Proper patching, Firefox, NAV, and Ad-Aware. Web mail. WinSCP for file copying to remote shells. A hardware firewall to block unnecessary services from my internal network. I have no security issues on Windows. And Linux won't even install on my personal box.
Stop thinking Windows users are idiots. Properly secured, a Windows client is perfectly acceptable. If I could only train the people who used the machines I set up to stop running stupid things, life would be good, but if I put them on a Linux box, they'd find a way to run a rm -rf * script somehow.
BlenderCAD? You're seriously suggesting that? Qcad is at least a functional CAD program, but it doesn't do what I need. COordinate GeOmetry. COGO. No COGO, no Linux.
The real question is, what trust should be placed in someone who can't type the S key properly?
An emac is not comparable to what I'm talking about, but that's okay.
I own a G3 powerbook. It works for me, because all the apps I want to use it with were ones I didn't own on the PC anyway, so buying them didn't cost me 'extra'. If I'd owned a PC version, it would have cost me 'extra' compared to just upgrading my hardware.
Now stop being a fucking fanboy. The same argument works against switching from a Mac to something else. Switching is expensive. That *is* my case. Not that the Mac is more expensive than the PC. That switching is more expensive than upgrading.
No, the point is that if you could reliably add the bow noise back in later with no change in sound, would you want to make it and record it with the bow noise included, or would you want to not make the bow noise and record it clean, with the ability to add it in later? I've run into this with musicians, especially guitarists - they set up their stack of effects and say "This is how it needs to be recorded". Since I like recording, I just do it eventually, but I always try first with "try it clean, we can run the clean sound through the chain later to add in your effects, but starting clean gives us more options". Sometimes they listen, and they're rarely sorry.
Bow noise is somewhat different, in that its more about *playing* than the output. A closer analogy would be the choice of microphone used if you were miked for a performance - would you want a mike that changes your timbre, or would you want a neutral mike? Obviously, the neutral mike would be preferable - coloration and alteration of sound should *always* be intentional. Amps are not equivalent to bowing style, to strings or to pick. The purpose of the amp should be to make it *louder* without changing timbre, tempo, or any other quality than volume. If you want a particular tonal quality to come through the amp, it should be provided by something that is intended to make that tonal change. A box that changes tone is not an amplifier; it's a sound processor.
Further, audiophiles are rarely musicians; their choice is a lie, because they talk about 'transparency' while seeking a non-transparent sound. They talk about being 'true to the original' while seeking to change the original sound. Changing the sound to be pleasing to your ears is fine, but don't try to justify it by saying you're trying to get as close to the original as possible.
CAD link 1, and PCB links 2 and 3 are for commercial software on Linux. I know that exists. I was saying there aren't Free equivalents.
Checking through your links, there's still no Linux COGO at all. I figure there's probably a commercial package somewhere, but I don't know of it (then again, I don't do COGO full time, I just support a civil engineer's business, who does a lot of COGO).
Mechanical and electrical CAD is still often done on UNIX workstations, yes, but most of the civil engineers I know work entirely on Windows packages. Electricals can, and more and more are, working entirely in Windows. And mechanicals who don't need Pro/E style capabilities, who are just parts drafting in AutoCAD, often work entirely in Windows.
Gentoo tells me its a kernel panic of some variety. Mandrake doesn't tell me anything, it just locks up.
From what I can tell, its some sort of weird interaction between the Highpoint HPT372 RAID controller and the fact that I have 1GB of memory. Maybe. Maybe it's something else. I really don't know.
No, truly good musicians understand the following:
Perfect fidelity is GOOD, because you can *always* reintroduce the distortion in some other fashion if you decide you want it, but you can't remove it if you don't want it.
You can add a tube processor if you like the tube sound, and the processor+clean solid state amp will sound tubelike, but if you have a tube amp you will NEVER be able to remove the tube sound.
Then again, I blame that fact on Linux not being able to install on that machine. Any distro I've tried. Mandrake, Gentoo, and SuSE all lock up during install. It's strange.
Ditch your Windows box to buy a Mac. ($1500, let's say, for a decent Mac to use - and I'm being generous here)
Now run all your Mac programs. Oh, wait, you have to buy all new Mac programs? That's a new copy of Photoshop ($500), a new copy of Office ($400), plus whatever else you might need. Oh, and let's not forget buying VPC ($100, I think?)
So we're talking approaching $3000 for *the same functionality* as a Windows box, and marginally better security than a well-patched and secured Windows box. And this of course assumes that all your apps are available on the Mac; if you have something you have to run under VPC, hope it doesn't need to run all that fast.
I like Macs. I own one. But there are real financial considerations to switching from a PC to Mac environment, and those tend to rule over security considerations for most of the world.
It isn't always an issue of an equivalent app existing.
I know there are probably UNIXable equivalents out there to, for example, Eagle Point, a coordinate geometry solution for civil engineers using CAD. However, the small business I do occasional support for has a copy of a Windows license of Eagle Point, and they can't *afford* to switch. Hell, they're still running 98 because the version they own won't run on anything newer, and they can't *afford* to upgrade.
You want an office suite, email, and the web? Sure, there are free Linux apps for all that. And they're totally usable and fine.
For the rest of us, who work with specialized applications that probably cost as much as (or more than) we make in a month, WE CAN'T AFFORD TO GO BUY THE PROPRIETARY LINUX SOLUTION. We're still paying off the proprietary Windows solution.
Note to self - the ones that spin that fast are carbon fiber, not concrete. Concrete is used for slow high mass rotors, not for small high speed rotors.
Do you have any idea how much rotational energy they store into one of those? You're talking about a concrete disc spinning at something like 50-100,000 RPM. Stored rotational energy goes up as the square of the speed of rotation, so you can have a LOT of energy in a chunk of concrete.
Let's do the math. You take a 1 ft diameter 50 lb flywheel spinning at 100,000 RPM - these do exist (read up on Jack Bitterly). That's KE = 1/2 I omega^2, which for a circular disk of uniform density works out to:
I = 1/2 m r^2, or.5*22.67kg*(.1524m)^2 I = 0.2632 kg(m^2) w = 2*pi*(100000/(60)) rads/sec KE = 1/2 I w^2 KE = 14431554.9 Joules
Now, let's assume that the whole chunk of mass disintegrates and each piece gets an equal bit of the energy. A 1 lb chunk receives 1/50th the energy.
KEchunk = 288631 Joules KE =.5m*v^2
Initial velocity of the chunk will be v=sqrt(2*KE/m) v = 1132.6 m/s
Launched straight up, this chunk would go (neglecting air friction) up to KEinit=PEfinal, or PE = mgh h = PE/mg h = 288631/(9.8*.45) = 65.449km = 40.66 miles
It's going A KILOMETER PER SECOND at disintegration. And that's assuming an average piece - a piece towards the edge is actually going to have more of the KE than a piece towards the center.
No, they billed that many hours. You'd be surprised how that kinda shit works.
For instance, I once worked a job for a week. I *worked* a total of about 24 hours. I *billed* nearly 60, because of the way they agreed to charging hours (minimum callouts, etc.)
Nextel would be an expensive acquisition in an industry that's already in the middle of a cutthroat price war.
Remember, Nextel has (by many measures) the *best* customer base of any of the providers - they have the highest revenue per subscriber of any carrier, and tend towards more business customers (generally more stable than consumers). They might prove to be more expensive a company to acquire than the revenue warrants, though.
That said - were someone to purchase them, the only difference to the Nextel customers, at least at first, would be the name on the bill. My comment re: transition plans would go ahead in exactly the same manner if they get bought.
First off, you know GSM (original, not the 3G versions) is a type of TDMA standard, right? And that the 3GSM standard is generelly implemented as a CDMA?
Verizon/Sprint CDMA is actually called IS-95.
The TDMA you're thinking of (AT&T's) is actually IS-136, previously IS-54.
Nextel uses a modified standard (based on TDMA methods, but not compatible with IS-136/IS-54, which also run in the 800-900 MHz band). TDMA isn't getting shafted because it runs higher in the 800 MHz band (closer to 850), further away from emergency frequencies.
Nextel's plans, last I heard, involved overlaying a 3G network onto their existing spectrum. This will obviously be easier to do now that they've got new spectrum to work with. I would bet that it'll be like the AT&T TDMA-GSM transition - a long period of overlap, followed by a slow phaseout of the last few TDMA customers as their phones die.
They *bought* the patents in 97. This implies they bought pre-existing patents in 97. Odds are the patents were *granted* prior to 97, and quite likely prior to 94. Since prior art only applies to the grant of a patent, not a purchase, prior art isn't applicable (unless the patents are post 94.)
If you think this, the real question becomes: was the cost of the design appropriate for a 90 day rover, a 2 year rover, or is it indeterminable?
I'm sure they did make a 'safe' estimate, but I severely doubt that every single engineer at NASA involved in the program was willing to stay silent about lieing to the public. They're normal people, you know. We'd have heard something.
Slashdot Mirror
Proper patching, Firefox, NAV, and Ad-Aware. Web mail. WinSCP for file copying to remote shells. A hardware firewall to block unnecessary services from my internal network. I have no security issues on Windows. And Linux won't even install on my personal box.
Stop thinking Windows users are idiots. Properly secured, a Windows client is perfectly acceptable. If I could only train the people who used the machines I set up to stop running stupid things, life would be good, but if I put them on a Linux box, they'd find a way to run a rm -rf * script somehow.
BlenderCAD? You're seriously suggesting that? Qcad is at least a functional CAD program, but it doesn't do what I need. COordinate GeOmetry. COGO. No COGO, no Linux.
The real question is, what trust should be placed in someone who can't type the S key properly?
An emac is not comparable to what I'm talking about, but that's okay.
I own a G3 powerbook. It works for me, because all the apps I want to use it with were ones I didn't own on the PC anyway, so buying them didn't cost me 'extra'. If I'd owned a PC version, it would have cost me 'extra' compared to just upgrading my hardware.
Now stop being a fucking fanboy. The same argument works against switching from a Mac to something else. Switching is expensive. That *is* my case. Not that the Mac is more expensive than the PC. That switching is more expensive than upgrading.
No, the point is that if you could reliably add the bow noise back in later with no change in sound, would you want to make it and record it with the bow noise included, or would you want to not make the bow noise and record it clean, with the ability to add it in later? I've run into this with musicians, especially guitarists - they set up their stack of effects and say "This is how it needs to be recorded". Since I like recording, I just do it eventually, but I always try first with "try it clean, we can run the clean sound through the chain later to add in your effects, but starting clean gives us more options". Sometimes they listen, and they're rarely sorry.
Bow noise is somewhat different, in that its more about *playing* than the output. A closer analogy would be the choice of microphone used if you were miked for a performance - would you want a mike that changes your timbre, or would you want a neutral mike? Obviously, the neutral mike would be preferable - coloration and alteration of sound should *always* be intentional. Amps are not equivalent to bowing style, to strings or to pick. The purpose of the amp should be to make it *louder* without changing timbre, tempo, or any other quality than volume. If you want a particular tonal quality to come through the amp, it should be provided by something that is intended to make that tonal change. A box that changes tone is not an amplifier; it's a sound processor.
Further, audiophiles are rarely musicians; their choice is a lie, because they talk about 'transparency' while seeking a non-transparent sound. They talk about being 'true to the original' while seeking to change the original sound. Changing the sound to be pleasing to your ears is fine, but don't try to justify it by saying you're trying to get as close to the original as possible.
Totally unsuitable for my uses, but it looks like a cool program for basic 2D CAD.
I can't use it because there's no COGO integration and there's no 3D.
CAD link 1, and PCB links 2 and 3 are for commercial software on Linux. I know that exists. I was saying there aren't Free equivalents.
Checking through your links, there's still no Linux COGO at all. I figure there's probably a commercial package somewhere, but I don't know of it (then again, I don't do COGO full time, I just support a civil engineer's business, who does a lot of COGO).
A good example of current Windows COGO stuff is Eagle Point Software's packages.
Mechanical and electrical CAD is still often done on UNIX workstations, yes, but most of the civil engineers I know work entirely on Windows packages. Electricals can, and more and more are, working entirely in Windows. And mechanicals who don't need Pro/E style capabilities, who are just parts drafting in AutoCAD, often work entirely in Windows.
I'd love to attempt to be Free.
But unfortunately, my rent is not free.
Until Free can pay my rent, I can't be Free.
That is not a dumb concept.
Qcad is 2D. For the CAD apps I need to do/support, we need 3D - the real world is 3D, there's this thing called "elevation" when you do site surveys.
COGO - coordinate geometry. Used by civil engineers and surveyors, mainly. Almost totally useless without CAD integration.
It's pretty obvious you've never worked with the full Protel or Mentor setups, if you think EagleCAD is good...
Yeah.
Gentoo tells me its a kernel panic of some variety. Mandrake doesn't tell me anything, it just locks up.
From what I can tell, its some sort of weird interaction between the Highpoint HPT372 RAID controller and the fact that I have 1GB of memory. Maybe. Maybe it's something else. I really don't know.
No, truly good musicians understand the following:
Perfect fidelity is GOOD, because you can *always* reintroduce the distortion in some other fashion if you decide you want it, but you can't remove it if you don't want it.
You can add a tube processor if you like the tube sound, and the processor+clean solid state amp will sound tubelike, but if you have a tube amp you will NEVER be able to remove the tube sound.
Not on my machine...
Then again, I blame that fact on Linux not being able to install on that machine. Any distro I've tried. Mandrake, Gentoo, and SuSE all lock up during install. It's strange.
So, where's the Free CAD drafting program?
Free COGO suite to work with?
Hey, I know - a good Free PCB design toolchain!
Wait... do these exist? If they do, are they even remotely comparable to the proprietary equivalents?
For some of us, its about getting something useful done. Go be Free; I'm going to do something useful.
Yeah, that's a great solution.
Ditch your Windows box to buy a Mac. ($1500, let's say, for a decent Mac to use - and I'm being generous here)
Now run all your Mac programs. Oh, wait, you have to buy all new Mac programs? That's a new copy of Photoshop ($500), a new copy of Office ($400), plus whatever else you might need. Oh, and let's not forget buying VPC ($100, I think?)
So we're talking approaching $3000 for *the same functionality* as a Windows box, and marginally better security than a well-patched and secured Windows box. And this of course assumes that all your apps are available on the Mac; if you have something you have to run under VPC, hope it doesn't need to run all that fast.
I like Macs. I own one. But there are real financial considerations to switching from a PC to Mac environment, and those tend to rule over security considerations for most of the world.
It isn't always an issue of an equivalent app existing.
I know there are probably UNIXable equivalents out there to, for example, Eagle Point, a coordinate geometry solution for civil engineers using CAD. However, the small business I do occasional support for has a copy of a Windows license of Eagle Point, and they can't *afford* to switch. Hell, they're still running 98 because the version they own won't run on anything newer, and they can't *afford* to upgrade.
You want an office suite, email, and the web? Sure, there are free Linux apps for all that. And they're totally usable and fine.
For the rest of us, who work with specialized applications that probably cost as much as (or more than) we make in a month, WE CAN'T AFFORD TO GO BUY THE PROPRIETARY LINUX SOLUTION. We're still paying off the proprietary Windows solution.
Magnetism: Check the work NASA did with ProSEDS, an electrodynamic tether experiment in LEO. It has been worked on.
Note to self - the ones that spin that fast are carbon fiber, not concrete. Concrete is used for slow high mass rotors, not for small high speed rotors.
Do you have any idea how much rotational energy they store into one of those? You're talking about a concrete disc spinning at something like 50-100,000 RPM. Stored rotational energy goes up as the square of the speed of rotation, so you can have a LOT of energy in a chunk of concrete.
.5*22.67kg*(.1524m)^2
.5m*v^2
Let's do the math. You take a 1 ft diameter 50 lb flywheel spinning at 100,000 RPM - these do exist (read up on Jack Bitterly). That's KE = 1/2 I omega^2, which for a circular disk of uniform density works out to:
I = 1/2 m r^2, or
I = 0.2632 kg(m^2)
w = 2*pi*(100000/(60)) rads/sec
KE = 1/2 I w^2
KE = 14431554.9 Joules
Now, let's assume that the whole chunk of mass disintegrates and each piece gets an equal bit of the energy. A 1 lb chunk receives 1/50th the energy.
KEchunk = 288631 Joules
KE =
Initial velocity of the chunk will be
v=sqrt(2*KE/m)
v = 1132.6 m/s
Launched straight up, this chunk would go (neglecting air friction) up to KEinit=PEfinal, or
PE = mgh
h = PE/mg
h = 288631/(9.8*.45) = 65.449km = 40.66 miles
It's going A KILOMETER PER SECOND at disintegration. And that's assuming an average piece - a piece towards the edge is actually going to have more of the KE than a piece towards the center.
40 miles isn't so farfetched now, is it?
No, they billed that many hours. You'd be surprised how that kinda shit works.
For instance, I once worked a job for a week. I *worked* a total of about 24 hours. I *billed* nearly 60, because of the way they agreed to charging hours (minimum callouts, etc.)
Point 3:
PC laptops won't let you modify most of that crap either, except by going to a different base model.
Billable rarely if ever equates to "paycheck per hour". For one thing, no one ever works 40 billable hours in a 40 hour workweek.
Nextel would be an expensive acquisition in an industry that's already in the middle of a cutthroat price war.
Remember, Nextel has (by many measures) the *best* customer base of any of the providers - they have the highest revenue per subscriber of any carrier, and tend towards more business customers (generally more stable than consumers). They might prove to be more expensive a company to acquire than the revenue warrants, though.
That said - were someone to purchase them, the only difference to the Nextel customers, at least at first, would be the name on the bill. My comment re: transition plans would go ahead in exactly the same manner if they get bought.
Right, because Nextel didn't pay anything for their frequencies...
Of course Nextel deserves compensation for a frequency they licensed that they're being forced out of. How the hell can you argue against that?
First off, you know GSM (original, not the 3G versions) is a type of TDMA standard, right? And that the 3GSM standard is generelly implemented as a CDMA?
Verizon/Sprint CDMA is actually called IS-95.
The TDMA you're thinking of (AT&T's) is actually IS-136, previously IS-54.
Nextel uses a modified standard (based on TDMA methods, but not compatible with IS-136/IS-54, which also run in the 800-900 MHz band). TDMA isn't getting shafted because it runs higher in the 800 MHz band (closer to 850), further away from emergency frequencies.
Nextel's plans, last I heard, involved overlaying a 3G network onto their existing spectrum. This will obviously be easier to do now that they've got new spectrum to work with. I would bet that it'll be like the AT&T TDMA-GSM transition - a long period of overlap, followed by a slow phaseout of the last few TDMA customers as their phones die.
Jesus, not interesting at all.
They *bought* the patents in 97. This implies they bought pre-existing patents in 97. Odds are the patents were *granted* prior to 97, and quite likely prior to 94. Since prior art only applies to the grant of a patent, not a purchase, prior art isn't applicable (unless the patents are post 94.)
If you think this, the real question becomes: was the cost of the design appropriate for a 90 day rover, a 2 year rover, or is it indeterminable?
I'm sure they did make a 'safe' estimate, but I severely doubt that every single engineer at NASA involved in the program was willing to stay silent about lieing to the public. They're normal people, you know. We'd have heard something.