You may know that a 500 MHz G4 is faster than a 1 GHz PIII but the vast majority of computer buyers don't have a clue. Until that changes, Apple has to want (and Motorola had damn well better start supplying) faster chips, because it is going to continue to be difficult to sell Apples otherwise. ~tieguy
You forget that Darwin/OS X are built on a BSD core. All they have to do is use the BSD drivers- which are free, well tested, and available for a huge number of platforms. Now, graphics drivers will be a different issue (since OS X doesn't use Xfree) but motherboards, HDs, etc., so on, so forth, have all already been done for Apple. That's one of the beauties (for them) of basing things on a BSD platform. Don't be surprised if that has been in the works all along. ~luge P.S. Graphics drivers won't be easy to write, but don't be surprised if the hypercompetitive graphics card companies wouldn't jump at the chance to at least help Apple- in such a tight market, an additional 4-5% in market share would mean a huge jump for many of those companies.
Quartz (OS X's brand new imaging subsystem) would probably suffer a more noticable performance drop, given the extensive use of transparency.
Right on the money, I think. If you go back and look at the arstechnica stories on OS X, one of them mentions that OS performance took a huge hit when going from a G4 to a G3, even though the machines were comparably equipped (the G3 may have even been faster- don't recall.) Presumably (and ars may have mentioned this) it was because of Altivec and Quartz. ~luge
It's not like it is surprising or anything. We've known for quite some time that they intend to make boxes that do nothing except run AOL- this is just the logical predecessor to that box and it's cousins. AOL gets a little practice in distribution and organization for very little skin off their back. Besides, this isn't really a big threat. It's not like AOL can control HW makers in the same way MS does, especially once they start competing. What should be scary is their newfound control (via TimeWarner) of content. I can always get a box from VA or office depot or any of about a zillion online sites, and I'll be able to get them without AOL keyboards. I won't be able to get whole bunches of movies that the Turner Networks own with that box, though. I may not be able to read Time. And don't forget those NBA contracts... think espn.com might get shutout of the loop anytime soon? That is the kind of domination that has been staring us in the face for months, and that is what we ought to be scared of. ~luge
Actually, IIRC, at least one of Cyperpatrol's competitors (the one whose content is controlled by the Christian Coalition) really did put the ACLU on the banned list. There are any number of people who sincerely believe that groups like the ACLU who fight for civil rights for all groups are evil. Scary, really... ~luge
The problem is not that it isn't a law, it's that fast new processors get dropped on Intel/AMD every 18 months. They get dropped on me something like every four years.... ~luge
I fear the next perl upgrade... I mean, I don't think it'll be as bad as the last (hopefully, they'll have learned something from the last one) but one can never know. Damn... ~luge
Umm... I'll take beta open-source security over closed-source x.0 security any day, thanks. We all know security through binary obscurity doesn't work... ~luge
Umm... last I checked they didn't sell boxes with any OS other than RH, so while the BSD and Debian boxes are probably VA hardware, they aren't boxes you can buy from VA, which is a big and important difference. ~luge
As someone who downloads a nightly every day, I have to agree, but maybe not in the way you think- even very good coders have had problems sinking their teeth into the huge Mozilla codebase. So, I'm not sure that it'll ever get a huge developer base in that sense. However, it will very quickly have the most active and useful bug-reporting and feature-requesting community of any product (sans, maybe, Linux). It'll also be useful for a huge number of other projects- look for people to follow in the footsteps of Zope and become "developers," not in the normal sense that a developer works with the original codebase, but rather as people who extend the project into new and different products without really touching the original code at all. my two cents- ~luge(posted with mozilla build 2000030709)
I think the distinction that is important to remember is that if you are using Mozilla, you'll be getting developer-type support- either very technical or very "RTFM" style. If you choose to go with Netscape, you'll get more professional and more user-friendly support. Everything will be one nice, happy package, with a label that the average user can "trust." Mozilla will remain more cutting edge (I hope they continue doing nightlies) and (hopefully) continue to be the source and testbed for new development and new ideas, which is something that the "standard" Netscape will never become. Think of it as the stable and unstable versions, if you will. ~luge(posted from mozilla build 2000030709)
They'll have them soon. Many of the helix guys use Debian (some are even Debian developers), so it is a priority for them. As you may have noticed, though, making an rpm is about a zillion times easier than setting up a.deb. They are busy bees over there, so this'll get done reasonably soon, just not immediately. ~luge
If you want to see more details on this issue, take a look at the primary SMP bug here. To comment on some of the other notes on this SMP thread: 1) Yes, it is multi-threaded. The problem is that certain key functions arae not thread safe. 2) The problem does show up on non-SMP boxes, but is rare- it is greatly exagerrated on SMP, both Linux and NT. 3) There is a lot of work underway on this problem, since several of the developers use SMP boxes at home. Unfortunately, while the solution is reasonably straightforward, it will require a lot of work. You'll note that the bug is marked beta1- which means it is a priority. We'll see... ~luge
Just a few days ago they removed a bug in the timing code which had been quadrupling win32 rendering times. Oddly enough, the bug wasn't in Mozilla itself but rather in some debugging/timing code- once that was disabled, the slowdowns went away. ~luge
You are kidding, right? Any way to document that? I'd love to add that to the arsenal of crud against SurfWatch (as, I'm sure, would the guys in Holland...) ~luge
Whenever Debian freezes it will ship with Mozilla (not sure which build, though.) Advancement of Mozilla is pretty crucial for Debian, since (unlike other distributions that would ship Mozilla as a supplement to Netscape) Debian doesn't distribute Netscape with the core of the distro. Until there is a usable Mozilla, Debian will continue to ship without a "serious" browser. Grain of salt: I'm posting this from yesterday's build, so I (personally) consider Mozilla pretty damn fine stuff. But it's just not quite ready for mainstream acceptance (which is my Debian isn't in great shape, web-wise.) ~luge
BigBaldGuy- That's great to hear. I knew that part of the problem was proprietary (read: unfreeable) stuff, and I'm sure your intentions are good. My one concern is whether or not I'll be able to use it with nightlies (since I use those and not the M builds.) Any idea if that'll be the case? ~luge
One important note: the crypto in M14 will not be Open Source. Rather, M14 will incorporate hooks which will be usable with a binary-only crypto module from iPlanet (the offspring of the Netscape-Mozilla alliance.) For more on the situation, read this mozillazine post.
I think if you search the mozillazine archives, you'll find a suggestion on how to do it. You'll need the newer nightly binaries, though- M13 won't cut it. ~luge
Besides the frequent complaints about X and the GUI situation on Linux, the next most frequent usability complaint is the lack of documentation, and/or it's poor quality. Do you have any thoughts or comments on the role of documentation in a complete system? Should there even need to be docs for a well-designed GUI? ~luge
But you don't have an 80% chance of a failure a day. If that were true, 10 machines with a ten percent failure rate would guarantee a failure every day, and that just doesn't make sense. My challenge to someone whose stats are fresh still stands... ~luge
Maybe it's just that it is early in the morning (sort of) but your math seems a little off... it would seem that if each box has a 10% chance of going down on a given day, and you have eight of them, your odds of sucess are closer to 20% (100%-(8*10%)) than they are to 99.999999%. Obviously, both of these numbers are wrong... anyone who has taken stats more recently than me is welcome to correct them... ~luge
Slashdot Mirror
You may know that a 500 MHz G4 is faster than a 1 GHz PIII but the vast majority of computer buyers don't have a clue. Until that changes, Apple has to want (and Motorola had damn well better start supplying) faster chips, because it is going to continue to be difficult to sell Apples otherwise.
~tieguy
You forget that Darwin/OS X are built on a BSD core. All they have to do is use the BSD drivers- which are free, well tested, and available for a huge number of platforms. Now, graphics drivers will be a different issue (since OS X doesn't use Xfree) but motherboards, HDs, etc., so on, so forth, have all already been done for Apple. That's one of the beauties (for them) of basing things on a BSD platform. Don't be surprised if that has been in the works all along.
~luge
P.S. Graphics drivers won't be easy to write, but don't be surprised if the hypercompetitive graphics card companies wouldn't jump at the chance to at least help Apple- in such a tight market, an additional 4-5% in market share would mean a huge jump for many of those companies.
Quartz (OS X's brand new imaging subsystem) would probably suffer a more noticable performance drop, given the extensive use of transparency.
Right on the money, I think. If you go back and look at the arstechnica stories on OS X, one of them mentions that OS performance took a huge hit when going from a G4 to a G3, even though the machines were comparably equipped (the G3 may have even been faster- don't recall.) Presumably (and ars may have mentioned this) it was because of Altivec and Quartz.
~luge
It's not like it is surprising or anything. We've known for quite some time that they intend to make boxes that do nothing except run AOL- this is just the logical predecessor to that box and it's cousins. AOL gets a little practice in distribution and organization for very little skin off their back.
Besides, this isn't really a big threat. It's not like AOL can control HW makers in the same way MS does, especially once they start competing. What should be scary is their newfound control (via TimeWarner) of content. I can always get a box from VA or office depot or any of about a zillion online sites, and I'll be able to get them without AOL keyboards. I won't be able to get whole bunches of movies that the Turner Networks own with that box, though. I may not be able to read Time. And don't forget those NBA contracts... think espn.com might get shutout of the loop anytime soon? That is the kind of domination that has been staring us in the face for months, and that is what we ought to be scared of.
~luge
It's CYBERSitter. It blocks, among other things, the ACLU, the National Organization for Women, and Mother Jones. Great stuff.
~luge
Actually, IIRC, at least one of Cyperpatrol's competitors (the one whose content is controlled by the Christian Coalition) really did put the ACLU on the banned list. There are any number of people who sincerely believe that groups like the ACLU who fight for civil rights for all groups are evil.
Scary, really...
~luge
The problem is not that it isn't a law, it's that fast new processors get dropped on Intel/AMD every 18 months. They get dropped on me something like every four years....
~luge
I fear the next perl upgrade... I mean, I don't think it'll be as bad as the last (hopefully, they'll have learned something from the last one) but one can never know. Damn...
~luge
Umm... I'll take beta open-source security over closed-source x.0 security any day, thanks. We all know security through binary obscurity doesn't work...
~luge
Umm... last I checked they didn't sell boxes with any OS other than RH, so while the BSD and Debian boxes are probably VA hardware, they aren't boxes you can buy from VA, which is a big and important difference.
~luge
Yes, it's a troll, but this is still just about the damn funniest thing I've ever seen. How can you possibly moderate it down? For shame...
~luge
As someone who downloads a nightly every day, I have to agree, but maybe not in the way you think- even very good coders have had problems sinking their teeth into the huge Mozilla codebase. So, I'm not sure that it'll ever get a huge developer base in that sense. However, it will very quickly have the most active and useful bug-reporting and feature-requesting community of any product (sans, maybe, Linux). It'll also be useful for a huge number of other projects- look for people to follow in the footsteps of Zope and become "developers," not in the normal sense that a developer works with the original codebase, but rather as people who extend the project into new and different products without really touching the original code at all.
my two cents-
~luge(posted with mozilla build 2000030709)
I think the distinction that is important to remember is that if you are using Mozilla, you'll be getting developer-type support- either very technical or very "RTFM" style. If you choose to go with Netscape, you'll get more professional and more user-friendly support. Everything will be one nice, happy package, with a label that the average user can "trust." Mozilla will remain more cutting edge (I hope they continue doing nightlies) and (hopefully) continue to be the source and testbed for new development and new ideas, which is something that the "standard" Netscape will never become. Think of it as the stable and unstable versions, if you will.
~luge(posted from mozilla build 2000030709)
They'll have them soon. Many of the helix guys use Debian (some are even Debian developers), so it is a priority for them. As you may have noticed, though, making an rpm is about a zillion times easier than setting up a .deb. They are busy bees over there, so this'll get done reasonably soon, just not immediately.
~luge
If you want to see more details on this issue, take a look at the primary SMP bug here. To comment on some of the other notes on this SMP thread: 1) Yes, it is multi-threaded. The problem is that certain key functions arae not thread safe. 2) The problem does show up on non-SMP boxes, but is rare- it is greatly exagerrated on SMP, both Linux and NT. 3) There is a lot of work underway on this problem, since several of the developers use SMP boxes at home. Unfortunately, while the solution is reasonably straightforward, it will require a lot of work. You'll note that the bug is marked beta1- which means it is a priority. We'll see...
~luge
Just a few days ago they removed a bug in the timing code which had been quadrupling win32 rendering times. Oddly enough, the bug wasn't in Mozilla itself but rather in some debugging/timing code- once that was disabled, the slowdowns went away.
~luge
Well, sure- it has been in 2.2 unstable since times unknown (M6, IIRC.) But I'm pretty sure it wasn't in slink.
~luge (proud woody user)
You are kidding, right? Any way to document that? I'd love to add that to the arsenal of crud against SurfWatch (as, I'm sure, would the guys in Holland...)
~luge
Whenever Debian freezes it will ship with Mozilla (not sure which build, though.) Advancement of Mozilla is pretty crucial for Debian, since (unlike other distributions that would ship Mozilla as a supplement to Netscape) Debian doesn't distribute Netscape with the core of the distro. Until there is a usable Mozilla, Debian will continue to ship without a "serious" browser.
Grain of salt: I'm posting this from yesterday's build, so I (personally) consider Mozilla pretty damn fine stuff. But it's just not quite ready for mainstream acceptance (which is my Debian isn't in great shape, web-wise.)
~luge
BigBaldGuy-
That's great to hear. I knew that part of the problem was proprietary (read: unfreeable) stuff, and I'm sure your intentions are good. My one concern is whether or not I'll be able to use it with nightlies (since I use those and not the M builds.) Any idea if that'll be the case?
~luge
One important note: the crypto in M14 will not be Open Source. Rather, M14 will incorporate hooks which will be usable with a binary-only crypto module from iPlanet (the offspring of the Netscape-Mozilla alliance.) For more on the situation, read this mozillazine post.
I think if you search the mozillazine archives, you'll find a suggestion on how to do it. You'll need the newer nightly binaries, though- M13 won't cut it.
~luge
Besides the frequent complaints about X and the GUI situation on Linux, the next most frequent usability complaint is the lack of documentation, and/or it's poor quality. Do you have any thoughts or comments on the role of documentation in a complete system? Should there even need to be docs for a well-designed GUI?
~luge
But you don't have an 80% chance of a failure a day. If that were true, 10 machines with a ten percent failure rate would guarantee a failure every day, and that just doesn't make sense. My challenge to someone whose stats are fresh still stands...
~luge
Maybe it's just that it is early in the morning (sort of) but your math seems a little off... it would seem that if each box has a 10% chance of going down on a given day, and you have eight of them, your odds of sucess are closer to 20% (100%-(8*10%)) than they are to 99.999999%. Obviously, both of these numbers are wrong... anyone who has taken stats more recently than me is welcome to correct them...
~luge