Well, I certainly won't claim that I am absolutely right. However, from the GPL FAQ:
If I distribute GPL'd software for a fee, am I required to also make it available to the public without a charge?
No. However, if someone pays your fee and gets a copy, the GPL gives them the freedom to release it to the public, with or without a fee. For example, someone could pay your fee, and then put her copy on a web site for the general public.
That seems pretty clear, but I couldn't have told you that from reading the actual license.
With my limited understanding of the GPL (IANAL), I believe that a vendor can charge what he/she wants for the binary version of the software, and is then obligated to provide the source (to that same person) at a nominal additional cost. The GPL only requires providing the source to those parties that receive the binaries. The customer, however, has no obligation (under the GPL) to keep the source code secret, and can redistribute the code at any cost (including none) to other parties. If MV adds an additional license to the GPL'd source that prevents redistribution, *that* would be a violation of the GPL.
So where are all these MV customers and why haven't they provided the MV kernel source for the rest of us?
According to this article (as well as many others via google):
Assembly, test, launch, and a year of operations of each rover cost about $425 million, or about the same amount of money as it cost to make the movies Titanic (1997) and Pearl Harbor (2001). This amount was also equivalent to what it costs to launch a single space-shuttle mission...
This sort of space exploration is realively cheap, considering the payoffs involved. I suggest we knock off a couple "bridges to nowhere" from our budget, or ask for some money back from the Big Dig, instead.
This is a good point. The total probablity of mission success would be the wieghted sum of all possible events minus thier mutual probablity. To get through the first few events (transit to orbit and landing), the amount of over-engineering required could easily render the probabilty of failure later on negligible.
P(failure)=P(crash) +P(land){P(!survive landing)+P(all other failure events after landing)-overlap}
It's not clear if the tech in the article is a simple beamformer-type array, or something closer to an acoustic bullet array. Because of the size of the individual array elements (1 m^2),I would assume that the array is designed for low frequency operation ( 20Hz). It would be hard to beamform at such a low frequency (due to large wavelengths -- 20 Hz translates to around 19 meter spacing for a lambda/4 array). Acoustic bullet research focuses on the radiator element and how to efficiently launch high-energy acoustic waves into the medium (water in this case).
And if as has been suggested most PPCs go into embedded systems, Yellow Dog is totally unsuitable. There are plenty of embedded solutions for Linux out there and I expect they all work quite well already.
YDL is a handy development platform for embedded PPC targets (at least for PPC 750s and 74xxs). I'm not sure about the other embedded PPC processors.
Of course, speaking of the Boomers, I fear that my generation (I'm 28) might be one of those unlucky historical examples of one which didn't get to do jack shit because they were so busy catering to the needs of their wealthy elders while trying to patch up the disasterous debts they left us. By the time they start to croak en masse it'll be too late to do anything all that interesting -- we'll be too old and too unimaginative, left only with the shadow of the dreams we once entertained.
I used to think the same thing (I'm 42). I remember double digit infation, and things looked pretty bleak. Only recently have I begun to look back at how life has changed since then, and how change is (usually) a gradual process. Most of the big changes that have occurred were unexpected, so the place to look for the future is not the present.
(BTW) The August issue of IEEE Spectrum has a interesting article on space elevators, which could really sidestep the shuttle debate and render it mute.
"CacheLogic conducted direct packet monitoring of Internet backbones and ISPs data streams via Layer 7 packet analysis. 61.44% of current peer-to-peer traffic is video, 11.34% is audio, 27.22% belongs to other category. On a global scale, 46% of P2P traffic is video in Microsoft formats. 65% of all audio files by volume of traffic are still traded in the MP3 format, but a surprising 12.3% are in the open-source OGG file format (almost all of which are exclusively traded on the BitTorrent network, particularly in Asia)."
It seems that the stats are by packet volume, not by file count.
I think the parent was refering to a level of abstraction, where a "modifier" is used to change the default action to an alternate. This maps easily from a one-button+keyboard modifier to a multibutton mouse.
That being said (it's been a while since I've used MacOS) you used to be able to get the contextual menu in MacOS by holding the mouse button down for a fraction of a second longer than a simple click (click-hold), so that the mouse implemented 4 events without any modifiers: click (select), double click (action), click-hold (contextual menu), and drag (move/multiselect). The two-button mice give you the same 4 events, but with more buttons: l-click (selection), l-double click (action), r-click (contextual menu), and l-click-drag (move/multiselect).
The mouse wheel adds a new dynamic, which I like, but I don't see it as an essential part of the man-machine interface.
It seems that the apple site hasn't updated it's specs page for the powermac. While the specs page shows 4 configurations (1.8 single, 2.0 dual, 2.3 dual, and 2.7 dual), the apple store only has the dual processor models for sale(!).
Well that should be enough for 85% of the people who read/. -- seems like it's already a done-deal. The other 15% would probably go with whatever Xbox-360 comes with...
There are two that I'm familiar with: XScibble and XStroke. I used to use these on my Stylistic 1200 (120 MHz Pentium tablet). I liked the alphabet xstroke used over the one in xscribble, plus it had a nice full screen capability (no special window to write in).
This is where the (free) market jumps in. If I wish to pay someone to maintain a certain piece of software, I will look at the potental costs of ownership of that software before I select in the first place. If one software product appears to have poor documentation and a propensity to crash, I would have an incentive to look for a "cheaper" alternative. Acceptance/adoption of any software package would necessarily depend on how easily it can be supported.
According to an independent study, it takes an average of four months to watch the latest episodes of top-rated shows like Lost and Desperate Housewives.
I must have even more lag in my over-the-air one-way distribution network. When did a show about housewives who are lost and desperate start airing?
I know, I know. Mod away: (-1) Dumb Joke. Goodbye.
While Googling for robotic applications for the gumstix sbc, I came across this guy's page. (actually, his new page is here.) Pretty cool. He's been designing and building a small 2-wheeled bot called the Belligerator that looks like a fun little project.
The better question is "how productive is that?" Honestly, if I had to deal with waiting a few seconds every time I wanted a context menu I would be pretty annoyed and I'm sure I'm not alone in that. Using control as a modifier to create a contextual click I can understand.
The actual delay should be tunable (for me a holding the mouse down and still a fraction of a second longer would suffice). The idea of a context menu is that you're doing something non-standard with the object that requires extra attention, otherwise you would have clicked or double-clicked for the default behaviour. If you are constantly using contextual menus, perhaps the UI needs reworking? The bottom line is not how many buttons are on the mouse (the apple mouse is one, big button now), but effective UI design. I would *love* to see a mouse with a twiddler on it.;^)
Using a control modifier would require using two hands, which I would think be more awkward (Apple does use a control/command key modifier, but for less-used options like opening a new window while closing the parent). As for the wheel -- I love my mouse wheel, but that's not necessarily the best thing for all users. The wheel is especially handy when the UI is set for focus-follows-mouse, but that is not the default behavior in MacOS (as far as I recall).
Exactly. This is almost as bad as the apples-are-too-expensive posts. I have a Power mac with both the origninal one-button optical mouse as well as a Logitec wheel mouse. When my 5-year-old daughter wants to draw in the GIMP, guess which mouse she reaches for? Guess which mouse my wife uses (an which one she refuses to use)?
Software developers could easily adapt to a one-button mouse model, if only they wanted to. Hold down the mouse for more than a few seconds and generate a "left-button" event. How hard is that?
The first mother board only has 1 133MHz PCI-X slot and 2 66MHz slots, the Apple has 1 133 MHz PCI-X slot and 2 100 MHz PCI-X (full length) slots. It has Firewire-400 while the Apple has Firewire-800, and the Apple case is specially designed to be quiet. I assume the USB on board is 2.0.
The other MoBo has 2 133 MHz PCI-X (good) and one 66 MHz PCI-X, but only firewire-400.
The third only has one PCI-X at 100MHz (it seems to be targeted to the 1/2U server market), no firewire, and 8MB on-board ATI Rage graphics chip.
Slashdot Mirror
Well, I certainly won't claim that I am absolutely right. However, from the GPL FAQ:
If I distribute GPL'd software for a fee, am I required to also make it available to the public without a charge?
No. However, if someone pays your fee and gets a copy, the GPL gives them the freedom to release it to the public, with or without a fee. For example, someone could pay your fee, and then put her copy on a web site for the general public.
That seems pretty clear, but I couldn't have told you that from reading the actual license.
Wait, we're not talking about this OpenDoc?
Oh well, my old version of Cyberdog will have to do...
With my limited understanding of the GPL (IANAL), I believe that a vendor can charge what he/she wants for the binary version of the software, and is then obligated to provide the source (to that same person) at a nominal additional cost. The GPL only requires providing the source to those parties that receive the binaries. The customer, however, has no obligation (under the GPL) to keep the source code secret, and can redistribute the code at any cost (including none) to other parties. If MV adds an additional license to the GPL'd source that prevents redistribution, *that* would be a violation of the GPL.
So where are all these MV customers and why haven't they provided the MV kernel source for the rest of us?
This sort of space exploration is realively cheap, considering the payoffs involved. I suggest we knock off a couple "bridges to nowhere" from our budget, or ask for some money back from the Big Dig, instead.
This is a good point. The total probablity of mission success would be the wieghted sum of all possible events minus thier mutual probablity. To get through the first few events (transit to orbit and landing), the amount of over-engineering required could easily render the probabilty of failure later on negligible.
P(failure)=P(crash) +P(land){P(!survive landing)+P(all other failure events after landing)-overlap}
It's not clear if the tech in the article is a simple beamformer-type array, or something closer to an acoustic bullet array. Because of the size of the individual array elements (1 m^2),I would assume that the array is designed for low frequency operation ( 20Hz). It would be hard to beamform at such a low frequency (due to large wavelengths -- 20 Hz translates to around 19 meter spacing for a lambda/4 array). Acoustic bullet research focuses on the radiator element and how to efficiently launch high-energy acoustic waves into the medium (water in this case).
And if as has been suggested most PPCs go into embedded systems, Yellow Dog is totally unsuitable. There are plenty of embedded solutions for Linux out there and I expect they all work quite well already.
YDL is a handy development platform for embedded PPC targets (at least for PPC 750s and 74xxs). I'm not sure about the other embedded PPC processors.
my mistake. soury.
Of course, speaking of the Boomers, I fear that my generation (I'm 28) might be one of those unlucky historical examples of one which didn't get to do jack shit because they were so busy catering to the needs of their wealthy elders while trying to patch up the disasterous debts they left us. By the time they start to croak en masse it'll be too late to do anything all that interesting -- we'll be too old and too unimaginative, left only with the shadow of the dreams we once entertained.
I used to think the same thing (I'm 42). I remember double digit infation, and things looked pretty bleak. Only recently have I begun to look back at how life has changed since then, and how change is (usually) a gradual process. Most of the big changes that have occurred were unexpected, so the place to look for the future is not the present.
(BTW) The August issue of IEEE Spectrum has a interesting article on space elevators, which could really sidestep the shuttle debate and render it mute.
From the Article:
"CacheLogic conducted direct packet monitoring of Internet backbones and ISPs data streams via Layer 7 packet analysis. 61.44% of current peer-to-peer traffic is video, 11.34% is audio, 27.22% belongs to other category. On a global scale, 46% of P2P traffic is video in Microsoft formats. 65% of all audio files by volume of traffic are still traded in the MP3 format, but a surprising 12.3% are in the open-source OGG file format (almost all of which are exclusively traded on the BitTorrent network, particularly in Asia)."
It seems that the stats are by packet volume, not by file count.
I think the parent was refering to a level of abstraction, where a "modifier" is used to change the default action to an alternate. This maps easily from a one-button+keyboard modifier to a multibutton mouse.
That being said (it's been a while since I've used MacOS) you used to be able to get the contextual menu in MacOS by holding the mouse button down for a fraction of a second longer than a simple click (click-hold), so that the mouse implemented 4 events without any modifiers: click (select), double click (action), click-hold (contextual menu), and drag (move/multiselect). The two-button mice give you the same 4 events, but with more buttons: l-click (selection), l-double click (action), r-click (contextual menu), and l-click-drag (move/multiselect).
The mouse wheel adds a new dynamic, which I like, but I don't see it as an essential part of the man-machine interface.
It seems that the apple site hasn't updated it's specs page for the powermac. While the specs page shows 4 configurations (1.8 single, 2.0 dual, 2.3 dual, and 2.7 dual), the apple store only has the dual processor models for sale(!).
What firmware upgrade? Could you provide more info?
Sony's PS3 is to use the Blu-Ray format.
/. -- seems like it's already a done-deal.
Well that should be enough for 85% of the people who read
The other 15% would probably go with whatever Xbox-360 comes with...
There are two that I'm familiar with: XScibble and XStroke.
I used to use these on my Stylistic 1200 (120 MHz Pentium tablet). I liked the alphabet xstroke used over the one in xscribble, plus it had a nice full screen capability (no special window to write in).
Well, doesn't that beat all. Hopefully Intel will dump the whole x86 instruction set...
I'd say it shows that he's *wise*. Preemptive flattery goes a long way when you spend too much time in the garage.
This is where the (free) market jumps in. If I wish to pay someone to maintain a certain piece of software, I will look at the potental costs of ownership of that software before I select in the first place. If one software product appears to have poor documentation and a propensity to crash, I would have an incentive to look for a "cheaper" alternative. Acceptance/adoption of any software package would necessarily depend on how easily it can be supported.
According to an independent study, it takes an average of four months to watch the latest episodes of top-rated shows like Lost and Desperate Housewives.
I must have even more lag in my over-the-air one-way distribution network. When did a show about housewives who are lost and desperate start airing?
I know, I know.
Mod away: (-1) Dumb Joke.
Goodbye.
While Googling for robotic applications for the gumstix sbc, I came across this guy's page. (actually, his new page is here.) Pretty cool. He's been designing and building a small 2-wheeled bot called the Belligerator that looks like a fun little project.
The actual delay should be tunable (for me a holding the mouse down and still a fraction of a second longer would suffice). The idea of a context menu is that you're doing something non-standard with the object that requires extra attention, otherwise you would have clicked or double-clicked for the default behaviour. If you are constantly using contextual menus, perhaps the UI needs reworking? The bottom line is not how many buttons are on the mouse (the apple mouse is one, big button now), but effective UI design. I would *love* to see a mouse with a twiddler on it.
Using a control modifier would require using two hands, which I would think be more awkward (Apple does use a control/command key modifier, but for less-used options like opening a new window while closing the parent).
As for the wheel -- I love my mouse wheel, but that's not necessarily the best thing for all users. The wheel is especially handy when the UI is set for focus-follows-mouse, but that is not the default behavior in MacOS (as far as I recall).
You can download YDL iso's free, you just have to wait for Terrasoft to post them a while after the release to the subscribers on ydl.net.
The YDL yum repository is also free, and somewhat faster with updates.
Exactly. This is almost as bad as the apples-are-too-expensive posts. I have a Power mac with both the origninal one-button optical mouse as well as a Logitec wheel mouse. When my 5-year-old daughter wants to draw in the GIMP, guess which mouse she reaches for? Guess which mouse my wife uses (an which one she refuses to use)?
Software developers could easily adapt to a one-button mouse model, if only they wanted to. Hold down the mouse for more than a few seconds and generate a "left-button" event. How hard is that?
Another example:
d uc ts_id=168
http://www.micronux.com/catalog/product.php?pro
The base price is around $1200, but configure the thing to compare to a dual 2.0 GHz PPC970 system and you get more like:
Viper 2875 Dual Opteron Workstation
Dual AMD Opteron 248
512MB DDR400 ECC Registered (1x512M)
TYAN S2875 AMD-8111 GbE Audio
NVIDIA FX5200 128M TV DVI DH
SoundBlaster Audigy
Intel PRO1000MT Gigabit Adapter
56K V.90 Hardware Modem
Maxtor 120GB SATA 7200RPM 8M
Samsung 16X DVD±RW DL Black
TAC-T01 Aluminum Tower Case Black
ThermalTake Silent PurePower 420W
Logitech Keyboard Black
Logitech Optical Mouse Black
Fedora Core 3 x86-64 4CD
Price: $2,368.00
Which sounds about right
I certainly could be wrong (flame away)
:
Mother board Price: ~$500
The first mother board only has 1 133MHz PCI-X slot and 2 66MHz slots, the Apple has 1 133 MHz PCI-X slot and 2 100 MHz PCI-X (full length) slots. It has Firewire-400 while the Apple has Firewire-800, and the Apple case is specially designed to be quiet. I assume the USB on board is 2.0.
The other MoBo has 2 133 MHz PCI-X (good) and one 66 MHz PCI-X, but only firewire-400.
The third only has one PCI-X at 100MHz (it seems to be targeted to the 1/2U server market), no firewire, and 8MB on-board ATI Rage graphics chip.
Processor: 2.2GHz Opteron compares 2.0 GHz PPC970
(At Outpost.com)
AMD Opteron 248 OSA248AUBOX Retail Boxed Processor
AMD Opteron Processor:
Outpost #: 4064692
$530 ea
Graphics:
The Apple has an NVIDIA GeForce FX 5200 Ultra with 64MB of DDR SDRAM. Add around $50 for an equivalent card.
so we're talking at least $1600 for MoBo+processors+graphics card.
+$100 160GB+ SATA HD
+$120 8xDVD+/-RW
+$80 512MB RAM
brings you up to $1900 without a case, which should be made for silent operation (7+ low rpm fans).
I guess Apple forces you to buy the OS ($100 extra?), but the Apple is perhaps only $200-$300 more expensive than building it yourself.