No, Apple does not say Intel-based Macs will have a BIOS. It just so happens that these devkits' machines have a Phoenix BIOS. The pdf that everyone's been referring to only says that there is no Open Firmware, there is no mention of BIOS.
You're assuming too many things will stay the way they are from the devkits' architecture.
I guess I "put up" with it 'cuz extra shit has never happened to me while I'm on a laptop with a trackpad. Maybe your fingers are too big, or you're used to having your thumbs rest below the spacebar as you type. I just looked at my hands while on a laptop and my thumbs don't touch the pad--they just hover above the spacebar.
It took me about 2 minutes to learn the finesse of doing 1-1/2 taps that are required for click lock for dragging--tap once then doing 1/2 a tap by pressing your finger on the pad and while not releasing move your finger.
Touchpads are much better because with the correct driver you can do everything with a touchpad that you can with a mouse/trackball including click, drag, and click/drag lock without using even the buttons. A truly solid state device with no moving parts. Apple does one better by allowing you to scroll using 2 fingers on the pad without the use of dedicated regions.
I have never liked the trackpoints, especially with my tendonitis. Doing click drags just feels unnatural.
It's really no use these days as most people use the Internet for a lot of stuff. When power goes out, your connection goes with it unless you have a UPS.
Apple's behaivor has zero to do with the point I was trying to make.
On the contrary, your use of "does the bare minimum," and "inadequate documentation" covertly implies that Apple's behavior is the problem. After observing this issue so far, I think Apple has done a lot for KHTML and the OSS community. The KHTML team finally took the code Apple gave them to heart and integrated a good portion back.
No, it doesn't. These "super" CDs use a Microsoft utility to bind common files in all forms of Windows XP (for instance, there are ones for Office, Windows 2000, 2003) to one physical location on the CD instead of different and separate locations. Saves a lot of space as there are a lot of common files.
Oh goodie, one of these "nya nya nya, mine's bigger than yours" questions.
Hey, I have 1 computer for every room in my house, including the attic. Surf anywhere on 1GBe HW for everything. Wireless everywhere. Servers everywhere. Mine's bigger than yours. Well, I'm kidding of course.
Anything more than 1 computer/person and it's a luxury.
Not quite. Congress, the legislative branch makes the laws; police/armed forces, the executive branch enforces the laws; courts, the judicial branch interprets the laws.
I think it goes like this: Modern hard drives require a thin film of air for the heads to ride on. Breather holes on the tops (usually) of HDs let the air in--they get filtered.
Yup, I agree with you. What the original submitter doesn't realize is if you get a good mobo with onboard NIC (GBe or faster) that doesn't go through the PCI bus, he will get much better throughput. If all the cards are used simultaneously, there will be unwanted PCI bus contention. Probably not an issue with PCIe, but it is probably wise to shop for a good "all-in-one." Of course, there are cheaper mobos where onboard NICs still go through the PCI bus--I try to avoid those.
Is this an old Dell or one from around 3 years ago? I have found that recent Dells have switched to standard PC power supplies (old ones had different pin layout) and have drive rails for all available bays.
What bites is some of their low end servers have a custom 8-lane PCIe slot that prevents you from installing a 16-lane PCIe video board without serious hacking (literally).
Well, it's not much of a feature if your other drivers require a reboot. This problem manifests itself whenever you do a reinstall of the operating system. First thing you would do is install Windows, followed by getting all your hardware devices recognized via driver installs, then go to Windows Update. In many cases, each time a driver is installed, you need to reboot: sound, graphics, chipset, etc. What should happen is the driver installations should be smart enough so that they can be batched in one swoop like what happens for Windows Update.
I have done some installs where I click "No" when the system asked me if I wanted to restart. I never felt comfortable with that because in the back of my mind, I'm thinking "maybe some program will do some cleanup as I reboot" and that program won't run if I do another install. I have also found that even after a reboot immediately after installation, temporary directories aren't cleaned up.
Normal maintenance is not too big a deal to restart, however if a server is busy providing services, you would have to schedule downtime.
I eat to live. In this day and age of corporate mongering by upper level execs (some exceptions noted: Xilinx among a few), I would voice agreement with Michael Rubin in the article: "I'd rather see the money in my paycheck."
I've had direct experience with developing middleware that accepted SQL and have that SQL talk to all of the prevalent RDBMSes back in 1996 or so. It can be a pain to support some of the more esoteric features from one vendor in another, but not really that hard. If we got stuck, we consulted with Oracle, Sybase, RedBrick, whoever. Doing this will add value to your product; and if done right, you can reap a lot of revenue from it.
Well, talk about uncommented code. I'm sure maybe a good portion (yes, talking outta my ass here) of the OSS projects out there are in this category. How many times have you've read "well, the code is a mess, and really needs to be reorganized, but here it is--that's why it's labeled version 0.x"?
Depending on the relative skills between two engineers, one may find code fragment A obvious while another don't. Yes, comments are valuable to a certain extent, but Apple's code is not an academic exercise where they're being graded on how their code looks.
It seems from reading page 10 of the article that John described the way Spotlight should behave like how Microsoft has decided to implement their version in Longhorn. Rather than coupling MS's solution with just the file system and be relegated to just working with files, Longhorn will integrate it with web searches, and other stuff. Seems reasonable that MS heard about Spotlight when its Entourage team was working with Apple on this aspect of Tiger.
the main problem is that Apple's internal cvs history is peppered with references they can't open to the public
Therein lies the rub. Off the top of my head, if I were an Apple developer and I wanted to send the source and logs out to KHTML, I would have to get not only the source but the cvs logs cleared by both management and lawyers to ensure I don't leak anything out. Imagine having management and lawyers looking at code and deciding what to black out. The overhead required would make any developer cringe. Probably the release to KHTML would be months after Apple's initial release, especially after a major OS upgrade like Tiger.
No, Apple does not say Intel-based Macs will have a BIOS. It just so happens that these devkits' machines have a Phoenix BIOS. The pdf that everyone's been referring to only says that there is no Open Firmware, there is no mention of BIOS.
You're assuming too many things will stay the way they are from the devkits' architecture.
It took me about 2 minutes to learn the finesse of doing 1-1/2 taps that are required for click lock for dragging--tap once then doing 1/2 a tap by pressing your finger on the pad and while not releasing move your finger.
A UPS will take care of this situation, no?
Yes, you're the only one.
Touchpads are much better because with the correct driver you can do everything with a touchpad that you can with a mouse/trackball including click, drag, and click/drag lock without using even the buttons. A truly solid state device with no moving parts. Apple does one better by allowing you to scroll using 2 fingers on the pad without the use of dedicated regions.
I have never liked the trackpoints, especially with my tendonitis. Doing click drags just feels unnatural.
It's really no use these days as most people use the Internet for a lot of stuff. When power goes out, your connection goes with it unless you have a UPS.
No, it doesn't. These "super" CDs use a Microsoft utility to bind common files in all forms of Windows XP (for instance, there are ones for Office, Windows 2000, 2003) to one physical location on the CD instead of different and separate locations. Saves a lot of space as there are a lot of common files.
It shows just because something is old doesn't automatically mean it's useless.
I'm really curious if this is indeed an incremental patch or Mozilla's idea of one--namely a complete download of the product.
Oh goodie, one of these "nya nya nya, mine's bigger than yours" questions.
Hey, I have 1 computer for every room in my house, including the attic. Surf anywhere on 1GBe HW for everything. Wireless everywhere. Servers everywhere. Mine's bigger than yours. Well, I'm kidding of course.
Anything more than 1 computer/person and it's a luxury.
Not quite. Congress, the legislative branch makes the laws; police/armed forces, the executive branch enforces the laws; courts, the judicial branch interprets the laws.
Maybe Gates is talking about this: http://www.mobilegazette.com/motorola-e790-rumours .htm
What about this badboy? http://www.dpreview.com/news/0503/05030901samsung_ schv770.asp
http://www.infosyncworld.com/news/n/5879.html
I think it goes like this: Modern hard drives require a thin film of air for the heads to ride on. Breather holes on the tops (usually) of HDs let the air in--they get filtered.
Yup, I agree with you. What the original submitter doesn't realize is if you get a good mobo with onboard NIC (GBe or faster) that doesn't go through the PCI bus, he will get much better throughput. If all the cards are used simultaneously, there will be unwanted PCI bus contention. Probably not an issue with PCIe, but it is probably wise to shop for a good "all-in-one." Of course, there are cheaper mobos where onboard NICs still go through the PCI bus--I try to avoid those.
Is this an old Dell or one from around 3 years ago? I have found that recent Dells have switched to standard PC power supplies (old ones had different pin layout) and have drive rails for all available bays.
What bites is some of their low end servers have a custom 8-lane PCIe slot that prevents you from installing a 16-lane PCIe video board without serious hacking (literally).
The Arstechnica review describes scaling under Tiger.
Well, it's not much of a feature if your other drivers require a reboot. This problem manifests itself whenever you do a reinstall of the operating system. First thing you would do is install Windows, followed by getting all your hardware devices recognized via driver installs, then go to Windows Update. In many cases, each time a driver is installed, you need to reboot: sound, graphics, chipset, etc. What should happen is the driver installations should be smart enough so that they can be batched in one swoop like what happens for Windows Update.
I have done some installs where I click "No" when the system asked me if I wanted to restart. I never felt comfortable with that because in the back of my mind, I'm thinking "maybe some program will do some cleanup as I reboot" and that program won't run if I do another install. I have also found that even after a reboot immediately after installation, temporary directories aren't cleaned up.
Normal maintenance is not too big a deal to restart, however if a server is busy providing services, you would have to schedule downtime.
Your reply sounds like it comes from a geek, or perhaps wannabe geek. Well, just check out the source and tell us. http://www.opensource.apple.com/darwinsource/10.4/ launchd-106
I eat to live. In this day and age of corporate mongering by upper level execs (some exceptions noted: Xilinx among a few), I would voice agreement with Michael Rubin in the article: "I'd rather see the money in my paycheck."
I skipped this section. After reading it, I would say "yes."
They're available, as John has posted the link in the Ars discussion thread. http://www.opensource.apple.com/darwinsource/10.4
I've had direct experience with developing middleware that accepted SQL and have that SQL talk to all of the prevalent RDBMSes back in 1996 or so. It can be a pain to support some of the more esoteric features from one vendor in another, but not really that hard. If we got stuck, we consulted with Oracle, Sybase, RedBrick, whoever. Doing this will add value to your product; and if done right, you can reap a lot of revenue from it.
Well, talk about uncommented code. I'm sure maybe a good portion (yes, talking outta my ass here) of the OSS projects out there are in this category. How many times have you've read "well, the code is a mess, and really needs to be reorganized, but here it is--that's why it's labeled version 0.x"?
Depending on the relative skills between two engineers, one may find code fragment A obvious while another don't. Yes, comments are valuable to a certain extent, but Apple's code is not an academic exercise where they're being graded on how their code looks.
It seems from reading page 10 of the article that John described the way Spotlight should behave like how Microsoft has decided to implement their version in Longhorn. Rather than coupling MS's solution with just the file system and be relegated to just working with files, Longhorn will integrate it with web searches, and other stuff. Seems reasonable that MS heard about Spotlight when its Entourage team was working with Apple on this aspect of Tiger.
Therein lies the rub. Off the top of my head, if I were an Apple developer and I wanted to send the source and logs out to KHTML, I would have to get not only the source but the cvs logs cleared by both management and lawyers to ensure I don't leak anything out. Imagine having management and lawyers looking at code and deciding what to black out. The overhead required would make any developer cringe. Probably the release to KHTML would be months after Apple's initial release, especially after a major OS upgrade like Tiger.