
g .h tml
http://www.caldera.com/scosource/gartner_warnin
some quotes:
"The SCO Group's suit against IBM could be a way to make SCO a more attractive takeover target"
"In Gartner's opinion, SCO's claim that IBM misappropriated trade secrets from AIX will be difficult to prove, because an enterprise OS consists of many components, including high-availability features, diagnostics, security, kernel hardening, scheduling and queue management."
it's in brasil's constitution since 1988. it's called "reply right" and states that everyone has the right of reply _proportional to the damage_
this means that if someone badmouths me, i have the right to defend myself.
is usual to see this right being used by politicians during campaigns. ans since the arcticle is generic, it applies to _ALL KINDS_ of media. doesn't matter if is writen, spoken or digital. you badmouth me in your blog and your blog is in brasil, i'll get a reply right. if you don't publish my reply i'll get a warrant to shovel the reply down your server's throat. simple, efficient and it's there since before internet went public.
maybe because the market found that that was a company name SCO after all and that it was a penny stock company playing with the big guys.
now that investors already multiplied their money the stocks are plumeting back to what they were. my guess is that sooner than later SCO'll be a penny stock company again.
don't know about other countries, but here in Sao Paulo (brasil) works like this:
every DSL card installed in a telco central accepts 10 lines. suposing all of them are 256 kbps down/128 kbps up, they'll do traffic shaping allowing 256 kbps down/128 kbps up for the WHOLE CARD. this way if all 10 users have kazaa running using all the bandwith available, each one will be using 25/12.8 kbps... crappy modem speed. and it's written in the contract they send to you when you sign that only 10% of the nominal speed is guaranteed, so unless it stops completelly you'll have to accept it.
they charge 90 brasilian bucks (30 US bucks) for the service, which means they make 900 from 10 users. since a 256 kbps frame-relay cost here about 900 bucks a month... they're at least not loosing money.
other bandwidth providers probably do the same thing.
is just like dial-up providers that have more users than modems and the summed bandwidth off all modems are biger than their connection to the internet.
my comment was taken from the movie because the example fitted my idea.
i'm not paranoid with the possibility of geting cancer from radio waves but I know ppl who are so i posted to get more info on the subject.
anyone here knows a trusted (as in neutral) source of information on the subject of radio waves x health ?
before moding me down, read...
remember keanu reeve's johnny mnemonic ? a cyberpunk data courrier carrying the cure for a brain disease caused by excessive exposure to information ?
science fiction appart, we're more and more exposed to wireless communications and other forms of electro-magnetic interference since the radio was invented and no one knows exactly what kind of consequences this exposure has over humans (if some one knows, tell me. all I heard until now is noise. ppl from the industry says one thing, environmentalist another...).
what if all this electro-magnetic noise around us proves to be harmfull ? remember that early in the 20th century doctors used to prescribe cigaretes to relieve stress, and it took a century or more before everyone, including the tobacco industry, to agree smoking was not exactly healthy. how many time it'll take before we're sure about radio signals all over us ?
original starwars was an unbelievably good trilogy, i agree. trminator isn't complete yet (i guess T3 is comming this year, right ?) so we can't judge. Aliens wasn't a trilogy since the whole series has four flics, and the third sucks, while the fourth stinks. so take only the first 2 movies in account when discussing the Aliens series.
from my own experience, SMTP is good for trusted environments, where don't have to worry with ppl abusing the system, this means a pre-commercial internet when the only users were govt, big companies and academia.
now the situation is completelly diferent, and SPAM exists because is easy to abuse the SMTP standard and forge messages.
some flaws inherent to SMTP:
- it knows nothing about users.
There's nothing in RFC822 (or even RFC2822) that says a mail transport agent (MTA. think on sendmail, exchange...) must check if the user sending the message really exists. things like authentication exists only to ensure the server isn't being used by unauthorized people but it leaves the task of filling the "From:" header to the mail user application (MUA, think eudora, outlook, pine, mutt...). the correct would be the MTA to authenticate the user and fill the "From:" header itself leaving only the "Reply-to:" as an optional to the MUA to fill.
- There's no way for the receiving MTA to check if the message is valid.
One way to implement this is to make the "Message-Id:" mandatory and formated in a standard way, and implement in the messaging protocal a way to check it against the sending domain's MX (Mail eXchange). it'll work like this: mydomain.com.br receives a message with "Message-Id: 1234567890@your-domain.com". then my MTA would do a DNS lookup for your-domain.com's MX and ask it if it realy sent mesage "1234567890@your-domain.com". if the answer is "no" my MTA would drop the message.
- The envelope can be easilly forged.
The envelope is the part of message that contains information such as the sender, the subject, from where the message came, and so on.
Each MTA fills the envelope it's own way (IMHO Qmail is lousy on this). the envolope should should follow a rigid standard, containing information of sender, originating IP, user and maybe even a digital signature of both the user and the server issued by some acredited institution.
even messages generated by some proccess running on the same server (such as mailforms or webmail applications that write messages directly to the mail queue) would be subject to a rigid header checking/rewriting by the MTA to ensure it's valid.
- Messages from the obsolete (current) version of the protocol would be tagged accordingly leaving to the recipient the decision of accepting or refusing them.
how would this few additions to the mail protocol help stop SPAM ? well, it wont. but it'll help prevent forged messages or spoofed headers and make it easy to filter messages while they'll still on the server.
making it hard for spammers to hide themselves behind fake From's or Received's would reduce the efficiency of their junk, leaving only honest, respectuff mail-marketers (if there is such a thing) on the business.
but then, this is just my humble opinion. I still have to work this out, but i think it'd be a nice subject for a colege term paper, and i still have 3 years to prepare it.
who used to promote thin clients with just enough to boot a barebones system and that would load everything else from the network, probably from a subscription based service ?
well, if it is, he was wrong then and is probably wrong now. my guess is that software industry still have a long road (10, maybe 15 years more) ahead before most of the IT industry moves to a service-centric model built around open-source code.
it works for TV and radio.
will it work for mp3 music ?
maybe...
but for what kind of music ? will some company sponsor the kind of music I want to hear ?
is M$ fault. their software doesn't use ipaq's power managment the way it should, which results in 24 hour batery time. with familia 0.6 I get 4X the battery time.
hmmmm...
millions of windows users, maybe ?
seriously now, ppl. is M$ software reliable enough to such operation ?
heck, the software they supplied to ISS caused serious problems to astronauts and cosmonauts before, and a movie theatre isn't as close as sensitive as a space station. I'd hate go to a movie and leave frustrated because the projector BSoDed, even if I receive my money back.
before I wiped WinCE in my ipaq to install linuux I noticed several anoying bugs, one of them the shor lived battery. owners of top-of-the-line BMWs also are being annoyed by bugs in the M$ software that runs on the car. M$ code simply cannot be trusted to drive some things.
Backups exist not only to prevent dataloss in case of disk malfunction. it exists also to protect against PEBKAC erros.
Last time I managed an office network I had to deal with at least "can you restore file xxxyyy.dwg from the backup ???"
so if disks are geting bigger, it's better for quantum to increase the size of their DLT tapes too.
some webservers already allow you to resume broken http downloads. some web browsers are too brain damaged to do it right. in this case use a download manager or copy the URL into as a wget parameter and let it handle it.
or the cat will make YOU nuts simply by ignoring the stuff, just like mine did when I bought an optical mouse.
go figure...