So you actually think that a manufacturing delay is going to translate into Apple giving you a free upgrade to a better processor that they haven't even released yet?
Absolutely. Apple is notorious for dropping production early and shipping the new product to the people who had ordered the old one, for the same price. It's a win-win situation: they get to avoid being stuck with a load of old product on hand, and the customer gets a better system than they expected.
I use iTunes and the iTMS because I know I can legally and in a supported way get out from under the DRM if I need to.
Apple's old iTunes slogan tells you how.
Oh, and Barnett said the scrutiny of Apple 'provides a useful illustration of how an attack on intellectual property rights can threaten dynamic innovation.'.
How about how the unreasoning enforcement of intellectual "property" rights threatens dynamic innovation. If iTunes was as closed as (for example) Sony's old protection schemes for their mini-discs, it would have met the same fate.
It still needs to be said, from an earlier thread:
Some people just seem to have a negative emotional response to the idea that there may be hundreds of planets in the Solar System, and that emotional response seems to be what has won the day here. There is no science behind this at all, it is a definition that is arbitrarily designed to permanently cap the number of planets to a small, manageable number. I see at least two big problems with this, first, the opening up to the possibility of there being hundreds of planets would force educators to rethink how these concepts are taught in elementary school. It could be something other than a memorize-these-nine-things exercise, but rather an opportunity to teach the basic concepts. Now it will just become a memorize-these-eight-things exercise. The other thing is that many of the newly discovered objects really merit study, and I don't see any Congress appropriating funds to send a probe to something that is just a "dwarf planet". The worst thing, though, is the pervasive media certainty that this is the end of the debate, five hundred really smart people voted on the issue and now it is settled for all eternity. I think that this new definition is deeply flawed in a number of ways, and we really need to treat this as a debate that has just begun.
The Old Mill, the classic pre-Fantasia musical short from Disney, is available for $1.99. If you haven't seen it, it's worth the price of admission. If you have, you're probably heading over there to get it already and didn't bother reading this far.:)
would you be able to somehow mitigate the their single-threaded nature by assigning the respective processes to it's own core?
First, pretty much any application on the Mac is multithreaded just because of the way the user interface works. Apple's OpenGL implementation is partly software, for example... this is why you can run hardware T&L on the Mac mini with its GMA950 GPU - the OS does that in software on the second core even in single-threaded games.
Second, OS X does a pretty goodjob of distributing applications to cores without having to explicitly bind them. Binding an application to a core would most likely slow it down... unless the program has been written to use a lot of fined grained shared state between threads... and what you're doing with processor affinity is *preventing* it from multiprocessing.
Processor affinity is like 64 bit. Unless you're doing something on the edge you probably don't need it, and if you need it you're probably already doing it.
Here's the summary:
The bad news is that OS X doesn't provide a hook for processor affinity. The good news is that Mach does support it, and you could use the Darwin sources to figure out how to implement it in OSX using direct Mach calls. The bad news is that it's really hard. The good news is you don't need to do it unless you're trying to prevent multiprocessing anyway.
I have an old G3 iMac and use XCode to develop AppleScript Studio applications.
I wouldn't expect that application programs in an interpreted language would do anything to cause panics.
I mean, at Finals Week at Berkeley in 1980 we had over 60 people at a time logged in and developing "C" code on a single PDP-11/70... the EE/CS undergrads all shared the one machine in the basement of Cory Hall.
What's most likely to cause panics is driver problems, file system problems (HFS+ isn't all that robust, I've had it blow chunks when I've left it running at less than 10% free space), and hardware problems. Application level stuff is really unlikely to panic the kernel.
Apple doesn't fill the in between market like dell tries to do. You either buy a workstation or a light, friendly desktop.
Apple doesn't *sell* a regular desktop.
What you're calling the "in between" market is most of the market. Look at what people are actually buying at any computer store: all-in-one PCs flunk in the market, every time, and the only reason Apple can sell them at all is because they make it the only option if you want a Mac... and they're losing sales from everyone who doesn't want a Mac quite that badly as a result.
They need a Mac mini "Pro", and it can't cost much more than the current mini. But it seems like Jobs is having second thoughts on going back on his "no ugly monitors" statement.
Nah, it'll have more integration with the iTunes Music Store.
It'll integrate some kind of Dashboard-like visualizer based plugin mechanism for controls that'll work better on Mac OS than Windows.
It'll have "video podcast" support and a way to let you drop your own files into the podcast tab to automatically share and stream them via.mac as podcasts posted to the iTMS.
And a new DRM scheme to allow people to put restrictions on their Podcasts and vidcasts.
It appears to be based on real world testing and not some crappy faux retro Hi-Fi panel designed by software engineers in a vacuum.
That's why the iPod interface is all but useless for adjusting by feel, and has to be locked when it's in your pocket becase just brushing it is enough to make it respond.
I gave my iPod Mini to my daughter and got the iPod Shuffle with its conventional d-Pad interface that's actually got signs of having been designed for *use* rather than *looks*.
No, what made me get an iPod at all instead of something else is that I can actually get accessories for it. And I expect I'll be able to get accessories for it in a year, and if I get a different iPod in a year the accessories I got now will still work.
I've got boxes of accessories that I'll never be able to use because nobody's every going to make a PDA compatible with the 1st generation iPaq, the Jornada, or the Visor, ever again. My old Nokia accessories don't work with my new Nokia phone. The charging cable for my old CLie doesn't work on my new one. But I can get an iPod now that will work with this docking cable from three years ago.
"Innovative" touch pad.. try scrolling through ten thousand songs precisely. Not. Happening.
I got an iPod Shuffle and gave my iPod Mini to my daughter precisely because the touch wheen is such a horrid user interface, and the D-pad on the shuffle is so much more practical for 99% of the uses.
Apple needs to get together with Sony and do an iPod that has a jog-wheel and a D-pad but is otherwise an iPod... syncs with iTunes, plays AAC and protected AAC files, and so on.
I just said that you need to provide additional efforts for synchronization for one end of the elevator. What is not clear about it?
Why you think you need to make extra efforts for synchronization of one end of the elevator. That's what's not clear, and that's what I'm trying to get across to you.
You started to talk about the usage of it. It does not matter.
The usage is exactly what does matter. You wouldn't ever expend energy to keep the elevator in an unstable orbit... you leave the elevator in a stable orbit. The centerpoint of the inter-orbital elevator is where freely-falling installations are synchronized to, not the end-points. Both the top end and bottom end are (from an energy standpoint) at the "bottom" of the elevator's "shaft". The "top of the shaft" is in the center.
This is, by the way, just like the "skyhook" style of elevator, except that they use an anchor mass to make the "top end" part of the structure shorter. For an inter-orbit elevator that's actually a disadvantage.
None of this addresses the design problems... some of them at the application level... that have caused the problem.
I didn't say, for example, that the XP firewall didn't work. What I said was that the XP firewall is a single layer of security between the Internet and the service. In UNIX, a daemon is typically capable of being configured so it's bound to a specific interface... either directly or through tcpwrappers or xinetd... whether you have the firewall running or not. In Windows, the OS API allows this, but the services don't actually give you a way to specify this, so you need the firewall to get the same iind of isolation for services that UNIX has without a firewall.
Similarly, internal application firewalls are an attempt to work around a problem in application design, one created by MS with Internet Explorer, and depended on or emulated by others since. They don't keep an application from being compromised in the first place... they just make it harder for the application to do bad things once it is compromised.
This would be a second layer of protection in other operating systems: the integration between the browser and the shell that makes Windows so susceptible to spyware and web-spread viruses is vanishingly rare outside Windows.
So the problem I'm getting at is not that the Windows XP firewall doesn't work, or that Vista's firewall doesn't work, it's that it may well be the very best solution... to the wrong problem. You're better off using a non-IE-based browser and an external NATting router (which typically costs $20-40 and usually includes a wireless access point) and Windows 2000, than with Vista and Vista's firewall and IE.
I've found a plethora of cross-platform libraries, the problem has been that they only solve a small part of the problem (typically graphics or providing a POSIX API), don't use the native user interface, or require a specific implementation language. Most of them hit all three roadbumps, even before considering the quality of the API or the license.
That's why I end up using Tk. It's not a swiss-army-knife toolkit, but it's got bindings for multiple scripting languages and *they* have bindings to multiple compiled languages and portable APIs for networking and filesystem, and it has native look and feel. Plus it's got an agreeable license.
Everyone waited forever for a credible OS like OSX, and now Apple's hardware lineup has gone to middle of the road crap. Why?
Apple's hardware has almost always been middle of the road at best, from the Apple ][ or the original badly-underpowered Macintosh onwards. They've occasionally had a high-end machine that's competitive on the hardware side, but it's been more the exception than the rule.
What I want is for them to provide at least a competant middle-of-the-road box that doesn't lock me in. I hoped the Mac mini was a sign that Apple was shedding some of Jobs' obsession with style over everything. Yes, it was stylish, but at least he had to go back on the "no ugly monitors on nice Macs" promise. But they need a "Mini Pro" with a video card slot and room for a 3.5" hard drive. It wouldn't cost them any more to make than the Mini, and would bring an awful lot of fence-sitters into the fold.
Because 90% of iMac buyers (all-in-one consumer grade machine) will never want a second hard drive.
How many potential Mac buyers are remaining potential Mac buyers because Apple's not giving them a choice?
When I got my Mac mini it was the first consumer grade Mac since the cube that was even vaguely close to my sweet spot, and the new Mini is further away. You might ask "How many Mac mini buyers (compact entertainment machine) will ever want to upgrade the video card" just as easily... and you'd be missing the point. The point is that there's a lot of people who just can't find a Mac that can replace a PC, because Apple hasn't had a machine that targets the majority of PC buyers since they discontinued the Beige G3 desktop. Most people don't want a quad-core monster, but they don't want to be locked in either.
I'm sorry if you're frustrated. I know it's frustrating when someone seems to be doing that to me, but I honestly don't understand what you're getting at.
You wrote: Besides, it will cost you to synchronize two orbits with different R, because, you know, T~R^1.5
Can you explain what you mean by this in a bit more detail, because it sounds like you're making a wildly incorect assumption about inter-orbit elevators and how they're used... but I don't know what that assumption is... and obviously I'm doing a lousy job of guessing.
So, coudl you describe what you THINK I'm talking about here?
Surely there's a cross platform API that's better than the godawful bend-over-for-Microsoft Win32 API.
I've always written for the base UNIX API (which is available under Windows, and will be a native API in Vista if (as reported) Interix is included) with the GUI written using a scripting language and Tk. This produces apps that have a native user interface under UNIX, Windows, Mac OS X, and even (with care) old Mac OS. Surely I can't be the only one.
That is what I said: you have to put additional efforts to synchronize the ends of the elevator.
That happens automatically through what are commonly known as tidal forces.
The only effort that matters is the energy spent changing your velocity using a reaction drive. Since the low end moves below orbital velocity, you need less delta-v and thus less fuel and acceleration to dock with it from a lower orbit. Since the high end moves above orbital velocity, it takes less fuel and acceleration to dock with it from a higher orbit. The result is a net reduction in the energy required to get to space... to the point where an air-breathing hypersonic vehicle can do the job.
This means *less* effort overall, and *less* effort to dock with the elevator.
The point of inter-orbital elevators is to have the end-points moving above or below orbital velocity for that orbit.
Besides, it will cost you to synchronize two orbits with different R, because, you know, T~R^1.5
Not if you don't circularize the orbits first. If you're going to "land" on the low-end of an inter-orbit elevator it's a lot cheaper to just target it at apogee and grab it when you get there. The longer the elevator is, the more elliptical your intercept orbit can be, and the less fuel you have to use to synchronise with it.
And of course the opposite is true for the high end of the elevator and the perigee of your transfer orbit.
The most obvious application of this is an elevator you can reach from a hypersonic mostly-air-breathing craft on a ballistic arc out of the atmosphere.
Slashdot Mirror
I'm still stunned at the level of personal attachment people feel over planetary names.
You can't call a planet "Bob"!
So you actually think that a manufacturing delay is going to translate into Apple giving you a free upgrade to a better processor that they haven't even released yet?
Absolutely. Apple is notorious for dropping production early and shipping the new product to the people who had ordered the old one, for the same price. It's a win-win situation: they get to avoid being stuck with a load of old product on hand, and the customer gets a better system than they expected.
Yep.
I use iTunes and the iTMS because I know I can legally and in a supported way get out from under the DRM if I need to.
Apple's old iTunes slogan tells you how.
Oh, and Barnett said the scrutiny of Apple 'provides a useful illustration of how an attack on intellectual property rights can threaten dynamic innovation.'.
How about how the unreasoning enforcement of intellectual "property" rights threatens dynamic innovation. If iTunes was as closed as (for example) Sony's old protection schemes for their mini-discs, it would have met the same fate.
It still needs to be said, from an earlier thread:
Some people just seem to have a negative emotional response to the idea that there may be hundreds of planets in the Solar System, and that emotional response seems to be what has won the day here. There is no science behind this at all, it is a definition that is arbitrarily designed to permanently cap the number of planets to a small, manageable number. I see at least two big problems with this, first, the opening up to the possibility of there being hundreds of planets would force educators to rethink how these concepts are taught in elementary school. It could be something other than a memorize-these-nine-things exercise, but rather an opportunity to teach the basic concepts. Now it will just become a memorize-these-eight-things exercise. The other thing is that many of the newly discovered objects really merit study, and I don't see any Congress appropriating funds to send a probe to something that is just a "dwarf planet". The worst thing, though, is the pervasive media certainty that this is the end of the debate, five hundred really smart people voted on the issue and now it is settled for all eternity. I think that this new definition is deeply flawed in a number of ways, and we really need to treat this as a debate that has just begun.
Pluto is still Pluto.
Just because it has a number doesn't make the name go away. Let's see... I like the "Jelly Sandwich":
Mother Very Thoughtfully Makes A Jelly Sandwich Using No Persimmons... Everyday.
(Earnestly if you don't care for 'Terra')
(Jelly Sandwiches instead of A Jelly Standwich to keep the Asteroids out)
The Old Mill, the classic pre-Fantasia musical short from Disney, is available for $1.99. If you haven't seen it, it's worth the price of admission. If you have, you're probably heading over there to get it already and didn't bother reading this far. :)
in XP Pro go to Task Manager->Processes then right click on the process and choose "Set Affinity..."
:)
The good news is that Windows NT (from Windows 2000 at least) has had pretty good support for processor affinity.
The bad news is that you're almost always going to do better by letting the OS manage it, for any modern OS, and Windows is no exception.
Oh, wait, that's really good news too!
would you be able to somehow mitigate the their single-threaded nature by assigning the respective processes to it's own core?
First, pretty much any application on the Mac is multithreaded just because of the way the user interface works. Apple's OpenGL implementation is partly software, for example... this is why you can run hardware T&L on the Mac mini with its GMA950 GPU - the OS does that in software on the second core even in single-threaded games.
Second, OS X does a pretty goodjob of distributing applications to cores without having to explicitly bind them. Binding an application to a core would most likely slow it down... unless the program has been written to use a lot of fined grained shared state between threads... and what you're doing with processor affinity is *preventing* it from multiprocessing.
Processor affinity is like 64 bit. Unless you're doing something on the edge you probably don't need it, and if you need it you're probably already doing it.
Here's the summary:
The bad news is that OS X doesn't provide a hook for processor affinity. The good news is that Mach does support it, and you could use the Darwin sources to figure out how to implement it in OSX using direct Mach calls. The bad news is that it's really hard. The good news is you don't need to do it unless you're trying to prevent multiprocessing anyway.
Summary of the summary: Don't worry, be happy.
I have an old G3 iMac and use XCode to develop AppleScript Studio applications.
... the EE/CS undergrads all shared the one machine in the basement of Cory Hall.
I wouldn't expect that application programs in an interpreted language would do anything to cause panics.
I mean, at Finals Week at Berkeley in 1980 we had over 60 people at a time logged in and developing "C" code on a single PDP-11/70
What's most likely to cause panics is driver problems, file system problems (HFS+ isn't all that robust, I've had it blow chunks when I've left it running at less than 10% free space), and hardware problems. Application level stuff is really unlikely to panic the kernel.
What do you want it to do, make coffee for you?
Yep, iceCoffee is one of the extensions I use.
Apple doesn't fill the in between market like dell tries to do. You either buy a workstation or a light, friendly desktop.
Apple doesn't *sell* a regular desktop.
What you're calling the "in between" market is most of the market. Look at what people are actually buying at any computer store: all-in-one PCs flunk in the market, every time, and the only reason Apple can sell them at all is because they make it the only option if you want a Mac... and they're losing sales from everyone who doesn't want a Mac quite that badly as a result.
They need a Mac mini "Pro", and it can't cost much more than the current mini. But it seems like Jobs is having second thoughts on going back on his "no ugly monitors" statement.
1. It will be easier to configure and use.
.mac as podcasts posted to the iTMS.
How could it get any easier?
Nah, it'll have more integration with the iTunes Music Store.
It'll integrate some kind of Dashboard-like visualizer based plugin mechanism for controls that'll work better on Mac OS than Windows.
It'll have "video podcast" support and a way to let you drop your own files into the podcast tab to automatically share and stream them via
And a new DRM scheme to allow people to put restrictions on their Podcasts and vidcasts.
Most of the time, there is something equal or better out than the Apple product for a lower price.
Hardware-wise, pretty much all of the time.
Software-wise? Apple's been the leader pretty much since Jaguar came out.
Apple's a software company. Their hardware is an expensive "dongle" for their software.
You don't read slashdot for the comments?
I'll bet you read the articles in Playboy too!
It appears to be based on real world testing and not some crappy faux retro Hi-Fi panel designed by software engineers in a vacuum.
That's why the iPod interface is all but useless for adjusting by feel, and has to be locked when it's in your pocket becase just brushing it is enough to make it respond.
I gave my iPod Mini to my daughter and got the iPod Shuffle with its conventional d-Pad interface that's actually got signs of having been designed for *use* rather than *looks*.
No, what made me get an iPod at all instead of something else is that I can actually get accessories for it. And I expect I'll be able to get accessories for it in a year, and if I get a different iPod in a year the accessories I got now will still work.
I've got boxes of accessories that I'll never be able to use because nobody's every going to make a PDA compatible with the 1st generation iPaq, the Jornada, or the Visor, ever again. My old Nokia accessories don't work with my new Nokia phone. The charging cable for my old CLie doesn't work on my new one. But I can get an iPod now that will work with this docking cable from three years ago.
"Innovative" touch pad.. try scrolling through ten thousand songs precisely. Not. Happening.
I got an iPod Shuffle and gave my iPod Mini to my daughter precisely because the touch wheen is such a horrid user interface, and the D-pad on the shuffle is so much more practical for 99% of the uses.
Apple needs to get together with Sony and do an iPod that has a jog-wheel and a D-pad but is otherwise an iPod... syncs with iTunes, plays AAC and protected AAC files, and so on.
I just said that you need to provide additional efforts for synchronization for one end of the elevator. What is not clear about it?
Why you think you need to make extra efforts for synchronization of one end of the elevator. That's what's not clear, and that's what I'm trying to get across to you.
You started to talk about the usage of it. It does not matter.
The usage is exactly what does matter. You wouldn't ever expend energy to keep the elevator in an unstable orbit... you leave the elevator in a stable orbit. The centerpoint of the inter-orbital elevator is where freely-falling installations are synchronized to, not the end-points. Both the top end and bottom end are (from an energy standpoint) at the "bottom" of the elevator's "shaft". The "top of the shaft" is in the center.
This is, by the way, just like the "skyhook" style of elevator, except that they use an anchor mass to make the "top end" part of the structure shorter. For an inter-orbit elevator that's actually a disadvantage.
Vista addresses this is a very big way. [...]
MS's WinXP firewall DID work, [...]
None of this addresses the design problems... some of them at the application level... that have caused the problem.
I didn't say, for example, that the XP firewall didn't work. What I said was that the XP firewall is a single layer of security between the Internet and the service. In UNIX, a daemon is typically capable of being configured so it's bound to a specific interface... either directly or through tcpwrappers or xinetd... whether you have the firewall running or not. In Windows, the OS API allows this, but the services don't actually give you a way to specify this, so you need the firewall to get the same iind of isolation for services that UNIX has without a firewall.
Similarly, internal application firewalls are an attempt to work around a problem in application design, one created by MS with Internet Explorer, and depended on or emulated by others since. They don't keep an application from being compromised in the first place... they just make it harder for the application to do bad things once it is compromised.
This would be a second layer of protection in other operating systems: the integration between the browser and the shell that makes Windows so susceptible to spyware and web-spread viruses is vanishingly rare outside Windows.
So the problem I'm getting at is not that the Windows XP firewall doesn't work, or that Vista's firewall doesn't work, it's that it may well be the very best solution... to the wrong problem. You're better off using a non-IE-based browser and an external NATting router (which typically costs $20-40 and usually includes a wireless access point) and Windows 2000, than with Vista and Vista's firewall and IE.
I want a Mac mini Pro, iSlab, Macstation, or New Cube.
Whatever it's called, give it room for a 3.5" hard drive and a PCI-E video card.
I've found a plethora of cross-platform libraries, the problem has been that they only solve a small part of the problem (typically graphics or providing a POSIX API), don't use the native user interface, or require a specific implementation language. Most of them hit all three roadbumps, even before considering the quality of the API or the license.
That's why I end up using Tk. It's not a swiss-army-knife toolkit, but it's got bindings for multiple scripting languages and *they* have bindings to multiple compiled languages and portable APIs for networking and filesystem, and it has native look and feel. Plus it's got an agreeable license.
Everyone waited forever for a credible OS like OSX, and now Apple's hardware lineup has gone to middle of the road crap. Why?
Apple's hardware has almost always been middle of the road at best, from the Apple ][ or the original badly-underpowered Macintosh onwards. They've occasionally had a high-end machine that's competitive on the hardware side, but it's been more the exception than the rule.
What I want is for them to provide at least a competant middle-of-the-road box that doesn't lock me in. I hoped the Mac mini was a sign that Apple was shedding some of Jobs' obsession with style over everything. Yes, it was stylish, but at least he had to go back on the "no ugly monitors on nice Macs" promise. But they need a "Mini Pro" with a video card slot and room for a 3.5" hard drive. It wouldn't cost them any more to make than the Mini, and would bring an awful lot of fence-sitters into the fold.
Because 90% of iMac buyers (all-in-one consumer grade machine) will never want a second hard drive.
How many potential Mac buyers are remaining potential Mac buyers because Apple's not giving them a choice?
When I got my Mac mini it was the first consumer grade Mac since the cube that was even vaguely close to my sweet spot, and the new Mini is further away. You might ask "How many Mac mini buyers (compact entertainment machine) will ever want to upgrade the video card" just as easily... and you'd be missing the point. The point is that there's a lot of people who just can't find a Mac that can replace a PC, because Apple hasn't had a machine that targets the majority of PC buyers since they discontinued the Beige G3 desktop. Most people don't want a quad-core monster, but they don't want to be locked in either.
I'm sorry if you're frustrated. I know it's frustrating when someone seems to be doing that to me, but I honestly don't understand what you're getting at.
You wrote: Besides, it will cost you to synchronize two orbits with different R, because, you know, T~R^1.5
Can you explain what you mean by this in a bit more detail, because it sounds like you're making a wildly incorect assumption about inter-orbit elevators and how they're used... but I don't know what that assumption is... and obviously I'm doing a lousy job of guessing.
So, coudl you describe what you THINK I'm talking about here?
Surely there's a cross platform API that's better than the godawful bend-over-for-Microsoft Win32 API.
I've always written for the base UNIX API (which is available under Windows, and will be a native API in Vista if (as reported) Interix is included) with the GUI written using a scripting language and Tk. This produces apps that have a native user interface under UNIX, Windows, Mac OS X, and even (with care) old Mac OS. Surely I can't be the only one.
That is what I said: you have to put additional efforts to synchronize the ends of the elevator.
That happens automatically through what are commonly known as tidal forces.
The only effort that matters is the energy spent changing your velocity using a reaction drive. Since the low end moves below orbital velocity, you need less delta-v and thus less fuel and acceleration to dock with it from a lower orbit. Since the high end moves above orbital velocity, it takes less fuel and acceleration to dock with it from a higher orbit. The result is a net reduction in the energy required to get to space... to the point where an air-breathing hypersonic vehicle can do the job.
This means *less* effort overall, and *less* effort to dock with the elevator.
The point of inter-orbital elevators is to have the end-points moving above or below orbital velocity for that orbit.
Besides, it will cost you to synchronize two orbits with different R, because, you know, T~R^1.5
Not if you don't circularize the orbits first. If you're going to "land" on the low-end of an inter-orbit elevator it's a lot cheaper to just target it at apogee and grab it when you get there. The longer the elevator is, the more elliptical your intercept orbit can be, and the less fuel you have to use to synchronise with it.
And of course the opposite is true for the high end of the elevator and the perigee of your transfer orbit.
The most obvious application of this is an elevator you can reach from a hypersonic mostly-air-breathing craft on a ballistic arc out of the atmosphere.