...is that it's getting harder and harder to tell what's a put-on. More and more things I thought were parody at first turn out to be real.
Anyway, say what you will about Jobs, but he certainly got a LOT more done in his 20's and 30's than our current President, and by all accounts wasn't that much more of a jerk.
And Jobs's rescue of Apple certainly shows that he has an extraordinary ability to balance short and long term needs. Given what they've had to work with, technically, from Motorola for the last few years, can you believe that Apple is not only extant, but profitable?
Anyway, I'd probably vote for him over a fair number of other politicians. While we know a lot of his youthful indiscretions, I think that's just because he's been famous for so long. I imagine our previous two presidents were just as wild in their youth. The real question is how good a job who he is now could do, and I'd say the evidence is promising, or at least intriguing.
For all the "Jobs is a visionary" rhetoric, running a company on a knifes-edge like Apple has been for the last half decade implies a good ability to roll with the punches, and be flexible when appropriate.
Yes, the default license doesn't include the codecs, but there is an attachment that will give you access to them. The license is somewhat more restrictive. Among other things, there is a $0.25 fee per decoder for commercial products.
But, if you want to look at the source code for the codecs, its availalbe. Certainly not free software, but arguably open source.
https://www.helixcommunity.org/content/licenses# po rting
While there is legacy support for the 1995 RealAudio codecs, the modern stuff is really very good.
Anyone encoding music or soundtracks should be using the RealAudio Stereo Music 8 codecs. At lower data rates, this uses Real's in-house "Cook" codec, and at higher data rates a streaming-optimized version of Sony's ATRAC3 from miniDisc.
While Ogg is a fine format for download and CD-ROM type applications, today it isn't anywhere near mature as Real is for real-time streaming over lossy networks. Of course, with the sub-band stuff, Ogg could get a LOT better for this with further development.
I can see what you mean. Perhaps I'm overeducated, but that striked me as a very important part of the human condition. And they talk about things that matter - we're not in Waiting for Godot territory, here.
Still in the Mars trilogy
The bring down the darn beanpole! Have a revolution! Twice! A character dies while wearing a glide suit trying to save someone else! A scientist is captured, tortured, and saved with a daring rescue involving explosives. And the same scientist immolates the torture facility in a particularly clever and nasty way, years later. One of the moons of Mars gets shot off into space. Lots of sabotage. People suffocating as domed cities collapse. Heck, the whole thing kicks off with a murder!
That said, the characters certainly do talk a lot as well. I believe some of those scenes in Blue Mars were a concious reflection of the deliberation at the constitutional convention.
If you like the Classic SF stuff, I think the best modern linear extrapolation of that is Kim Stanley Robinson. His early stuff like Icehenge and A Short Sharp Shock was interesting, but self-conciously postmodern. But his stuff of the last decade is modern in the best way. It takes the best of the science-based and future history extrapolation of the Heinlien's and Asimov's, with a real curiosity and passion about human beings. The time I really realized how good he was was the third book of the "Three Californias" trilogy, "Pacific Edge." It was a largely utopian world - the biggest battle of the book was whether to build an enviormentally friendly business development on top of a hill with a good view. But the heart of the book was about the main character's loss of the woman who loved him to another man. As wonderful as the world was, a utopia can't prevent the real pain of life.
I've always felt that SF was fundamentally an exploration of the human condition, refracted through ideas of how humans would live in radically different times and worlds. Robinson fufills that promise better than any other I've read.
The one thing he isn't is funny, though. Humane, but not funny.
Robinson has the good habit of writing lots of short stories and single books, with two trilogies so far. None of those 10-volumes of 1000 pages each some authors can inflict on us.
For starts, the Mars trilogy is unbelievibly good (Red Mars, Green Mars, Blue Mars). It's about the first few centuries of human habitation of Mars, told from a variety of viewpoints. I just finished the excellent "Years of Rice and Salt," an alternate history assuming that the black death killed 99% of Europe, and how history would have been different through about 2100 or so. Both these books are striking both by their compassionate affection towards humanity and their characters, and the absolutely astounding amount of research that must have gone into them. Robinson, in the best possible way, makes writing great SF look HARD.
Anyone who lives in California should read "Three Californias" (The Wild Shore, Gold Coast, Pacific Edge), which posit three alternative futures for Orange County, California. Wonderful stuff.
The short story collections are also awesome, like Planet on the Table, and Remaking History. The short story "a sensitive dependence on initial conditions" is an absolute classic, analyzing many, many different possibilities of what could have happened if the Hiroshima bombing hadn't happened. My favorite of his experimental works.
As for AAC or Vorbis sounding better, it really depends on the AAC implementation and the data rate. There is a point of "good enough" where all codecs sound just fine. The interesting and hard area is sub 96 Kbps over lossy networks.
The manufacturer of the camera will have to pay $0.25 per unit, after the first 50,000, up to $1M. You don't have to pay anything,
If you make a site that is "renumerated" (like pay per view or subscriber), you owe $0.02 per hour, or $0.25 per subscriber per year (your choice), after 50,000 users per year, with a $1M cap per year per entity.
So, in general most users of MPEG-4 won't ever have to pay anything. Only big, commercial sites will have to pay anything for content.
XM Radio uses this, and are able to do nearly CD quality at 48 Kbps. By nearly, I mean casual listeners don't notice artifacts, although it certainly isn't mathematically identical to the original. Better than FM.
I don't have any public samples I can distribute, but I've heard it myself, and it sounds darn cool.
Essentially, AAC-SBR bolts on the Spectral Band Replication features of MP3 Pro, but uses the much more efficient AAC to encode the base frequencies instead of MP3. It's really pretty straightforward, with big payoffs.
I'd say MPEG-4 is Open, but not Free, in FSF terminology. Different strokes for different folks. MPEG-4 probably has more engineer-hours in it that the Linux kernel, and a lot of those companies wouldn't have participated if they hadn't thought they'd get money back on licensing terms. But yes, more flexible licenses would certainly help. We'll see what happens.
And I expect a lot of "black sheep" apps ala MP3, to exist for Linux. Check out MPEG4IP for a LAME-equivalent.
As for Theora, who knows? It isn't even in beta yet. It's VP3 based, and unless they enhance that code a LOT, it isn't going to be quality competitive with the best MPEG-4 implementations. But maybe they are enhancing it a lot.
Video codecs are a lot harder than audio codecs. And the new MPEG-4 audio codec (AAC-SBR) is a LOT better than Vorbis.
Well, you could say the same thing about Doom 3, or the Linux kernel, or pretty much any other big software project!
The key is finding the right balance of tools that can provide optimal quality while keeping the complexity required by a decoder as low as possible.
A video compression engineer once told me that designing codecs is 10% R&D, 10% Sun worship, and 80% alchemy.
R&D is all the PhD's thinking of new techniques, like wavelets, new motion estimation patters, etcetera.
Sun worship is spending a LOT of time doing test encodes on Sun workstations.
And alchemy is trying different combinations of tools in different orders, and seeing which work better in practice. It's by far the hardest and most time consuming element.
No, DivX Networks pays their licensing fee to MPEG-LA. If you write an app that uses their codec, you don't have to pay an additional fee for the video codec.
Computer playback is a relatively minor aspect of MPEG-4. MPEG-4 projects are in progress on integrating playback in everything from replacing the GSM codecs for audio transmission in cell phones, to HD DVD with red laser, to replacing MPEG-2 in set top boxes, to replacing Flash for interactive presentations.
MPEG-4 is really meant to replace MPEG-1 and MPEG-2, not QuickTime, Real, and Windows Media. Of course, given those open standards (with HIGHER licensing fees) are responsible for probably 98% of all digital video watched worldwide, that's the real game. MPEG-1 and MPEG-2 are used in VideoCD, DVD, digital cable, etcetera.
Windows Media 9 is incredibly good for computer-based authoring and playback, but is a Win32 only system right now. MPEG-4 already works on all kinds of devices.
MPEG-4 is open because all implementation details are public. You can get a copy of the standard, and build your encoder, decoder, server, etcetera based on it. No NDA's to sign or anything. You have to pay license fees in some cases if you distribute commerical products, but writing the software is something anyone can do.
This isn't true with Windows Media 9. While some details are avalable, not all are, and some are under restrictive licenses.
The licensing terms aren't that bad, and getting better for newer versions. For example, the forthcoming AVC MPEG-4 codec will be free to implement in all no-cost software. Even now, you get a pass on the first 50,000 distributed players. MPEG-4 is less difficult do deal with than MP3 licensing, and there are certainly lots of stuff in the Free Software community that can author and play back MP3 files!
MPEG-4 is open because full implementation details are public. While you certainly need to pay to do commerical products with MPEG-4, all details are available for implementation. This NOT true of Windows Media 9. There are nearly a dozen companies today competing to develop the best MPEG-4 encoder. But the only company that can produce the WM9 codec is Microsoft.
And Ogg Theora is still vaporware, with a public release not until this summer. It's based on VP3 and Vorbis, neither of which are as efficient as today's MPEG-4, let alone the next generation codecs like AVC and the AAC-SBR audio codec, both of which should be in products this year.
This probably isn't neccessary at this point. Apple has made it clear that they see MPEG-4 as the future, and so a port of QuickTime isn't really needed. Instead, good ISMA MPEG-4 decoders are needed for Linux. MPEG4IP is a great first start.
Indeed. Exactly the kind of thing to be covered in a firearm licensing class.
It's my expectation that most gun enthusiasts would handily pass this class, and they should. But it should help deal with folks who impulsively buy a gun for "protection" without know what they're doing.
Ah, you're one of those. Anyway, if I might extract the rational versions of a few of your arguments, and ignore the rest...
Well, either original intent OR the text is what's important, and neither imply an absolutist reading of the amendement. If we ignore intent, the amendement could easily be read as saying that the state can authorize some memebers of the national guard to carry some weapons, as the feds think is appropriate. You may feel there is a personal right to own big full auto firearms without any restrictions, but there is no constitutional right to that in the text, nor in case law.
And seriously, even in SE DC, how many cases a year are there of non-criminal civilians successfully defending themselves by firing a weapon? I'm sure it happens every now and then, but not often.
If you want an example of an ACTUAL tyrant, there really aren't any in any modern democratic societies - Nixon was about the closest we can as to a president trying to subvert democratic mechanisms, and he certainly wasn't a tyrant per se. The correlation between free speech and democracy is massive. The correlation between gun laws and democracy verges on non-existant. There are lots of democracies with lax gun laws, and lots of democracies with very stringent ones. Doesn't seem to change politics very much, but it certainly does the gun homicide rate.
It sounds like you're making the circular argument of "we need guns to keep the government from taking our guns away" which is more common than it is coherant.
You also seem to think I'm saying you can't have guns. You can have guns (well, if you're honestly that willing to shoot police offiers, than no, YOU can't have guns, but others who aren't willing to murder for politics can. But I assume that was a little free speech hyperbole on your part). I am saying there is a valid state interest in making sure those who carry guns outside of shooting ranges know what they're doing in some basic capacity.
Well, the second amendement didn't specify firearms either, just "arms." That's probably what they were thinking of, but they were probably thinking of swords as well, and certainly bayonets. Maybe grenades? Probably not cannons. Certainly not Class III firearms.
And while defeating tyrants was something some of them were thinking about, they were probably just as aware of the danger on the frontier. The odds of a civilian needing to fire a weapon in self defense were probably a couple orders of magnitude higher back then. Also, the US didn't have a substantial standing army, so getting a "well-regulated militia" was definitely a substantial goal of theirs, and the second amendement.
I won't invoke Godwin's Law.
But the important difference between speech and guns is the directness of the violence. Speech that directs others to kill is conspiracy, not speech, and illegal. Firing a gun at another human being is attempted murder. Saying society would be better off without someone is speech. Firing at a target is a hobby. The intent and the foreseeable consequences make a BIG difference.
A substantial number (hundreds?) of Americans so around every day parroting what Hilter said, to little consequence. If the same number of people deicded to copycat the DC area snipers, that' be one of the greatest tragedies in the history of this nation. Hilter was enormously worse than Malvo, but him having free speech wasn't the problem.
And as for tyrants, free speech has proven remarkably more effective in taking them down (i.e. Nixon). While guns have killed a fair number of politicians, their track record of killing the ones that we would have been better off without is essentially nil. Nor can I think of any cases where the threat of vilgilante murder against politicians has been a productive political force, but the threat of press criticism or exposure certainly has.
So, from a public utility perspective, in the political arena, speech is a lot more beneficial, with a lot fewer downsides, than firearms. This justifies a much different governmental stance for prior restraint.
I certainly would rather live in a nation with free speech and no handguns than a nation without free speech but with many available handguns. Lots of the former societies do just fine, but none of the latter do.
Slashdot Mirror
...is that it's getting harder and harder to tell what's a put-on. More and more things I thought were parody at first turn out to be real.
Anyway, say what you will about Jobs, but he certainly got a LOT more done in his 20's and 30's than our current President, and by all accounts wasn't that much more of a jerk.
And Jobs's rescue of Apple certainly shows that he has an extraordinary ability to balance short and long term needs. Given what they've had to work with, technically, from Motorola for the last few years, can you believe that Apple is not only extant, but profitable?
Anyway, I'd probably vote for him over a fair number of other politicians. While we know a lot of his youthful indiscretions, I think that's just because he's been famous for so long. I imagine our previous two presidents were just as wild in their youth. The real question is how good a job who he is now could do, and I'd say the evidence is promising, or at least intriguing.
For all the "Jobs is a visionary" rhetoric, running a company on a knifes-edge like Apple has been for the last half decade implies a good ability to roll with the punches, and be flexible when appropriate.
Folks,
Go to http://www.helixcommunity.org
Just started at 11am PST. They're taking live questions.
Actually, Microsoft is moving away from MMS and towards RTP/RTSP for Windows Media 9.
They both support H.263 video, MPEG-4, QuickTime, etcetera. Quite a lot of overlap.
Guys,
# po rting
Yes, the default license doesn't include the codecs, but there is an attachment that will give you access to them. The license is somewhat more restrictive. Among other things, there is a $0.25 fee per decoder for commercial products.
But, if you want to look at the source code for the codecs, its availalbe. Certainly not free software, but arguably open source.
https://www.helixcommunity.org/content/licenses
While there is legacy support for the 1995 RealAudio codecs, the modern stuff is really very good.
Anyone encoding music or soundtracks should be using the RealAudio Stereo Music 8 codecs. At lower data rates, this uses Real's in-house "Cook" codec, and at higher data rates a streaming-optimized version of Sony's ATRAC3 from miniDisc.
While Ogg is a fine format for download and CD-ROM type applications, today it isn't anywhere near mature as Real is for real-time streaming over lossy networks. Of course, with the sub-band stuff, Ogg could get a LOT better for this with further development.
I can see what you mean. Perhaps I'm overeducated, but that striked me as a very important part of the human condition. And they talk about things that matter - we're not in Waiting for Godot territory, here.
Still in the Mars trilogy
The bring down the darn beanpole! Have a revolution! Twice! A character dies while wearing a glide suit trying to save someone else! A scientist is captured, tortured, and saved with a daring rescue involving explosives. And the same scientist immolates the torture facility in a particularly clever and nasty way, years later. One of the moons of Mars gets shot off into space. Lots of sabotage. People suffocating as domed cities collapse. Heck, the whole thing kicks off with a murder!
That said, the characters certainly do talk a lot as well. I believe some of those scenes in Blue Mars were a concious reflection of the deliberation at the constitutional convention.
Not to mention the GIRLS who read Gor books!
Intriguing. I think you need to give some specific examples!
If you like the Classic SF stuff, I think the best modern linear extrapolation of that is Kim Stanley Robinson. His early stuff like Icehenge and A Short Sharp Shock was interesting, but self-conciously postmodern. But his stuff of the last decade is modern in the best way. It takes the best of the science-based and future history extrapolation of the Heinlien's and Asimov's, with a real curiosity and passion about human beings. The time I really realized how good he was was the third book of the "Three Californias" trilogy, "Pacific Edge." It was a largely utopian world - the biggest battle of the book was whether to build an enviormentally friendly business development on top of a hill with a good view. But the heart of the book was about the main character's loss of the woman who loved him to another man. As wonderful as the world was, a utopia can't prevent the real pain of life.
I've always felt that SF was fundamentally an exploration of the human condition, refracted through ideas of how humans would live in radically different times and worlds. Robinson fufills that promise better than any other I've read.
The one thing he isn't is funny, though. Humane, but not funny.
Robinson has the good habit of writing lots of short stories and single books, with two trilogies so far. None of those 10-volumes of 1000 pages each some authors can inflict on us.
For starts, the Mars trilogy is unbelievibly good (Red Mars, Green Mars, Blue Mars). It's about the first few centuries of human habitation of Mars, told from a variety of viewpoints. I just finished the excellent "Years of Rice and Salt," an alternate history assuming that the black death killed 99% of Europe, and how history would have been different through about 2100 or so. Both these books are striking both by their compassionate affection towards humanity and their characters, and the absolutely astounding amount of research that must have gone into them. Robinson, in the best possible way, makes writing great SF look HARD.
Anyone who lives in California should read "Three Californias" (The Wild Shore, Gold Coast, Pacific Edge), which posit three alternative futures for Orange County, California. Wonderful stuff.
The short story collections are also awesome, like Planet on the Table, and Remaking History. The short story "a sensitive dependence on initial conditions" is an absolute classic, analyzing many, many different possibilities of what could have happened if the Hiroshima bombing hadn't happened. My favorite of his experimental works.
Well, that's a pretty big "just"
As for AAC or Vorbis sounding better, it really depends on the AAC implementation and the data rate. There is a point of "good enough" where all codecs sound just fine. The interesting and hard area is sub 96 Kbps over lossy networks.
The manufacturer of the camera will have to pay $0.25 per unit, after the first 50,000, up to $1M. You don't have to pay anything,
If you make a site that is "renumerated" (like pay per view or subscriber), you owe $0.02 per hour, or $0.25 per subscriber per year (your choice), after 50,000 users per year, with a $1M cap per year per entity.
So, in general most users of MPEG-4 won't ever have to pay anything. Only big, commercial sites will have to pay anything for content.
I meant that MPEG-1 and MPEG-2 are 98%, and that QuickTime, Windows Media, and RealMedia at best share 2%.
What do you want?
XM Radio uses this, and are able to do nearly CD quality at 48 Kbps. By nearly, I mean casual listeners don't notice artifacts, although it certainly isn't mathematically identical to the original. Better than FM.
I don't have any public samples I can distribute, but I've heard it myself, and it sounds darn cool.
Essentially, AAC-SBR bolts on the Spectral Band Replication features of MP3 Pro, but uses the much more efficient AAC to encode the base frequencies instead of MP3. It's really pretty straightforward, with big payoffs.
I'd say MPEG-4 is Open, but not Free, in FSF terminology. Different strokes for different folks. MPEG-4 probably has more engineer-hours in it that the Linux kernel, and a lot of those companies wouldn't have participated if they hadn't thought they'd get money back on licensing terms. But yes, more flexible licenses would certainly help. We'll see what happens.
And I expect a lot of "black sheep" apps ala MP3, to exist for Linux. Check out MPEG4IP for a LAME-equivalent.
As for Theora, who knows? It isn't even in beta yet. It's VP3 based, and unless they enhance that code a LOT, it isn't going to be quality competitive with the best MPEG-4 implementations. But maybe they are enhancing it a lot.
Video codecs are a lot harder than audio codecs. And the new MPEG-4 audio codec (AAC-SBR) is a LOT better than Vorbis.
Well, you could say the same thing about Doom 3, or the Linux kernel, or pretty much any other big software project!
The key is finding the right balance of tools that can provide optimal quality while keeping the complexity required by a decoder as low as possible.
A video compression engineer once told me that designing codecs is 10% R&D, 10% Sun worship, and 80% alchemy.
R&D is all the PhD's thinking of new techniques, like wavelets, new motion estimation patters, etcetera.
Sun worship is spending a LOT of time doing test encodes on Sun workstations.
And alchemy is trying different combinations of tools in different orders, and seeing which work better in practice. It's by far the hardest and most time consuming element.
No, DivX Networks pays their licensing fee to MPEG-LA. If you write an app that uses their codec, you don't have to pay an additional fee for the video codec.
Computer playback is a relatively minor aspect of MPEG-4. MPEG-4 projects are in progress on integrating playback in everything from replacing the GSM codecs for audio transmission in cell phones, to HD DVD with red laser, to replacing MPEG-2 in set top boxes, to replacing Flash for interactive presentations.
MPEG-4 is really meant to replace MPEG-1 and MPEG-2, not QuickTime, Real, and Windows Media. Of course, given those open standards (with HIGHER licensing fees) are responsible for probably 98% of all digital video watched worldwide, that's the real game. MPEG-1 and MPEG-2 are used in VideoCD, DVD, digital cable, etcetera.
Windows Media 9 is incredibly good for computer-based authoring and playback, but is a Win32 only system right now. MPEG-4 already works on all kinds of devices.
No.
MPEG-4 is open because all implementation details are public. You can get a copy of the standard, and build your encoder, decoder, server, etcetera based on it. No NDA's to sign or anything. You have to pay license fees in some cases if you distribute commerical products, but writing the software is something anyone can do.
This isn't true with Windows Media 9. While some details are avalable, not all are, and some are under restrictive licenses.
The licensing terms aren't that bad, and getting better for newer versions. For example, the forthcoming AVC MPEG-4 codec will be free to implement in all no-cost software. Even now, you get a pass on the first 50,000 distributed players. MPEG-4 is less difficult do deal with than MP3 licensing, and there are certainly lots of stuff in the Free Software community that can author and play back MP3 files!
MPEG-4 is open because full implementation details are public. While you certainly need to pay to do commerical products with MPEG-4, all details are available for implementation. This NOT true of Windows Media 9. There are nearly a dozen companies today competing to develop the best MPEG-4 encoder. But the only company that can produce the WM9 codec is Microsoft.
And Ogg Theora is still vaporware, with a public release not until this summer. It's based on VP3 and Vorbis, neither of which are as efficient as today's MPEG-4, let alone the next generation codecs like AVC and the AAC-SBR audio codec, both of which should be in products this year.
This probably isn't neccessary at this point. Apple has made it clear that they see MPEG-4 as the future, and so a port of QuickTime isn't really needed. Instead, good ISMA MPEG-4 decoders are needed for Linux. MPEG4IP is a great first start.
Indeed. Exactly the kind of thing to be covered in a firearm licensing class.
It's my expectation that most gun enthusiasts would handily pass this class, and they should. But it should help deal with folks who impulsively buy a gun for "protection" without know what they're doing.
Basically, it should cover:
Safe gun handling
Gun security
Basic marksmanship.
Due to the latter, it should be fine to use a gun on a firing range when unlicensed.
It's fine to leave it there if you're in the bedroom, and it's in a place any kids around couldn't access while you're sleeping.
It's not okay to leave it there, loaded, when you go to work.
Ah, you're one of those. Anyway, if I might extract the rational versions of a few of your arguments, and ignore the rest...
Well, either original intent OR the text is what's important, and neither imply an absolutist reading of the amendement. If we ignore intent, the amendement could easily be read as saying that the state can authorize some memebers of the national guard to carry some weapons, as the feds think is appropriate. You may feel there is a personal right to own big full auto firearms without any restrictions, but there is no constitutional right to that in the text, nor in case law.
And seriously, even in SE DC, how many cases a year are there of non-criminal civilians successfully defending themselves by firing a weapon? I'm sure it happens every now and then, but not often.
If you want an example of an ACTUAL tyrant, there really aren't any in any modern democratic societies - Nixon was about the closest we can as to a president trying to subvert democratic mechanisms, and he certainly wasn't a tyrant per se. The correlation between free speech and democracy is massive. The correlation between gun laws and democracy verges on non-existant. There are lots of democracies with lax gun laws, and lots of democracies with very stringent ones. Doesn't seem to change politics very much, but it certainly does the gun homicide rate.
It sounds like you're making the circular argument of "we need guns to keep the government from taking our guns away" which is more common than it is coherant.
You also seem to think I'm saying you can't have guns. You can have guns (well, if you're honestly that willing to shoot police offiers, than no, YOU can't have guns, but others who aren't willing to murder for politics can. But I assume that was a little free speech hyperbole on your part). I am saying there is a valid state interest in making sure those who carry guns outside of shooting ranges know what they're doing in some basic capacity.
Well, the second amendement didn't specify firearms either, just "arms." That's probably what they were thinking of, but they were probably thinking of swords as well, and certainly bayonets. Maybe grenades? Probably not cannons. Certainly not Class III firearms.
And while defeating tyrants was something some of them were thinking about, they were probably just as aware of the danger on the frontier. The odds of a civilian needing to fire a weapon in self defense were probably a couple orders of magnitude higher back then. Also, the US didn't have a substantial standing army, so getting a "well-regulated militia" was definitely a substantial goal of theirs, and the second amendement.
I won't invoke Godwin's Law.
But the important difference between speech and guns is the directness of the violence. Speech that directs others to kill is conspiracy, not speech, and illegal. Firing a gun at another human being is attempted murder. Saying society would be better off without someone is speech. Firing at a target is a hobby. The intent and the foreseeable consequences make a BIG difference.
A substantial number (hundreds?) of Americans so around every day parroting what Hilter said, to little consequence. If the same number of people deicded to copycat the DC area snipers, that' be one of the greatest tragedies in the history of this nation. Hilter was enormously worse than Malvo, but him having free speech wasn't the problem.
And as for tyrants, free speech has proven remarkably more effective in taking them down (i.e. Nixon). While guns have killed a fair number of politicians, their track record of killing the ones that we would have been better off without is essentially nil. Nor can I think of any cases where the threat of vilgilante murder against politicians has been a productive political force, but the threat of press criticism or exposure certainly has.
So, from a public utility perspective, in the political arena, speech is a lot more beneficial, with a lot fewer downsides, than firearms. This justifies a much different governmental stance for prior restraint.
I certainly would rather live in a nation with free speech and no handguns than a nation without free speech but with many available handguns. Lots of the former societies do just fine, but none of the latter do.