I thought the whole beauty of open source was that the users/developers would make drivers as they needed them? The linuxppc guys can reverse engineer the mac motherboards, XFree86 people write their own video drivers when none are made available, but no one can take the time to figure out the modems that seem to appear in almost every shipping PC? What gives?
I thought the whole thing was that people liked the units but wanted to do stuff with them that didn't require their service? What's the difference between 3 months and 2 years if you have no intention of ever using it?
They really just should sell the machines at cost + 15% or 20% (regular profit) and then offer multihundred dollar rebates to people that sign on for 3 years of service, a la MSN, AOL, Compuserve, etc... If you want their service, you'd get the rebate. If you don't, then you buy the machine from them at a cost that makes everyone happy.
No one should be expected to lose money selling their goods, just because that's what the customers want.... Which seems to be the case of netppliance. They offered a good deal, didn't realize that people would find a way to use their device without their service and pretty much stood to get completely screwed, losing a few hundred bucks per machine with no hope of recouping it through ISP fees.
I just put together a system as well... It ended up being a Power Mac 8100/80, 72 megs of RAM, FWB Jackhammer SCSI card, 2x4GB Barracuda's, and a Radius VideoVision Studio card... All bought on eBay separately for about $700.00. The Mac has built in ethernet, so i can just ftp the files back and forth to my main (Win NT Workstation) machine... Overall, I'd say I'm quite happy with the set up and quality... I have a big investment in S-VHS and Hi8 gear that i didn't want to have to replace yet, and that saved it!
Actually, for video, RAM isn't really all that important... Drive bandwidth rules... as well as processing power. Why? Most any video of any length and quality is going to exeed the amount of RAM in the system, and in the end, worst case scenario (no compression) broadcast video is only going to need around 30 megabytes/second. That's an incredibly small piece of bandwidth, ram wise. Compressed video will be 5-7 megs a second (tv quality), which most hard drives can handle these days.
The processors important especially if you're working with compressed video... But while in the editting phase of a project, don't use any codec's like Sorenson or Cinepak... They just take TOO LONG to compress. Apple's generic "video" codec works fine, and doesn't hit the CPU nearly as hard.
The iMac DV's are nice little self contained systems... Their main drawbacks would be the lack of a second monitor and lack of slots, which limits them to only being able to use only FireWire camera's. iMovie may be lame, but you can buy Adobe Premiere and have a nice, cute little system...
Further up the scale, a Mac G4 couple with a Targa 1000 or 2000 video card would be a great choice for editting, because you've got all the expandability of the G4's, plus the ease of use and plug and play of the Mac... Video capture cards are very finnicky creatures, so it's nice to not have to worry too much about if device 1 will work with device 2 while on motherboard 3.
If you really want to have 3D effects, an NT based system would probably be the way to go, since most of the 3d developers target NT workstation at the low end, and either Solaris or Irix at the high end. Yes, there is some stuff available for the mac (Lightwave, though it lacks the 3rd party plug in support of NT, Infini-D, Electric Image, Strata Studio Pro all ship on the mac... missing from that list is 3D Studio, most notably).
Moving past the low end, you mightt also want to check out systems from Avid and Media 100... They sell turnkey solutions, based on the Mac OS, Windows NT, and Irix.
Without knowing budgets or goals, it's hard to recommend a video editting solution... One thing is, there just aren't any open source tools, or tools that run on the open source operating systems, that can stack up against the proprietary tools.
Lastly... If you end up on a Mac or Windows machine, you'll probably want Adobe Premiere and Adobe After Effects for your editting and compositing. Throw in Photoshop and Illustrator for titling, and you've hit almost $2000 on just software, so be warned it's not at all cheap!
I too would really like to start using OpenBSD... It just seem like an concept I'm ready for... But aside from installing the MacOS, Windows 95, 98, NT Workstation 4, NT Server 4, OpenLinux and Redhat Linux, I'm just not sure if i've the actual know-how to pull off an OpenBSD install... Can anyone share their experiences with me/us?
Is it a really technical process, or is it like the others where you just plop in a CD, choose some options, set some settings, sit back and babysit and viola! you've got an almost working machine? Somehow i doubt it's like that, though.
The same exact thing can be said of Linus and Linux. Just look at the hordes of "cultists" that conglomerate slashdot. Linus', ESR's and RMS' emails are almost scriptures of the open-source community.
The same could be said of the many BSDer's who rush to defend their OS of choice.
And of the many still amiga-philes.
Just because you don't like apple doesn't mean it's bad.
By mainstream, he meant consumer not server. Linux is not mainstream by that definition. Neither is FreeBSD. Neither is Windows NT. Neither is Netware.
Your number might be correct about market shares, but the last time i checked, 5% of the desktops is a WHOLE LOT MORE than 17% of servers.
But do SPARCs even exist that run at 500 MHz? And how much do they cost? And according to your calculations, which i recreated and got the same results with, you scaled the speed of the entire system, rather than just the speed of the processor. I don't think the system bus speed in a SPARC system scales linerally with the addition of faster processors, so anything that leaves the L1 cache would hit the memory bus and slow down a little bit.
It's hard to predict the speed difference of machines unless you take everything into account. A faster CPU is just one part of the equation.
And what if i don't WANT the source? I have to sit through an extra block of time downloading stuff i have absolutely zero interest in? And isn't that one of the issues people have with microsoft, shoving unwanted things down everyone's throats?
Leave the source where it is... if someone wants it they can get it... if they don't want it, they dont' have to have it.
You're forgetting that MS doesn't have monopoly positions in Server OS's or Embedded OS's...
In your regulation of Microsoft, what committee would determine which features, utilities, and API's would be appropriate for an operating system and which should go into applications? How quickly would that committee act?
It seems that Microsoft was found to have a monopoly on Windows. I think the best solution is a horizontal breakup, where 3 or 4 companies would be created, each with access to all of Microsofts patents, trademarks, and copyrights. Prevent employees from moving between companies for 10 years. Then set those companies lose on each other.
They'll all have the cash to compete. They can do whatever they want... If that means lowering their prices, maybe we'll see $15 windows licenses. Maybe one of the companies might decide to open source all of their non-patented softwares.
I don't think a DOJ solution should impose Linux as a competitor to windows and completely tilt the playing field for Linux by saying things to Microsoft like "No including browsers. No including development tools. No including servers, etc..." because every other company which has an OS does the same...
We just need 3 versions of windows to ensure that hey all have competition, keep prices in check, etc... There could be some clause in there saying that if in 5 or 10 years any of those companies had decimated the others, some other intervention could be sought.
I don't know. I'm not a lawyer. I just think the punishment should focus on consumers and windows, and forget about linux.
In case you hadn't, all the Linux darling have been taking quite a beating as of late:
ANDN's at 23% of it's high CALD's at 71% of it's high COBT's at 27% of it's high CORL's at 23% of it's high LNUX's at 19% of it's high RHAT's at 28% of it's high
Where as with the dot com's,
AMZN's at 59% of it's high LCOS's at 75% of it's high YHOO's at 68% of it's high
In all those cases of me saying high, i meant 52 week highs...But those are the flagships of the Linux companies and the dotcoms. Hate to say it, but Linux has lost it's ring, it appears. The new trends are PDA's and wireless... Handspring's well poised to take advantage of that infatuation.
I still do wish that Handspring would manage to ship their products in quantity prior to their IPO... Palm had been shipping Pilots for years before 3Com spun them off. Handspring just develops a device and then says "time for the IPO!"
I wonder what the perfomance of a MAME running on Windows 98 VM running inside of Vmware running on Windows NT running inside of Vmware running on Redhat running inside of Connectixes Virtual PC running on a Mac G4/450 would be?
Aside from MAME, it'd be interesting to run some standard benchmark, or nearly standard, like Bytemark or SPEC (if someone with the $$$ for SPEC was interested) to see how much the CPU bogs down with each layer of emulation. Theoretically, if VMWare just passes x86 instructions to the processor, each additional instance should only suffer a small bit, rather than getting completely mangled.
Alternatively, one could run slackware inside of debian inside of openlinux inside of redhat inside of turbo linux... yikes!
It's all dependant on completely arbitrary factors. Like, who are the types of visitors? Are they loyal? Do they have money to spend? Is your site based on any of the now popular keywords? Is the domain of any value of itself?
Although it's kind of extreme, look at the market caps vs. actual revenues of Lycos vs. Yahoo... Yahoo has like 10x more visitors, 10x more market cap, but only 2x more earnings... So it's hard to quantify what makes a site valuable.
With ORBS, one could test a server to see if it will relay mail for you. If it does, you know, and orbs tells them. You can spam for up to 30 days before ORBS notifies the rest of the spammers, that that machine is now available. The spamming continues until the machine is blackholed.
Why publicize it? Why not just wait until the machine is actually used for spam, and then gothrough the steps of RBLing it? It's that whole innocent until proven guilty thing... They haven't done anything wrong, there's just potential for wrong to be done.
What is Darwin OS? Darwin is a complete operating system based on the foundation technologies in Mac OS X Server. It is an advanced BSD Unix system which offers advanced networking, services such as the Apache web server, and support for both Macintosh and Unix filesystems.
The reason that they and other censor ware vendors keep their lists secret is because that's their only asset. Access to the list is the only thing they're selling. If vendor A came along and released their list of 10,000,000 sites, in an open fashion, then vendor B could take those 10,000,000 sites, add 5,000,000 more, and have a 15,000,000 site list... C could come along and do nothing but still release a 15,000,000 name list.
I know, that's the same issue that's being faced with GPLed software, as companies like Mandrake and LinuxOne show, but I don't think that things such as support are nearly as much of an issue for Microsystems as compared to Redhat. Microsystems just needs to hold someones hands for 15 minutes, if that, where as Redhat has a much larger role to play with their customers.
While in the long term, security through obscurity doesn't work, in the short term, I'd think that it's a workable solution. I wish that the lot of ethical hackers would announce their discovery's of bugs and holes to the company's whose products have them and, depending on the complexitiy of the issue, hold off on making any announcements until a fix is at least available.
I mean, yes, the more skilled crackers will still have a chance to find out and exploit the holes, but it might at least prevent some of the script kiddie utilities from popping up, until at least a fix is available. If people get hit and there is no fix, it's basically terrorism (on a much larger scale). If people want to browse the web, they've got few real choices, so it's not like they can go "oh... IE sucks, i'll switch to netscape. Oh, netscape sucks too, i'll switch to Opera, oh... opera sucks, what do i do now?"
If fixes are available, and people still get "hit" by the script kiddies, at least they have a some one to point their finger at, with that person being themselves.
I think it all has to do with who installed it and who's using it. My friends have WinNT boxes that crash constantly, as well as Win98 boxes that seem to crash every other time they attempt to connect to the internet, where as my Win98 box has been running fine since 98 came out. The only hiccup it's ever had was when i upgraded the video card... Which unfortunately had the side effect of curbing my interest in Linux, being that the card doesn't seem to want to work with Redhat 6.0 in even VGA mode... command line is all i can get to now.
It all goes to say, it doesn't really matter so much about the operating system, it's who's using the machine.
A simple break up just won't do anything... There'll still be one company with a monopoly on the desktop OS market, another one with a monopoly on the productiviy suite market, and an internet company will a few billion in it's coffers. That's far from leveling the playing field.
So far as shares go... I'd almost think people would want MSFT broken up if the performance on ATT and the baby bells are any indication of what would happen to the Baby Bills. In the time frame since the breakup, IBM has increased something like 200% in value, while all the baby bells and ATT went up 800 or 1400% in the same period... I don't remember where i read it, though it was probably on news.com.
As far as I've always known... Tour's only really make money for the REALLY SMALL and REALLY BIG BANDS... all the medium sized bands tour mainly to promote their record sales. And record stores take a much larger chunk than just the $1. It's more around $3-4 on up.
But you're into music right? Read an essay by steve albini (a la big black, shellac, rapeman fame... along with engineering a slew of albums). It pretty much shows artists getting hosed already... But taking away their CD royalties pretty much destroys their means of recording as a means to support their careers.
That's what i meant... cpu's as in chips... shoulda been clearer, i guess. But what i meant was if you need more than 8 CPU's, you really don't want to use intel boxes...
Slashdot Mirror
I thought the whole beauty of open source was that the users/developers would make drivers as they needed them? The linuxppc guys can reverse engineer the mac motherboards, XFree86 people write their own video drivers when none are made available, but no one can take the time to figure out the modems that seem to appear in almost every shipping PC? What gives?
I thought the whole thing was that people liked the units but wanted to do stuff with them that didn't require their service? What's the difference between 3 months and 2 years if you have no intention of ever using it?
They really just should sell the machines at cost + 15% or 20% (regular profit) and then offer multihundred dollar rebates to people that sign on for 3 years of service, a la MSN, AOL, Compuserve, etc... If you want their service, you'd get the rebate. If you don't, then you buy the machine from them at a cost that makes everyone happy.
No one should be expected to lose money selling their goods, just because that's what the customers want.... Which seems to be the case of netppliance. They offered a good deal, didn't realize that people would find a way to use their device without their service and pretty much stood to get completely screwed, losing a few hundred bucks per machine with no hope of recouping it through ISP fees.
I just put together a system as well... It ended up being a Power Mac 8100/80, 72 megs of RAM, FWB Jackhammer SCSI card, 2x4GB Barracuda's, and a Radius VideoVision Studio card... All bought on eBay separately for about $700.00. The Mac has built in ethernet, so i can just ftp the files back and forth to my main (Win NT Workstation) machine... Overall, I'd say I'm quite happy with the set up and quality... I have a big investment in S-VHS and Hi8 gear that i didn't want to have to replace yet, and that saved it!
Actually, for video, RAM isn't really all that important... Drive bandwidth rules... as well as processing power. Why? Most any video of any length and quality is going to exeed the amount of RAM in the system, and in the end, worst case scenario (no compression) broadcast video is only going to need around 30 megabytes/second. That's an incredibly small piece of bandwidth, ram wise. Compressed video will be 5-7 megs a second (tv quality), which most hard drives can handle these days.
The processors important especially if you're working with compressed video... But while in the editting phase of a project, don't use any codec's like Sorenson or Cinepak... They just take TOO LONG to compress. Apple's generic "video" codec works fine, and doesn't hit the CPU nearly as hard.
The iMac DV's are nice little self contained systems... Their main drawbacks would be the lack of a second monitor and lack of slots, which limits them to only being able to use only FireWire camera's. iMovie may be lame, but you can buy Adobe Premiere and have a nice, cute little system...
Further up the scale, a Mac G4 couple with a Targa 1000 or 2000 video card would be a great choice for editting, because you've got all the expandability of the G4's, plus the ease of use and plug and play of the Mac... Video capture cards are very finnicky creatures, so it's nice to not have to worry too much about if device 1 will work with device 2 while on motherboard 3.
If you really want to have 3D effects, an NT based system would probably be the way to go, since most of the 3d developers target NT workstation at the low end, and either Solaris or Irix at the high end. Yes, there is some stuff available for the mac (Lightwave, though it lacks the 3rd party plug in support of NT, Infini-D, Electric Image, Strata Studio Pro all ship on the mac... missing from that list is 3D Studio, most notably).
Moving past the low end, you mightt also want to check out systems from Avid and Media 100... They sell turnkey solutions, based on the Mac OS, Windows NT, and Irix.
Without knowing budgets or goals, it's hard to recommend a video editting solution... One thing is, there just aren't any open source tools, or tools that run on the open source operating systems, that can stack up against the proprietary tools.
Lastly... If you end up on a Mac or Windows machine, you'll probably want Adobe Premiere and Adobe After Effects for your editting and compositing. Throw in Photoshop and Illustrator for titling, and you've hit almost $2000 on just software, so be warned it's not at all cheap!
I too would really like to start using OpenBSD... It just seem like an concept I'm ready for... But aside from installing the MacOS, Windows 95, 98, NT Workstation 4, NT Server 4, OpenLinux and Redhat Linux, I'm just not sure if i've the actual know-how to pull off an OpenBSD install... Can anyone share their experiences with me/us?
Is it a really technical process, or is it like the others where you just plop in a CD, choose some options, set some settings, sit back and babysit and viola! you've got an almost working machine? Somehow i doubt it's like that, though.
The same exact thing can be said of Linus and Linux. Just look at the hordes of "cultists" that conglomerate slashdot. Linus', ESR's and RMS' emails are almost scriptures of the open-source community.
The same could be said of the many BSDer's who rush to defend their OS of choice.
And of the many still amiga-philes.
Just because you don't like apple doesn't mean it's bad.
By mainstream, he meant consumer not server. Linux is not mainstream by that definition. Neither is FreeBSD. Neither is Windows NT. Neither is Netware.
Your number might be correct about market shares, but the last time i checked, 5% of the desktops is a WHOLE LOT MORE than 17% of servers.
But do SPARCs even exist that run at 500 MHz? And how much do they cost? And according to your calculations, which i recreated and got the same results with, you scaled the speed of the entire system, rather than just the speed of the processor. I don't think the system bus speed in a SPARC system scales linerally with the addition of faster processors, so anything that leaves the L1 cache would hit the memory bus and slow down a little bit.
It's hard to predict the speed difference of machines unless you take everything into account. A faster CPU is just one part of the equation.
And what if i don't WANT the source? I have to sit through an extra block of time downloading stuff i have absolutely zero interest in? And isn't that one of the issues people have with microsoft, shoving unwanted things down everyone's throats?
Leave the source where it is... if someone wants it they can get it... if they don't want it, they dont' have to have it.
You're forgetting that MS doesn't have monopoly positions in Server OS's or Embedded OS's...
In your regulation of Microsoft, what committee would determine which features, utilities, and API's would be appropriate for an operating system and which should go into applications? How quickly would that committee act?
It seems that Microsoft was found to have a monopoly on Windows. I think the best solution is a horizontal breakup, where 3 or 4 companies would be created, each with access to all of Microsofts patents, trademarks, and copyrights. Prevent employees from moving between companies for 10 years. Then set those companies lose on each other.
They'll all have the cash to compete. They can do whatever they want... If that means lowering their prices, maybe we'll see $15 windows licenses. Maybe one of the companies might decide to open source all of their non-patented softwares.
I don't think a DOJ solution should impose Linux as a competitor to windows and completely tilt the playing field for Linux by saying things to Microsoft like "No including browsers. No including development tools. No including servers, etc..." because every other company which has an OS does the same...
We just need 3 versions of windows to ensure that hey all have competition, keep prices in check, etc... There could be some clause in there saying that if in 5 or 10 years any of those companies had decimated the others, some other intervention could be sought.
I don't know. I'm not a lawyer. I just think the punishment should focus on consumers and windows, and forget about linux.
In case you hadn't, all the Linux darling have been taking quite a beating as of late:
ANDN's at 23% of it's high
CALD's at 71% of it's high
COBT's at 27% of it's high
CORL's at 23% of it's high
LNUX's at 19% of it's high
RHAT's at 28% of it's high
Where as with the dot com's,
AMZN's at 59% of it's high
LCOS's at 75% of it's high
YHOO's at 68% of it's high
In all those cases of me saying high, i meant 52 week highs...But those are the flagships of the Linux companies and the dotcoms. Hate to say it, but Linux has lost it's ring, it appears. The new trends are PDA's and wireless... Handspring's well poised to take advantage of that infatuation.
I still do wish that Handspring would manage to ship their products in quantity prior to their IPO... Palm had been shipping Pilots for years before 3Com spun them off. Handspring just develops a device and then says "time for the IPO!"
I wonder what the perfomance of a MAME running on Windows 98 VM running inside of Vmware running on Windows NT running inside of Vmware running on Redhat running inside of Connectixes Virtual PC running on a Mac G4/450 would be?
Aside from MAME, it'd be interesting to run some standard benchmark, or nearly standard, like Bytemark or SPEC (if someone with the $$$ for SPEC was interested) to see how much the CPU bogs down with each layer of emulation. Theoretically, if VMWare just passes x86 instructions to the processor, each additional instance should only suffer a small bit, rather than getting completely mangled.
Alternatively, one could run slackware inside of debian inside of openlinux inside of redhat inside of turbo linux... yikes!
I"d much rather see software versioned by release year, or something like that...
Netscape 2000.1, netscape 2000.2 etc...
it'd make figuring out what version of what is really the newst and all that.
It's all dependant on completely arbitrary factors. Like, who are the types of visitors? Are they loyal? Do they have money to spend? Is your site based on any of the now popular keywords? Is the domain of any value of itself?
Although it's kind of extreme, look at the market caps vs. actual revenues of Lycos vs. Yahoo... Yahoo has like 10x more visitors, 10x more market cap, but only 2x more earnings... So it's hard to quantify what makes a site valuable.
Exactly. ORBS seems a bit useless.
With ORBS, one could test a server to see if it will relay mail for you. If it does, you know, and orbs tells them. You can spam for up to 30 days before ORBS notifies the rest of the spammers, that that machine is now available. The spamming continues until the machine is blackholed.
Why publicize it? Why not just wait until the machine is actually used for spam, and then gothrough the steps of RBLing it? It's that whole innocent until proven guilty thing... They haven't done anything wrong, there's just potential for wrong to be done.
Sound familiar, Napster fans?
But NT has a microkernel! And Linux doesn't!... Your analogy has left me completely stumped. You should think of a better one.
That's like saying "xxx OS" is based on a Microkernel because you can launch and quit different programs without rebooting.
What is Darwin OS?
Darwin is a complete operating system based on the
foundation technologies in Mac OS X Server. It is an advanced BSD Unix system which offers advanced
networking, services such as the Apache web server,
and support for both Macintosh and Unix filesystems.
That's a direct snippet from: Apples Darwin page.
Forgive the copy & paste, but since there don't appear to be any links to this yet, here it is.
The reason that they and other censor ware vendors keep their lists secret is because that's their only asset. Access to the list is the only thing they're selling. If vendor A came along and released their list of 10,000,000 sites, in an open fashion, then vendor B could take those 10,000,000 sites, add 5,000,000 more, and have a 15,000,000 site list... C could come along and do nothing but still release a 15,000,000 name list.
I know, that's the same issue that's being faced with GPLed software, as companies like Mandrake and LinuxOne show, but I don't think that things such as support are nearly as much of an issue for Microsystems as compared to Redhat. Microsystems just needs to hold someones hands for 15 minutes, if that, where as Redhat has a much larger role to play with their customers.
While in the long term, security through obscurity doesn't work, in the short term, I'd think that it's a workable solution. I wish that the lot of ethical hackers would announce their discovery's of bugs and holes to the company's whose products have them and, depending on the complexitiy of the issue, hold off on making any announcements until a fix is at least available.
I mean, yes, the more skilled crackers will still have a chance to find out and exploit the holes, but it might at least prevent some of the script kiddie utilities from popping up, until at least a fix is available. If people get hit and there is no fix, it's basically terrorism (on a much larger scale). If people want to browse the web, they've got few real choices, so it's not like they can go "oh... IE sucks, i'll switch to netscape. Oh, netscape sucks too, i'll switch to Opera, oh... opera sucks, what do i do now?"
If fixes are available, and people still get "hit" by the script kiddies, at least they have a some one to point their finger at, with that person being themselves.
I think it all has to do with who installed it and who's using it. My friends have WinNT boxes that crash constantly, as well as Win98 boxes that seem to crash every other time they attempt to connect to the internet, where as my Win98 box has been running fine since 98 came out. The only hiccup it's ever had was when i upgraded the video card... Which unfortunately had the side effect of curbing my interest in Linux, being that the card doesn't seem to want to work with Redhat 6.0 in even VGA mode... command line is all i can get to now.
It all goes to say, it doesn't really matter so much about the operating system, it's who's using the machine.
A simple break up just won't do anything... There'll still be one company with a monopoly on the desktop OS market, another one with a monopoly on the productiviy suite market, and an internet company will a few billion in it's coffers. That's far from leveling the playing field.
So far as shares go... I'd almost think people would want MSFT broken up if the performance on ATT and the baby bells are any indication of what would happen to the Baby Bills. In the time frame since the breakup, IBM has increased something like 200% in value, while all the baby bells and ATT went up 800 or 1400% in the same period... I don't remember where i read it, though it was probably on news.com.
Actually, no. Mine was with IE :P
As far as I've always known... Tour's only really make money for the REALLY SMALL and REALLY BIG BANDS... all the medium sized bands tour mainly to promote their record sales. And record stores take a much larger chunk than just the $1. It's more around $3-4 on up.
But you're into music right? Read an essay by steve albini (a la big black, shellac, rapeman fame... along with engineering a slew of albums). It pretty much shows artists getting hosed already... But taking away their CD royalties pretty much destroys their means of recording as a means to support their careers.
That's what i meant... cpu's as in chips... shoulda been clearer, i guess. But what i meant was if you need more than 8 CPU's, you really don't want to use intel boxes...