I'm not entirely sure about 10.2.x on down(I'm pretty sure they're stuck at 8), but 10.3(aka Panther) finally takes care of the issue officially. With Panther, Apple's finally gone to the *nix standard of shadow hashes, so you can have whatever long password you want.
10.2 still uses the standard crypt() algorithm without md5 support, so it's still limited to 8 characters. Glad to hear they're finally changing this.
First let me say I know nothing about Mac OSX, so YMMV. From what I understand 8 is the default for most *NIXs and to change it is a compile time option for the kernel which I'm supposing you can't do. If you have an/etc/login.defs, the PASS_MAX_LEN field only specifies how many characters are used when crypt() hashes the password for the md5 output.
Wow. Um, hate to say it, but you're completely and totally wrong. For starters, the standard crypt() algorithm, without md5 support, only works on passwords with no more than 8 characters; anything after 8 is ignored. And second, it has nothing whatsoever to do with the kernel; authentication happens entirely in userspace.
As for the Mac OS X-specific bit, the problem is that Mac OS X doesn't support md5.
Just use Firebird. I guarantee a pleasant experience!
You do realize that Firebird uses the same kind of plugins, right? So when Eolas is done with Microsoft, they could potentially come after The Mozilla Foundation?
No-floppy computers? Sorry, he did not. IBM sold crippled floppy-less PC's at the beginning. This is not a good idea: it is bad one that few copy. It is not a good thing to sell something without a useful feature. That's why most PC's have floppies (along with the DVD or CDRW burner).
Computers without floppy drives are not crippled, unless you need to use a floppy to interact with an old computer that can't use something better for some reason. For example, it might be a convenient way to transport small files (word processing documents, for example) to/from a computer that doesn't have Internet access, and I'll need to use floppies the next time I upgrade the OS on my 486. If you need these features, you can buy a floppy drive. But there's no reason for every computer in my house to have one.
USB? He did not bring us this. It would all over the place before the iMac.
USB was on motherboards; there were almost no USB peripherals. The first USB printer to hit the market was made by Epson, and it had translucent blue plastic to match the iMac.
These unreadable icons will die a quick death. Thankfully, few are following the lead. Look for Apple to ditch it when the paradigm of "make icons readable" returns.
I'm with you on that.
We could also mention an old Job's favorite: "eject the disk with a bent paperclip in a hole. It is so much better than an eject button". No one copied this.
Sun did, I believe.
Eventually, Jobs had to follow the leaders: there are now eject buttons on Macs.
There are now CD eject buttons on Mac keyboards. Previously, there had always been eject buttons on Mac CD-ROM drives, as well as other Mac removable media drives such as Zip drives, external floppy drives, etc. Newer Macs do not have accessible eject buttons on them, relying on the keyboard button instead. What does this have to do with your point?
It should be noted that nearly all removable media drives have a paperclip hole, except PC floppy drives which have a manual eject button. Try to find some that don't.
Actually, the one innovation you did name that finds favor outside of Apple's niche that Jobs was involved with was Firewire, which he did.... alongside Sony.
Well, I remember a time when nobody complained about editors not reading the articles, is that good enough?
I also remember when CowboyNeal had never been the default poll option, First Post started at comment ID#1, and yes, I even remember when www.slashdot.org didn't resolve. My how times have changed.
Not only that, but I happen to like having more than one mouse button. Fight the mono-button tyranny!!!
Give it up already. Buy a Mac, sell the mouse on eBay, and buy a new one for $20. Logitech makes very nice ones. Microsoft's have improved since their optical mouse was first introduced. Just to be different, I bought my last one from IBM. Seriously, quit whining. If you can afford a Mac, you can afford a third-party mouse to go with it.
Not only is FireWire 400 faster than USB 2, but FireWire 800 (IEEE1394b) is even faster than that. Built into new PowerMacs and PowerBooks (except the 12"PB), and available hereherehere and here (quick Google results).
FireWire doesn't divide the bandwidth among each device regardless of what they're using; I believe it's packet-based and each device uses only as much bandwidth as it needs.
It would be too simplistic to say that their profits would go down due to fewer people buying their hardware. They could, for example, realize greater profits from disproportionately greater software sales.
Releasing Mac OS X for x86 might be the best thing that could ever happen to Linux and the worst thing that could ever happen to Microsoft, but it could easily kill Apple.
Getting people to switch to Mac OS X on x86 would be like getting people to switch to Linux on x86, except that Linux is free and runs more applications. As soon as people start to realize that a non-Microsoft alternative exists, the majority will switch to the cheapest alternative they can find, which will be something free (as in beer).
Apple could make it work, but to do so, they'd need to completely change their entire business model. Consider this plan:
1) Release Mac OS X for x86 and PPC for free.
2) port Cocoa (and possibly Carbon) to Linux, FreeBSD and Win32 as well as Mac OS X for x86 and PPC. Charge developers licensing fees to bundle it.
3) Sell Xcode to developers. Make sure by default it builds dual-platform binaries, so compiled apps will run on both x86 and PPC natively, on any OS Cocoa has been ported to.
4) Port all the iApps and sell them. Become primarily a software company, which also sells hardware.
At this point Cocoa becomes the middleware Microsoft was so afraid of a decade ago. Cocoa applications can run on any operating system and architecture that Cocoa supports, so operating systems have to compete on technical merits rather than on application support. Likewise processor architectures - if the price/performance of IBM's PowerPCs is better than Intel's Pentiums, then people switch away from Intel.
It's a risky move, and it might not work. If it does, it could secure Apple's position as the computer industry leader. If it doesn't, it could completely kill any hope of ever making money again.
In Safari, select some text, then go up to the Safari menu and go to Services. Try Make New Sticky Note, or Mail/Send Selection, or Speech/Start Speaking Text.
This feature hasn't gotten much attention, so in addition to not being marketed it also isn't polished, but there's a lot of potential there.
Don't forget that HyperCard was definately one of the first visual UI editors - if not the first. When I used Visual Basic for the first time I couldn't believe how similar to HyperCard it was, and probably still is.
VB struck me as a weird cross between BASIC, HyperCard and JavaScript.
Verisign still gets your money - GoDaddy has to pay them $6/yr per domain. That's the wholesale fee Verisign charges to all registrars for.com/.net domains.
I completely agree with most of that. However, what about when the spammer lies to the client and tells them what they're doing is not spamming? Is the client still at fault? Obviously not; the spammer is - but going after the spammers hasn't really been effective so far.
Nice, but it doesn't need to be lethal - just make it bad enough that it makes the news. TRICKY PART: get the media to point the finger at spam in general, not just your actions.
How many of your complaints are sent via SpamCop? Those were the majority when I was doing abuse. AOL and RoadRunner were more threatening, of course.;-)
I just got a legitimate email returned because spamcop claims that the smtp server of the webhosting provider has an abnormal rate of spam.
Your e-mail was returned because whoever runs the mail server you were trying to deliver the message to has chosen to bounce mail from any IP in SpamCop's blacklist, which SpamCop has always recommended against. Complain to the people who made that decision, not SpamCop.
And, the reason the IP is listed in SpamCop's blacklist is probably because the server you're relaying your mail through has also been relaying spam, and people have complained about it (using SpamCop's reporting service). Go here to find out exactly why an IP is listed, along with sample e-mails that users have reported as spam and some statistics about how much spam has been reported from that IP.
The worse thing about spam is that filtering systems create false positives...
SpamCop says this is why their blacklist should not be used to block mail. Their list is entirely automated; it's based on reports from users, and SpamCop does not verify it. Read more on SpamCop's site about exactly how it works.
My provider requires authentication but everyone knows that you can create spam using a IP address from a well behaved smtp server.
SpamCop is really very good about identifying where a message actually came from, not just where it's been relayed through - unless there's something suspicious-looking about the server it's been relayed through (such as, for example, the hostname the server identifies itself as [the Dj line in sendmail.cf] doesn't resolve to the server's IP).
Spam exists because it works; enough people buy products that are advertised through spam that the increased sales more than make up for the cost of spamming.
Companies choose Microsoft solutions because Microsoft provides the most flexible, stable and secure systems, with lower TCO than the competition.
I believe both of these statements are false, but are believed to be true by people making the decisions. Why? Because spammers and (to a much lesser extent) Microsoft salespeople are dirty rotten lying scumbags out to make a buck by cheating whoever they can. On top of that, spammers also sell their service by claiming what they're selling is not spam - it's direct marketing to a targeted opt-in list of interested consumers over the Internet. We all know in reality it's completely untargetted and their definition of "opt-in" includes allowing your e-mail address to appear unobfuscated on any web page, using it to register a domain name or post to a newsgroup, or simply choosing an e-mail address that could be guessed at random. We know that, just like we know Windows almost never has a lower TCO than anything. But the people paying the money don't, because they simply don't know better.
Yeah, like KHTML in Safari, or the toolchain to build Darwin/OSX (standard gcc....).
KHTML isn't GPL, it's LGPL. Gcc would be a problem though; I hadn't considered that - still, IBM is working on making its own compiler gcc-compatible; I expect Apple to switch when this becomes a viable option.
Slashdot Mirror
I'm not entirely sure about 10.2.x on down(I'm pretty sure they're stuck at 8), but 10.3(aka Panther) finally takes care of the issue officially. With Panther, Apple's finally gone to the *nix standard of shadow hashes, so you can have whatever long password you want.
10.2 still uses the standard crypt() algorithm without md5 support, so it's still limited to 8 characters. Glad to hear they're finally changing this.
First let me say I know nothing about Mac OSX, so YMMV. From what I understand 8 is the default for most *NIXs and to change it is a compile time option for the kernel which I'm supposing you can't do. If you have an /etc/login.defs, the PASS_MAX_LEN field only specifies how many characters are used when crypt() hashes the password for the md5 output.
Wow. Um, hate to say it, but you're completely and totally wrong. For starters, the standard crypt() algorithm, without md5 support, only works on passwords with no more than 8 characters; anything after 8 is ignored. And second, it has nothing whatsoever to do with the kernel; authentication happens entirely in userspace.
As for the Mac OS X-specific bit, the problem is that Mac OS X doesn't support md5.
Just use Firebird. I guarantee a pleasant experience!
You do realize that Firebird uses the same kind of plugins, right? So when Eolas is done with Microsoft, they could potentially come after The Mozilla Foundation?
Ah... in NeXTStep it was called the Pasteboard?
No-floppy computers? Sorry, he did not. IBM sold crippled floppy-less PC's at the beginning. This is not a good idea: it is bad one that few copy. It is not a good thing to sell something without a useful feature. That's why most PC's have floppies (along with the DVD or CDRW burner).
Computers without floppy drives are not crippled, unless you need to use a floppy to interact with an old computer that can't use something better for some reason. For example, it might be a convenient way to transport small files (word processing documents, for example) to/from a computer that doesn't have Internet access, and I'll need to use floppies the next time I upgrade the OS on my 486. If you need these features, you can buy a floppy drive. But there's no reason for every computer in my house to have one.
USB? He did not bring us this. It would all over the place before the iMac.
USB was on motherboards; there were almost no USB peripherals. The first USB printer to hit the market was made by Epson, and it had translucent blue plastic to match the iMac.
These unreadable icons will die a quick death. Thankfully, few are following the lead. Look for Apple to ditch it when the paradigm of "make icons readable" returns.
I'm with you on that.
We could also mention an old Job's favorite: "eject the disk with a bent paperclip in a hole. It is so much better than an eject button". No one copied this.
Sun did, I believe.
Eventually, Jobs had to follow the leaders: there are now eject buttons on Macs.
There are now CD eject buttons on Mac keyboards. Previously, there had always been eject buttons on Mac CD-ROM drives, as well as other Mac removable media drives such as Zip drives, external floppy drives, etc. Newer Macs do not have accessible eject buttons on them, relying on the keyboard button instead. What does this have to do with your point?
It should be noted that nearly all removable media drives have a paperclip hole, except PC floppy drives which have a manual eject button. Try to find some that don't.
Actually, the one innovation you did name that finds favor outside of Apple's niche that Jobs was involved with was Firewire, which he did.... alongside Sony.
Then why didn't Sony win the Emmy?
Alyson Hannigan?
Well, I remember a time when nobody complained about editors not reading the articles, is that good enough?
I also remember when CowboyNeal had never been the default poll option, First Post started at comment ID#1, and yes, I even remember when www.slashdot.org didn't resolve. My how times have changed.
Unless, of course, they're solid state. ...which the iPod isn't.
Not only that, but I happen to like having more than one mouse button. Fight the mono-button tyranny!!!
Give it up already. Buy a Mac, sell the mouse on eBay, and buy a new one for $20. Logitech makes very nice ones. Microsoft's have improved since their optical mouse was first introduced. Just to be different, I bought my last one from IBM. Seriously, quit whining. If you can afford a Mac, you can afford a third-party mouse to go with it.
Exactly The Wrong Kind Of Pirate
Not only is FireWire 400 faster than USB 2, but FireWire 800 (IEEE1394b) is even faster than that. Built into new PowerMacs and PowerBooks (except the 12"PB), and available here here here and here (quick Google results).
FireWire doesn't divide the bandwidth among each device regardless of what they're using; I believe it's packet-based and each device uses only as much bandwidth as it needs.
It would be too simplistic to say that their profits would go down due to fewer people buying their hardware. They could, for example, realize greater profits from disproportionately greater software sales.
Releasing Mac OS X for x86 might be the best thing that could ever happen to Linux and the worst thing that could ever happen to Microsoft, but it could easily kill Apple.
Getting people to switch to Mac OS X on x86 would be like getting people to switch to Linux on x86, except that Linux is free and runs more applications. As soon as people start to realize that a non-Microsoft alternative exists, the majority will switch to the cheapest alternative they can find, which will be something free (as in beer).
Apple could make it work, but to do so, they'd need to completely change their entire business model. Consider this plan:
1) Release Mac OS X for x86 and PPC for free.
2) port Cocoa (and possibly Carbon) to Linux, FreeBSD and Win32 as well as Mac OS X for x86 and PPC. Charge developers licensing fees to bundle it.
3) Sell Xcode to developers. Make sure by default it builds dual-platform binaries, so compiled apps will run on both x86 and PPC natively, on any OS Cocoa has been ported to.
4) Port all the iApps and sell them. Become primarily a software company, which also sells hardware.
At this point Cocoa becomes the middleware Microsoft was so afraid of a decade ago. Cocoa applications can run on any operating system and architecture that Cocoa supports, so operating systems have to compete on technical merits rather than on application support. Likewise processor architectures - if the price/performance of IBM's PowerPCs is better than Intel's Pentiums, then people switch away from Intel.
It's a risky move, and it might not work. If it does, it could secure Apple's position as the computer industry leader. If it doesn't, it could completely kill any hope of ever making money again.
Which services are you referring to?
Yep, see? Failure to promote it.
In Safari, select some text, then go up to the Safari menu and go to Services. Try Make New Sticky Note, or Mail/Send Selection, or Speech/Start Speaking Text.
This feature hasn't gotten much attention, so in addition to not being marketed it also isn't polished, but there's a lot of potential there.
Cannons and Castles is my port of an Apple II game. Here's an old screen shot.
Don't forget that HyperCard was definately one of the first visual UI editors - if not the first. When I used Visual Basic for the first time I couldn't believe how similar to HyperCard it was, and probably still is.
VB struck me as a weird cross between BASIC, HyperCard and JavaScript.
Verisign still gets your money - GoDaddy has to pay them $6/yr per domain. That's the wholesale fee Verisign charges to all registrars for .com/.net domains.
$6 of that goes to Verisign; Hostway only makes $0.95.
I completely agree with most of that. However, what about when the spammer lies to the client and tells them what they're doing is not spamming? Is the client still at fault? Obviously not; the spammer is - but going after the spammers hasn't really been effective so far.
Nice, but it doesn't need to be lethal - just make it bad enough that it makes the news. TRICKY PART: get the media to point the finger at spam in general, not just your actions.
How many of your complaints are sent via SpamCop? Those were the majority when I was doing abuse. AOL and RoadRunner were more threatening, of course. ;-)
I just got a legitimate email returned because spamcop claims that the smtp server of the webhosting provider has an abnormal rate of spam.
Your e-mail was returned because whoever runs the mail server you were trying to deliver the message to has chosen to bounce mail from any IP in SpamCop's blacklist, which SpamCop has always recommended against. Complain to the people who made that decision, not SpamCop.
And, the reason the IP is listed in SpamCop's blacklist is probably because the server you're relaying your mail through has also been relaying spam, and people have complained about it (using SpamCop's reporting service). Go here to find out exactly why an IP is listed, along with sample e-mails that users have reported as spam and some statistics about how much spam has been reported from that IP.
The worse thing about spam is that filtering systems create false positives...
SpamCop says this is why their blacklist should not be used to block mail. Their list is entirely automated; it's based on reports from users, and SpamCop does not verify it. Read more on SpamCop's site about exactly how it works.
My provider requires authentication but everyone knows that you can create spam using a IP address from a well behaved smtp server.
SpamCop is really very good about identifying where a message actually came from, not just where it's been relayed through - unless there's something suspicious-looking about the server it's been relayed through (such as, for example, the hostname the server identifies itself as [the Dj line in sendmail.cf] doesn't resolve to the server's IP).
Spam exists because it works; enough people buy products that are advertised through spam that the increased sales more than make up for the cost of spamming.
Companies choose Microsoft solutions because Microsoft provides the most flexible, stable and secure systems, with lower TCO than the competition.
I believe both of these statements are false, but are believed to be true by people making the decisions. Why? Because spammers and (to a much lesser extent) Microsoft salespeople are dirty rotten lying scumbags out to make a buck by cheating whoever they can. On top of that, spammers also sell their service by claiming what they're selling is not spam - it's direct marketing to a targeted opt-in list of interested consumers over the Internet. We all know in reality it's completely untargetted and their definition of "opt-in" includes allowing your e-mail address to appear unobfuscated on any web page, using it to register a domain name or post to a newsgroup, or simply choosing an e-mail address that could be guessed at random. We know that, just like we know Windows almost never has a lower TCO than anything. But the people paying the money don't, because they simply don't know better.
Perhaps they actually mean it takes 5 seconds for a stream to start playing, if you're on a really fast connection.
Yeah, like KHTML in Safari, or the toolchain to build Darwin/OSX (standard gcc....).
KHTML isn't GPL, it's LGPL. Gcc would be a problem though; I hadn't considered that - still, IBM is working on making its own compiler gcc-compatible; I expect Apple to switch when this becomes a viable option.