I doubt Microsoft will license WM9 without the kind of OS support for strong DRM that's in NT kernel. That's easy to do in embedded devices, but it's not likely to happen on Mac OS X with its open-source kernel.
[insert conspiracy theory about Microsoft being responsible for getting Apple to switch to Intel to support WM9 DRM... it's no stranger than some of the ones that have been tossed around in/.]
It's Konfabulator done right - the widgets can be brought into view whenever you actually need them rather than having them hiding back on your desktop.
You could always use Expose' to do that. Trust Apple to come up with a new scheme that makes it impossible to have a widget that you want to keep an eye on sitting around all the time.
Ah well, hopefully someone will come up with a Dashboard/Konfabulator "Amnesty" hack that will let you use both kinds of widgets from whatever interface you want. There's already the Amnesty Widget browser that lets you free Dashboard widgets from the Dashboard... but unfortunately they tend to be CPU hogs...
One hopes they turn off the registration code soon.:)
Also, not specifically related to Konfabulator...
Once the download finishes, a disk image will be mounted automatically.
I hope that everyone has already disabled automatic mounting of downloaded disk images, along with all other "open safe files after download"-style actions.:)
Surely you wouldn't inadvertently use an inappropriate apostrophe when flaming about grammar, would you? That must have been teh deliberate.
Re:Why FreeBSD is not good for most businesses
on
Why FreeBSD
·
· Score: 1
What if it decides to evolve more rapidly in future?
It would be more likely that Linus would switch to a Windows-CE based kernel for Linux 2.8.
How are these different from core? Because they don't come with the distro now? [...] Why do I need seperate install proceedures for software that came with my distro and those that don't?
Because it's not a "distro", it's an "operating system". They're not separate packages, there's a complete source tree containing the kernel and all the userland components required for the OS. Even the "contrib" section is maintained by the FreeBSD project, not just turned into an RPM and bundled.
What if my OS update updates something relied upon by something that was installed later?
The OS will never update anything under/usr/local, so it's never going to clobber anything you've installed or that a package has installed. It also doesn't pull anything out that you might be depending on, new libraries go in with new version numbers. This kind of consistency does take more work from the FreeBSD Project, but it's one of the reasons people use it.
Re:Why FreeBSD is not good for most businesses
on
Why FreeBSD
·
· Score: 1
FreeBSD wasn't even mentioned until the interview. In fact, when I interviewed, I was asked, "have you ever heard of FreeBSD?" with a grimace like I was going to go, "WTF? FreeBZ whatis?" The fact I said, "Which build, the 4.x line or the newer 5.x line?" made them go, "WOW! WE FOUND ONE!"
That actually supports my point. If they went "wow, we found one" that sure sounds like they were looking for someone who knew FreeBSD, NOT someone who knew UNIX.
Sometimes there's not an RPM when we need it, but that's just a configure and a "make all && make install" away...
Not if you're packaging stuff for reproduction and distribution. On FreeBSD, installations never touch the core system. On Red Hat, there is no core system, and I eventually had to use custom RPMs and a hundred line install script to (carefully) back out several system packages and replace them with the right version, then install and configure the rest of the system. If I didn't do that, they'd get clobbered by the updater that didn't know about my "bandit" installs.
Both offer support. Both are answerable to customer pressure.
Microsoft? Answerable to customer pressure? You're pulling my leg, mate. If Microsoft was answerable to customer pressure, they'd have backed out the IE-Desktop integration in 1997, and we wouldn't STILL be fighting ActiveX and cross-zone attacks.
We can call Red Hat, call HP, get a live guy, and go, "What the heck is wrong with your HBA support an the 2.6 kernel?"
I get that kind of support from HP on Tru64 and HPUX, but not on Red Hat. There, I get "dunno, try another version"... when they call back after I've already fixed the problem myself. That's similar to the support I get from Microsoft, where their "support" people's advice left half my users unable to log on to the domain, and when I called back they told me to buy more support before they'd talk to me... no exceptions for problems *they* had caused. I got that one fixed myself before they called back with an apology and a free extension.
Re:Why FreeBSD is not good for most businesses
on
Why FreeBSD
·
· Score: 1
Really? That's funny. None of them answered ads on Dice or Careerbuilder.
For commercial UNIX expertise, or for FreeBSD expertise? I don't expect your typical Sun or AIX guy to automatically zone in on "hey, maybe I ought to respond to a FreeBSD ad".
A "completely different structure?"
Different installer, different packages, different startup scripts and so different configurations. Oh, and the whole "which desktop is our favorite" dance.
Now I didn't use Red Hat between 7.1 and RHES 3, so maybe they've finally settled that down a bit... most of my problems with RHES have been due to having to go find packages for software that's a higher "level" on RHN on various "we're not Red Hat we just package the same stuff" sites.
My experience with corporate support of Linux has been similar to my experience with corporate support of Windows... primarily a collection of artificial barriers to information to keep you coming back and paying for more support.
Yes, you can put attach external storage thru the usb2 and firewire ports but requiring usb and firewire bridgeboards on the external drives ups the cost of external storage considerably.
Funny. External 1394 and USB2 cases seem to be about the same as, or cheaper than external SATA cases. Googling around I'm actually finding a lot of cases where you can get the same case with SATA, Firewire, USB, or in some cases USB *and* Firewire... for the same price.
If the eMac had a decent monitor, I'd have bought one a long time ago. But since Steve Jobs took over, the quality of Apple's displays has taken a back seat to cool looks. Even the laptops have lower resoution than comparable Wintel laptops, except at the very low end of the range... but the eMac is really the pits.
My old 16" multisync had great picture quality for its time, and was a nice Trinitron tube with totally crisp and clear text. The eMac has a cheap-looking shadow mask that's distracting and painfully blurry no matter what the resolution. If I were to buy one I'd end up having to leave it on the floor like it was an overfed minitower and hook it in to something that doesn't make my eyes physically hurt after using it all day.
If Aple were to ship the eMac without the tube, as a kind of return to the NeXT Slab, I would be inclined to buy it over the Mini even if it was $650 by itself. But as it is... no.
Have you even looked at Apple's website?
on
Why FreeBSD
·
· Score: 1
All the sudden, it makes perfect sense why Apple and MS like the BSD license, and IBM, SGI, HP, and everyone else who competes with MS like the GPL.
Apple has released the source to every major and minor version of just about everything in Mac OS X below the Quartz layer, and a bunch of stuff above it. Even if BSD had been GPLed, they've gone far beyond the requirements of the GPL by not only releasing the source to the code that's linked with the kernel, but just about all the important utilities including applications like launchd that are part of their competitive advantage over other UNIX platforms. They have released not only their enhancements to KHTML but all the glue necessary to rebuild Webcore and virtually every important part of Safari, not only as tarballs but also through repository access. This is a HUGE amount of work from a company as small as Apple.
Apple didn't use BSD because they wanted to keep their source code private. The source code that they're keeping private is highly portable and was already running under Solaris and Windows, using standard APIs... they could have used any kernel they wanted. They could have even used Linux without getting into trouble with the GPL, the way they used GCC, but they decided to continue with their existing BSD code base and FreeBSD... and then kept it open source.
IBM? HP? SGI? Where can I download OpenAIX, OpenHPUX, and OpenIrix? THAT's the kind of code release that would be comparable to what Apple's done.
Microsoft? I wish they'd use MORE BSD code in their kernel, not less. They don't like any flavor of open source all that much, it makes it too hard for them to do things like maintaining an incompatible sockets library.
Where's that "low end Macs come first" quote?
on
New Apples Next Week
·
· Score: 1
I'm pretty sure Steve specifically said that the low end systems would go first.
I thought I remembered that but I couldn't find the quote.
It would be nice if someone would dig it up and put it to rest one way or the other.
Re:Kernel performance
on
Why FreeBSD
·
· Score: 2, Interesting
If you want to seriously compare two open-source Unix-like systems, the only instrinsic difference is the kernel. Arguing that one system is better because of the default configuration of network services, the package system, the organization of the rc scripts, and so on, is a red herring, because there is no reason you can't take all of the userspace from one system and run it on top of the kernel from the other -- and there are projects which do this.
I've looked at every "Linux kernel with a BSD-like userland" package I've found, and I haven't found a "Linux kernel with a BSD-like userland" yet. And I haven't seen one that claimed that that had the same userland two years running. The way Linux is built out of packages, rather than having a stable core OS that's more than just the kernel, ensures that.
Linux kernel performance has gotten better on some of these benchmarks, to the point where it's comparable to FreeBSD... ahead in some places, behind in others. The only place it seems to choke is a test that the author acknowledges isn't realistic: there's no reason to expect that mmapping every other page of a 200M file will behave similarly to performing the same number of mmaps on small files. Linux filesystem performance has often been very good, too: until FreeBSD got softupdates, Linux was clearly ahead there (albeit at the cost of stability).
The big problem with the Linux kernel is that it's unstable. Not unstable as in "it crashes too often", but unstable in "it changes too fast". 2.6 is changing so fast that Linus had to take time out to write a version control system that could handle his workload. And there's no indication what kernel APIs should be considered stable and which are subject to being pulled out from under you.
I've been able to debug problems in Tru64 UNIX (based on 4.3-Reno) using the FreeBSD (based on 4.4-Lite) source tree. Apple Merged NeXTstep and FreeBSD to form Darwin and has been pulling in chunks of FreeBSD kernel code into Darwin on a regular basis. These are source trees that forked years ago, and they're still close enough to make this kind of thing reasonable.
The BSD userland is similarly stable and reliable over the long term, and across separate systems. Linux? I went from Red Hat 2.1 to 4.1 to 6.0 to 7.1, to RHEL 3... and it was a different OS every time. And that doesn't begin to address the differences between Debian and Gentoo and Red Hat.
Re:Why FreeBSD is not good for most businesses
on
Why FreeBSD
·
· Score: 5, Insightful
The company searched for someone with FreeBSD experience.
That was a mistake. They should have earched for someone with mainstream UNIX experience. Anyone who's familiar with any commercial UNIX... Solaris, AIX, HPUX, whatever... will find FreeBSD a familiar environment. The details are different, but the BSD environment is baked into the genes of every commercial UNIX out there.
And there's lots of people who know UNIX who can pick up FreeBSD far far quicker than they can pick up Linux.
For example...
If FreeBSD had [...] a well-thought out directory structure (I have boot scripts in/etc and/usr/local/etc... and have you ever had to diagnose which one broke?),
That is a well-thought-out directory structure. You have the operating system, a fixed core that's evolved only gradually over the past 15 years, and add-on packages. You upgrade the OS, your packages don't get touched. You upgrade a package, the OS doesn't get touched. And your oldschool SunOS guys? They'll have no problem diagnosing which one broke.
I've used Red Hat versions since 2.1. Every major version has had a completely different structure. You don't have any border between the OS and add-ons, so when you go to upgrade you have to take all-or-nothing. Over the short term I can see the advantage of Red Hat's model, but over the long term you've got to start over again and again and again.
They could have called it the G5 Mobile. :)
on
New Apples Next Week
·
· Score: 1
If only they could find a chip to put in with lower power requirements in the performance range they need...
That would be the MPC8641 and MPC8641D, the high end e600 models from Freescale, which will be sampling this year and should be ready for production long before Apple could manage any major change like switching to a different processor architecture.
Unfortunately, it's unlikely that Apple will use this new chipset and will instead hold off any significant powerbook upgrade for another year. They may produce a slightly faster powerbook using the low-end e600 since it won't require a complete logic board redesign like the high-end models would.
And there's good reasons for designing it that way.
What if you were to merge/etc/passwd and/etc/group into a single file?
Well, if you were to do that you'd have one file containing two completely different types of records... one for users, and one for groups. You'd have to make sure you queried for the "user" or the "group" record, so instead of asking for "name=='fred'" you'd have to ask for "name=='fred' && type=='user'", to make sure you didn't get the wrong one. It sure sounds like you'd make it more complex to get information out of the file, regardless of the format. No, I think the reasons for having separate files for passwd and group are entirely sound.
You can't get an iBook for under $1000, unless you buy it used or refurbished. No, calling $999 "under $1000" is cheating.
You can get refurb sub-5-pound Thinkpad T21s for under $500. And that's *really* under $500, not "if you don't include tax" under $500. You can't even get a used G3 icebook for that.
One nice thing about Linux on a Mac is that you can still run OS X programs at almost native speed.
MoL? What does it do about video acceleration?
The nice thing about Mac OS X, of course, is that you can still run Linux programs at almost native speed. Well, um, really, that would be "at native speed", wouldn't it? ^_^;;;
dammit! I hate it when I make typos while I'm being a pedant!
You are simply embracing and promoting one of the oldest rules on the net. It's older than slashdot, older than the public Internet, it goes back to the earliest days of Usenet. It's called the spelling rule: every spelling flame must have at least one spelling misteak or grammatical error.
Don't be upset, be proud, it's a grand tradition you're following.
Once I'd factored in the SuperDrive (for making DVD's), Airport (for using it in the bedroom upstairs) - because you can't fit it yourself and bluetooth because if you're having the Airport installed you might as well and all kids these days have Bluetooth capable phones and some extra RAM as 256 Mb just isn't enough, it was a *LOT* more.
I got an external DVD burner that was DL-capable for less than the superdrive upgrade. Wifi and Bluetooth are available USB, and you're still better off drilling a hole in the floor and running a cable upstairs than using Wifi. The only non-optional upgrade for the Mini, really, is the extra 256M of RAM. That should be in the base unit... but since it's about a 10% price difference that's no biggy.
Personally, I think Apple should bring back the slab. Something like the NeXT slab or the Performa 475, but bigger than the Mini. An eMac without the monitor. Big enough for a 3.5" drive and two RAM slots, and maybe one PCI-Express slot for video. Stick it in between the Mini and the iMac in price.
MUCH better GPU, slightly faster CPU with a larger cache, dual-screen support, higher resolution screen on the two larger models.
My question is... apart from OSX, why would anyone buy a Powerbook or iBook? I occasionally see people running Linux on their 'books and wonder why they didn't get a good laptop like a Thinkpad instead.
I'm talking about nothing more than making XML as easy to use as flat database-style files. I'm not even asking you to do a better job, just to provide the capabilities including the ability to write forensic scripts that will work in single-user mode on a minimal system. You keep giving me solutions that impose significant costs in terms of the minimal requirements to actually get my work done.
But there are scripting languages that are aware of XML, and there have been for years.
There are scripting languages with fairly sizable XML libraries that have to be parsed and interpreted every time the script is run. This is unacceptable: too high a startup cost, too many files have to exist and be uncorrupted.
External programs don't help, because parsing and dequoting the output of the external program, and making sure you're not subject to eval- or quoting- exploits in the process of doing so, is actually harder than just reading a flat file safely.
Show me a native interface to a compiled parser (this C-language XML parser you're talking about) that doesn't itself require significant overhead in time, space, or required files.
You seemed to want it to be a replacement for 'awk' and 'sort' etc.
Go back and re-read the thread back to my first comment. Re-read the message I'm responding to. I'm not trying to make XML a substitute for make, etcetera, the person who was proposing to replace all the files in/etc with XML is. THAT is the topic of THIS subthread.
DO that, then try again. I'm not going to try and explain what I'm getting at with someone who's having a completely different discussion. For example, the fact that there's an XML library for C is irrelevant if you don't have a scripting language that's aware of it, but I can't explain why if you're thinking about "ant" while I'm talking about it.
Slashdot Mirror
I doubt Microsoft will license WM9 without the kind of OS support for strong DRM that's in NT kernel. That's easy to do in embedded devices, but it's not likely to happen on Mac OS X with its open-source kernel.
... it's no stranger than some of the ones that have been tossed around in /.]
[insert conspiracy theory about Microsoft being responsible for getting Apple to switch to Intel to support WM9 DRM
Ironic how you missed the comments about Dashboard being a CPU hog and how it'd be nice to have one program that could run any widget...
But of course you coudln't get your backhanded slap in if you actually read the comments before adding your own...
It's Konfabulator done right - the widgets can be brought into view whenever you actually need them rather than having them hiding back on your desktop.
You could always use Expose' to do that. Trust Apple to come up with a new scheme that makes it impossible to have a widget that you want to keep an eye on sitting around all the time.
Ah well, hopefully someone will come up with a Dashboard/Konfabulator "Amnesty" hack that will let you use both kinds of widgets from whatever interface you want. There's already the Amnesty Widget browser that lets you free Dashboard widgets from the Dashboard... but unfortunately they tend to be CPU hogs...
One hopes they turn off the registration code soon. :)
:)
Also, not specifically related to Konfabulator...
Once the download finishes, a disk image will be mounted automatically.
I hope that everyone has already disabled automatic mounting of downloaded disk images, along with all other "open safe files after download"-style actions.
To grammar nazi's
Grammer nazi's what?
Surely you wouldn't inadvertently use an inappropriate apostrophe when flaming about grammar, would you? That must have been teh deliberate.
What if it decides to evolve more rapidly in future?
/usr/local, so it's never going to clobber anything you've installed or that a package has installed. It also doesn't pull anything out that you might be depending on, new libraries go in with new version numbers. This kind of consistency does take more work from the FreeBSD Project, but it's one of the reasons people use it.
It would be more likely that Linus would switch to a Windows-CE based kernel for Linux 2.8.
How are these different from core? Because they don't come with the distro now? [...] Why do I need seperate install proceedures for software that came with my distro and those that don't?
Because it's not a "distro", it's an "operating system". They're not separate packages, there's a complete source tree containing the kernel and all the userland components required for the OS. Even the "contrib" section is maintained by the FreeBSD project, not just turned into an RPM and bundled.
What if my OS update updates something relied upon by something that was installed later?
The OS will never update anything under
FreeBSD wasn't even mentioned until the interview. In fact, when I interviewed, I was asked, "have you ever heard of FreeBSD?" with a grimace like I was going to go, "WTF? FreeBZ whatis?" The fact I said, "Which build, the 4.x line or the newer 5.x line?" made them go, "WOW! WE FOUND ONE!"
That actually supports my point. If they went "wow, we found one" that sure sounds like they were looking for someone who knew FreeBSD, NOT someone who knew UNIX.
Sometimes there's not an RPM when we need it, but that's just a configure and a "make all && make install" away...
Not if you're packaging stuff for reproduction and distribution. On FreeBSD, installations never touch the core system. On Red Hat, there is no core system, and I eventually had to use custom RPMs and a hundred line install script to (carefully) back out several system packages and replace them with the right version, then install and configure the rest of the system. If I didn't do that, they'd get clobbered by the updater that didn't know about my "bandit" installs.
Both offer support. Both are answerable to customer pressure.
Microsoft? Answerable to customer pressure? You're pulling my leg, mate. If Microsoft was answerable to customer pressure, they'd have backed out the IE-Desktop integration in 1997, and we wouldn't STILL be fighting ActiveX and cross-zone attacks.
We can call Red Hat, call HP, get a live guy, and go, "What the heck is wrong with your HBA support an the 2.6 kernel?"
I get that kind of support from HP on Tru64 and HPUX, but not on Red Hat. There, I get "dunno, try another version"... when they call back after I've already fixed the problem myself. That's similar to the support I get from Microsoft, where their "support" people's advice left half my users unable to log on to the domain, and when I called back they told me to buy more support before they'd talk to me... no exceptions for problems *they* had caused. I got that one fixed myself before they called back with an apology and a free extension.
You can't get that with FreeBSD.
http://www.freebsd.org/commercial/consult.html
Really? That's funny. None of them answered ads on Dice or Careerbuilder.
For commercial UNIX expertise, or for FreeBSD expertise? I don't expect your typical Sun or AIX guy to automatically zone in on "hey, maybe I ought to respond to a FreeBSD ad".
A "completely different structure?"
Different installer, different packages, different startup scripts and so different configurations. Oh, and the whole "which desktop is our favorite" dance.
Now I didn't use Red Hat between 7.1 and RHES 3, so maybe they've finally settled that down a bit... most of my problems with RHES have been due to having to go find packages for software that's a higher "level" on RHN on various "we're not Red Hat we just package the same stuff" sites.
My experience with corporate support of Linux has been similar to my experience with corporate support of Windows... primarily a collection of artificial barriers to information to keep you coming back and paying for more support.
http://eshop.macsales.com/shop/ministack/B X_888_intro.asp?pid=USBX-888
http://www.lacie.com/products/range.htm?id=10033
http://www.adstech.com/products/USBX-888/intro/US
http://www.micronet.com/General/minimate.asp
Yes, you can put attach external storage thru the usb2 and firewire ports but requiring usb and firewire bridgeboards on the external drives ups the cost of external storage considerably.
Funny. External 1394 and USB2 cases seem to be about the same as, or cheaper than external SATA cases. Googling around I'm actually finding a lot of cases where you can get the same case with SATA, Firewire, USB, or in some cases USB *and* Firewire... for the same price.
If the eMac had a decent monitor, I'd have bought one a long time ago. But since Steve Jobs took over, the quality of Apple's displays has taken a back seat to cool looks. Even the laptops have lower resoution than comparable Wintel laptops, except at the very low end of the range... but the eMac is really the pits.
My old 16" multisync had great picture quality for its time, and was a nice Trinitron tube with totally crisp and clear text. The eMac has a cheap-looking shadow mask that's distracting and painfully blurry no matter what the resolution. If I were to buy one I'd end up having to leave it on the floor like it was an overfed minitower and hook it in to something that doesn't make my eyes physically hurt after using it all day.
If Aple were to ship the eMac without the tube, as a kind of return to the NeXT Slab, I would be inclined to buy it over the Mini even if it was $650 by itself. But as it is... no.
All the sudden, it makes perfect sense why Apple and MS like the BSD license, and IBM, SGI, HP, and everyone else who competes with MS like the GPL.
... they could have used any kernel they wanted. They could have even used Linux without getting into trouble with the GPL, the way they used GCC, but they decided to continue with their existing BSD code base and FreeBSD... and then kept it open source.
Apple has released the source to every major and minor version of just about everything in Mac OS X below the Quartz layer, and a bunch of stuff above it. Even if BSD had been GPLed, they've gone far beyond the requirements of the GPL by not only releasing the source to the code that's linked with the kernel, but just about all the important utilities including applications like launchd that are part of their competitive advantage over other UNIX platforms. They have released not only their enhancements to KHTML but all the glue necessary to rebuild Webcore and virtually every important part of Safari, not only as tarballs but also through repository access. This is a HUGE amount of work from a company as small as Apple.
Apple didn't use BSD because they wanted to keep their source code private. The source code that they're keeping private is highly portable and was already running under Solaris and Windows, using standard APIs
IBM? HP? SGI? Where can I download OpenAIX, OpenHPUX, and OpenIrix? THAT's the kind of code release that would be comparable to what Apple's done.
Microsoft? I wish they'd use MORE BSD code in their kernel, not less. They don't like any flavor of open source all that much, it makes it too hard for them to do things like maintaining an incompatible sockets library.
I'm pretty sure Steve specifically said that the low end systems would go first.
I thought I remembered that but I couldn't find the quote.
It would be nice if someone would dig it up and put it to rest one way or the other.
If you want to seriously compare two open-source Unix-like systems, the only instrinsic difference is the kernel. Arguing that one system is better because of the default configuration of network services, the package system, the organization of the rc scripts, and so on, is a red herring, because there is no reason you can't take all of the userspace from one system and run it on top of the kernel from the other -- and there are projects which do this.
I've looked at every "Linux kernel with a BSD-like userland" package I've found, and I haven't found a "Linux kernel with a BSD-like userland" yet. And I haven't seen one that claimed that that had the same userland two years running. The way Linux is built out of packages, rather than having a stable core OS that's more than just the kernel, ensures that.
Linux kernel performance has gotten better on some of these benchmarks, to the point where it's comparable to FreeBSD... ahead in some places, behind in others. The only place it seems to choke is a test that the author acknowledges isn't realistic: there's no reason to expect that mmapping every other page of a 200M file will behave similarly to performing the same number of mmaps on small files. Linux filesystem performance has often been very good, too: until FreeBSD got softupdates, Linux was clearly ahead there (albeit at the cost of stability).
The big problem with the Linux kernel is that it's unstable. Not unstable as in "it crashes too often", but unstable in "it changes too fast". 2.6 is changing so fast that Linus had to take time out to write a version control system that could handle his workload. And there's no indication what kernel APIs should be considered stable and which are subject to being pulled out from under you.
I've been able to debug problems in Tru64 UNIX (based on 4.3-Reno) using the FreeBSD (based on 4.4-Lite) source tree. Apple Merged NeXTstep and FreeBSD to form Darwin and has been pulling in chunks of FreeBSD kernel code into Darwin on a regular basis. These are source trees that forked years ago, and they're still close enough to make this kind of thing reasonable.
The BSD userland is similarly stable and reliable over the long term, and across separate systems. Linux? I went from Red Hat 2.1 to 4.1 to 6.0 to 7.1, to RHEL 3... and it was a different OS every time. And that doesn't begin to address the differences between Debian and Gentoo and Red Hat.
The company searched for someone with FreeBSD experience.
/etc and /usr/local/etc... and have you ever had to diagnose which one broke?),
That was a mistake. They should have earched for someone with mainstream UNIX experience. Anyone who's familiar with any commercial UNIX... Solaris, AIX, HPUX, whatever... will find FreeBSD a familiar environment. The details are different, but the BSD environment is baked into the genes of every commercial UNIX out there.
And there's lots of people who know UNIX who can pick up FreeBSD far far quicker than they can pick up Linux.
For example...
If FreeBSD had [...] a well-thought out directory structure (I have boot scripts in
That is a well-thought-out directory structure. You have the operating system, a fixed core that's evolved only gradually over the past 15 years, and add-on packages. You upgrade the OS, your packages don't get touched. You upgrade a package, the OS doesn't get touched. And your oldschool SunOS guys? They'll have no problem diagnosing which one broke.
I've used Red Hat versions since 2.1. Every major version has had a completely different structure. You don't have any border between the OS and add-ons, so when you go to upgrade you have to take all-or-nothing. Over the short term I can see the advantage of Red Hat's model, but over the long term you've got to start over again and again and again.
If only they could find a chip to put in with lower power requirements in the performance range they need...
That would be the MPC8641 and MPC8641D, the high end e600 models from Freescale, which will be sampling this year and should be ready for production long before Apple could manage any major change like switching to a different processor architecture.
Unfortunately, it's unlikely that Apple will use this new chipset and will instead hold off any significant powerbook upgrade for another year. They may produce a slightly faster powerbook using the low-end e600 since it won't require a complete logic board redesign like the high-end models would.
Because its been designed that way.
/etc/passwd and /etc/group into a single file?
And there's good reasons for designing it that way.
What if you were to merge
Well, if you were to do that you'd have one file containing two completely different types of records... one for users, and one for groups. You'd have to make sure you queried for the "user" or the "group" record, so instead of asking for "name=='fred'" you'd have to ask for "name=='fred' && type=='user'", to make sure you didn't get the wrong one. It sure sounds like you'd make it more complex to get information out of the file, regardless of the format. No, I think the reasons for having separate files for passwd and group are entirely sound.
Can I get a sub-5-pound Thinkpad for under $1000?
You can't get an iBook for under $1000, unless you buy it used or refurbished. No, calling $999 "under $1000" is cheating.
You can get refurb sub-5-pound Thinkpad T21s for under $500. And that's *really* under $500, not "if you don't include tax" under $500. You can't even get a used G3 icebook for that.
The "apart from OS X" is a biggie, of course.
Well, yes. ^_^
One nice thing about Linux on a Mac is that you can still run OS X programs at almost native speed.
MoL? What does it do about video acceleration?
The nice thing about Mac OS X, of course, is that you can still run Linux programs at almost native speed. Well, um, really, that would be "at native speed", wouldn't it? ^_^;;;
dammit! I hate it when I make typos while I'm being a pedant!
You are simply embracing and promoting one of the oldest rules on the net. It's older than slashdot, older than the public Internet, it goes back to the earliest days of Usenet. It's called the spelling rule: every spelling flame must have at least one spelling misteak or grammatical error.
Don't be upset, be proud, it's a grand tradition you're following.
Once I'd factored in the SuperDrive (for making DVD's), Airport (for using it in the bedroom upstairs) - because you can't fit it yourself and bluetooth because if you're having the Airport installed you might as well and all kids these days have Bluetooth capable phones and some extra RAM as 256 Mb just isn't enough, it was a *LOT* more.
I got an external DVD burner that was DL-capable for less than the superdrive upgrade. Wifi and Bluetooth are available USB, and you're still better off drilling a hole in the floor and running a cable upstairs than using Wifi. The only non-optional upgrade for the Mini, really, is the extra 256M of RAM. That should be in the base unit... but since it's about a 10% price difference that's no biggy.
Personally, I think Apple should bring back the slab. Something like the NeXT slab or the Performa 475, but bigger than the Mini. An eMac without the monitor. Big enough for a 3.5" drive and two RAM slots, and maybe one PCI-Express slot for video. Stick it in between the Mini and the iMac in price.
why exactly you should anybody buy a PowerBook?
MUCH better GPU, slightly faster CPU with a larger cache, dual-screen support, higher resolution screen on the two larger models.
My question is... apart from OSX, why would anyone buy a Powerbook or iBook? I occasionally see people running Linux on their 'books and wonder why they didn't get a good laptop like a Thinkpad instead.
I'm talking about nothing more than making XML as easy to use as flat database-style files. I'm not even asking you to do a better job, just to provide the capabilities including the ability to write forensic scripts that will work in single-user mode on a minimal system. You keep giving me solutions that impose significant costs in terms of the minimal requirements to actually get my work done.
But there are scripting languages that are aware of XML, and there have been for years.
There are scripting languages with fairly sizable XML libraries that have to be parsed and interpreted every time the script is run. This is unacceptable: too high a startup cost, too many files have to exist and be uncorrupted.
External programs don't help, because parsing and dequoting the output of the external program, and making sure you're not subject to eval- or quoting- exploits in the process of doing so, is actually harder than just reading a flat file safely.
Show me a native interface to a compiled parser (this C-language XML parser you're talking about) that doesn't itself require significant overhead in time, space, or required files.
Here's a rather low-level one: Ciao Make.
You seemed to want it to be a replacement for 'awk' and 'sort' etc.
/etc with XML is. THAT is the topic of THIS subthread.
Go back and re-read the thread back to my first comment. Re-read the message I'm responding to. I'm not trying to make XML a substitute for make, etcetera, the person who was proposing to replace all the files in
DO that, then try again. I'm not going to try and explain what I'm getting at with someone who's having a completely different discussion. For example, the fact that there's an XML library for C is irrelevant if you don't have a scripting language that's aware of it, but I can't explain why if you're thinking about "ant" while I'm talking about it.