It hasn't really even begun to pay off and it's already being dumped!
Good riddance to bad rubbish! They should have done it much sooner, as it wasn't ever going to "pay off". In fact, they really should pull the plug on it right now. Long before they launched the first component, the plan for the station had been scaled down so much as to be esentially useless.
Perhaps they might claim that the distribution cost is $70, but that subscribers get a $20 coupon. This would seem to get around the requirement violation you state.
I thought about that, but I don't think so.
3b requires them to charge no more than the cost of distribution. Can they plausibly claim that the cost of burning a CD and mailing it to some people is $20 less than the cost of doing that for others? Maybe, but it seems fairly shaky.
But does not this mean that the third party presumably must be able to present the actual written offer that was initially distributed with the binary?
Where do you think it says that? It says they have to make a written offer (presumably "written" in a text file in the distribution would be OK), but why should they only need to honor the offer if a copy of the offer is sent back to them?
Read the actual GPL before you spout off incorrect statements about what you think it means.
Anyone redistributing GPL'd code, such as Sveasoft, is required to comply with at least one of the options in section 3. They no longer comply with 3a (object code accompanied by source code), and they aren't covered by 3c (noncommercial redistribution of object code you got from another party with an offer of source code). Thus they must comply with section 3b. Here's the actual text of section 3b (emphasis added):
b) Accompany it with a written offer, valid for at least three
years,
TO GIVE ANY THIRD PARTY, for a charge no more than your
cost of physically performing source distribution, a complete
machine-readable copy of the corresponding source code, to be
distributed under the terms of Sections 1 and 2 above on a medium
customarily used for software interchange; or,
So they don't have to send the written offer to a third party, but they have to honor it when a third party requests the source.
Then the GPL applies to recipients of the binaries. Only those third parties involved need apply.
Not necessarily. The old Sveasoft model was that customers paid $20 for a subscription, and got access to binaries "accompanied by" sources (both on the same download server), thus satisfying section 3a of the GPL.
Their new model of only providing sources to people who pay $50 apparently is an attempt to rely on section 3b of the GPL instead. Section 3b, however, requires that the source code be made available to "any third party", not just the party to whom they have distributed the binary.
Sveasoft has stated that they will only sell the $50 source code disc to customers who have bought the $20 subscription. This seems to violate the "any third party" requirement of section 3b.
Sveasoft are terminating the accounts of those that redistribute source. This seems like an attempt to prevent people redistributing source, which I understand to be illegal under the GPL
They are not "imposing any further restrictions on the recipients' exercise of the rights granted herein." (section 6 of the GPL). Terminating a subscription does not affect those rights in any way. They even got confirmation of that from the FSF.
They could also say that if you redistribute their version, they will use a voodoo doll to place a curse on you, and that wouldn't violate the GPL terms either. Both actions are entirely outside the scope of the GPL.
Except that when they distribute the object code to subscribers, it *IS* accompanied by the source code (another file available from the same server in the same directory). This satisifies section 3a of the GPL.
The part that I was missing is that this is no longer true. Future binary releases will not be accompanied by source releases. So they will in fact be relying on section 3b. However, they've also said that they will only sell the source code CD to subscribers who have paid for the binary. This DOES violate the GPL; section 3b requires that they provide the source code to any third party (but they can charge for it).
Except that when they distribute the object code to subscribers, it *IS* accompanied by the source code (another file available from the same server in the same directory). This satisifies section 3a of the GPL.
A distributor only has to satisify ONE of sections 3a, 3b, and 3c. Since Sveasoft satisfies section 3a, they are not required to satisfy section 3b, and 3b is the only section that requires that an offer be made to third parties.
They're offering the source at a cost; you still have access to the source, just at a price. This doesn't violate the GPL license.
Actually they provide the source code at no charge, to everyone that pays them $20/year for access to the binaries. Since the binaries are only provided to people who pay, and the source is provided to those people, they are under no obligation to make the source available to third parties (GPLv2 section 3a), though they cannot forbid those who receive the software from redistributing it.
And that's just for access to "prerelease" code. The official releases are available for no-charge download of both object and source code.
The essential point is then that Sveasoft stopped people distributing source.
But Sveasoft has NOT stopped people from distributing source. They sell a subscription. Anyone who buys the subscription gets the opportunity to download binaries and sources. As per the GPL, Sveasoft does NOT prevent redistribution of the source. However, they will terminate your subscription if you redistribute the source.
That policy does not conflict with the GPL in any way, because the GPL does not require Sveasoft to provide subscriptions.
When you purchaes that lightning fast 2400 baud modem you were making a quantum leap from that clunky 300 baud modem because the speeds had been predetermined for you.
A switch from 300 bps to 2400 bps dialup modems does not qualify as a quantum leap, because the 300 bps was not an inherently fixed rate. The "300 baud" modems did not actually
constrain the bit rate arbitrarily. They used Bell 103 FSK modulation, most commonly at rates of 110 and 300 baud, and less commonly at 134.5 and 450 baud. The modulator didn't care what the bit rate was, but if you raised it too high the demodulator wouldn't work. If you wanted to use the modem at 271.828 bps, it would work fine as long as the other end also wanted that bit rate.
Bell 202 FSK modulation did the same thing, typically at 1200 bps (half-duplex).
On the other hand, the 1200 bps full duplex modems (Bell 212, V.22) used synchronous PSK modulation, so they truly did have discrete bit rates. Technically they are 600 baud modems, with two bits per symbol.
A 2400 bps dialup modem (V.22bis modulation) is also only 600 baud, using QAM modulation with four bits per symbol (16 constellation points).
So a switch from a Bell 212 1200 bps modem to a V.22bis 2400 bps modem would meet your definition of a quantum leap.
Most higher-speed dialup modems other than "56K" also use QAM modulation, with a higher baud rate and more bits per symbol.
Ok, so part of this new deal is buyers can't resell what they bought. I assume the original license would handle that.
Maybe. Did you sign a license agreement when you bought the song?
I find restrictions on resale to be very obnoxious, and will avoid buying music that has such restrictions. By law (in the U.S.), if I buy a CD (or DVD, or book, etc.), I am allowed to resell it. Copyright law does not give the copyright holder the right to restrict or suppress the secondary market. This is known as the "first sale doctrine".
When I buy a CD or DVD, I am not signing a license agreement, or even agreeing to one via "shrink-wrap" or "click-wrap". I have purchased the physical medium, and a substantial set of legal rights to use, lend, or sell that medium.
Despite the supposed convenience of buying music online, I find it to have substantially less value than I get when buying a CD. I wind up being subject to a restrictive license, and the quality of the music sometimes suffers as a result of compression. Yet I am typically expected to pay as much or more money.
Vote with your wallet.
I should note that there are some musicians, bands, and even publishers that recognize that they can make money online without resulting to the typical draconian tactics of the RIAA cabal. It's my understanding that Magnatune "gets it". I haven't bought any music from them yet, but I think I'll give it a try.
Building an unmanned jet aircraft isn't unlawful either, nor is publishing the plans for one.
Everyone has reacted to this as if he has built something with a high explosive payload and aimed it at a city; nothing could be further from the truth. Get some sense of proportion, please.
In flagrantly building the missle (and continuing to do so) he was giving the finger to non-proliferation.
Can you cite a NZ law or regulation prohibiting building pulsejet-powered unmanned aircraft? That's all it was. That's not "proliferation" any more than anyone else building unmanned jet aircraft would be. He's certainly not the only civilian doing that.
Instead of distcc, I use
nc by
Steven Ellis. It seems to be more flexible, though I'm not an expert
on distcc, so I'm not certain.
I think nc can be used like distcc by redefining CC="nc gcc". However,
more commonly it is done by putting $(NC) at the beginning of the build
rules. Then you can use nc for any build rules, not just C compiles.
In addition to use with make, nc works well with
SCons.
After looking through hundreds if not thousands of job postings, everyone is looking for 3+ years of network admin experience or 5+ years of C++ experience even for an entry level position. How is one expected to gain that kind of experience when no one will hire you without the experience?
You can't get a job that way, even with a lot of experience. If a company has a job open, they'll interview candidates that respond to a job listing, but unless one of them really stands out, they'll hire someone from an employee referral instead.
Out of more than a dozen tech jobs I've held, I only ever got ONE though job listings, and that was because I was living in a backwater place at the time and the company had few applicants. All the rest were by knowing someone at the company (directly or indirectly). You don't necessarily have to know them well; a casual acquaintance is enough to get your foot in the door.
The companies DON'T CARE whether you can find a job or not. There are too many IT people on the market, so they can afford to only hire people with a lot of experience even for an entry level position. They believe (correctly or not) that if they get someone with less experience it will cost them more money.
If you really want to stick with this career path,
you need to find a company through friends, friends of friends, etc., that needs someone, possibly part time or as a consultant, and almost certainly for substandard pay. Work up from there.
A friend and I drove from Denver to San Jose CA last week (23 hours 7 minutes trip time). We stopped in Green River UT (because there are no services on I-70 for the next 107 miles). We ate at the Arby's
which is part of the Gas-N-Go at the west end of town. I was surprised that they provide free DSL internet access; on the wall under almost all of the tables one finds electrical outlets and phone and Ethernet jacks. Some of the tables have telephones.
They might have also had Wifi, but there was no sign indicating such, and I lost my 802.11g card a week earlier so I couldn't check for it. Anyhow, we were in a hurry so I didn't try the wired Ethernet either.
I don't mind that the reigstration requires cookies, but they should explicitly state that, especically if you try to submit a registration and the cookie is not present. Instead, they say something about the verification code not matching, and "Are you a robot?". Very unhelpful.
Why recommend the older when the newer is much faster with more memory?
And more bugs, too. For instance, when the program is longer than 260 lines, many GTO instructions that should be valid give an error 4. So the extra program memory isn't all that useful, unless you're writing mostly straight-line code. See the November/December 2003 issue of Datafile (the HPCC club magazine) for a list of more HP 12C Platinum bugs, most of which admittedly not as serious as the GTO problem.
Hewlett-Packard used to put a huge amount of effort into making sure that there were no obvious bugs. And when a nasty bug did slip through, they usually would provide ROM upgrades on request. But Hewlett-Packard no longer exists; it's been replaced by the new HP, which has abandoned the "HP Way" for what is apparently the Compaq Way. There's no word of any ROM upgrade for the HP 12C Platinum.
The existence of serious bugs in something as simple as the GTO instruction makes me rather skeptical of the accuracy of the financial calculations. Some of the financial calculations such as solving for interest rates are actually quite challenging to do accurately, and Hewlett-Packard spent a lot of time on numerical analysis to make sure that they did a good job of it. I rather doubt that the new HP has done nearly as much homework on the mathematical algorithms the HP 12C Platinum.
The original HP-12C also has a better keyboard. Hewlett-Packard invested more effort into making the calculators ergonomic. When you buy a $100 calculator, you should be able to get reliable keys with good tactile feedback and double-shot injection molding so that the legends don't wear off. But HP doesn't care about this. What they've forgotten is that by cutting corners and making the products as cheap as possible (but not any more inexpensive), they are not only failing to cultivate new customer loyalty, but are losing the loyalty of their existing customer base.
I have yet to meet someone that actually needs a faster 12C, or one with more memory. But I don't doubt that there are a few people somewhere that do. For everyone else, I strongly recommend the HP-12C over the HP 12C Platinum.
Slashdot Mirror
3b requires them to charge no more than the cost of distribution. Can they plausibly claim that the cost of burning a CD and mailing it to some people is $20 less than the cost of doing that for others? Maybe, but it seems fairly shaky.
So they don't have to send the written offer to a third party, but they have to honor it when a third party requests the source.
Their new model of only providing sources to people who pay $50 apparently is an attempt to rely on section 3b of the GPL instead. Section 3b, however, requires that the source code be made available to "any third party", not just the party to whom they have distributed the binary.
Sveasoft has stated that they will only sell the $50 source code disc to customers who have bought the $20 subscription. This seems to violate the "any third party" requirement of section 3b.
They could also say that if you redistribute their version, they will use a voodoo doll to place a curse on you, and that wouldn't violate the GPL terms either. Both actions are entirely outside the scope of the GPL.
A distributor only has to satisify ONE of sections 3a, 3b, and 3c. Since Sveasoft satisfies section 3a, they are not required to satisfy section 3b, and 3b is the only section that requires that an offer be made to third parties.
And that's just for access to "prerelease" code. The official releases are available for no-charge download of both object and source code.
That policy does not conflict with the GPL in any way, because the GPL does not require Sveasoft to provide subscriptions.
Bell 202 FSK modulation did the same thing, typically at 1200 bps (half-duplex).
On the other hand, the 1200 bps full duplex modems (Bell 212, V.22) used synchronous PSK modulation, so they truly did have discrete bit rates. Technically they are 600 baud modems, with two bits per symbol.
A 2400 bps dialup modem (V.22bis modulation) is also only 600 baud, using QAM modulation with four bits per symbol (16 constellation points).
So a switch from a Bell 212 1200 bps modem to a V.22bis 2400 bps modem would meet your definition of a quantum leap.
Most higher-speed dialup modems other than "56K" also use QAM modulation, with a higher baud rate and more bits per symbol.
I find restrictions on resale to be very obnoxious, and will avoid buying music that has such restrictions. By law (in the U.S.), if I buy a CD (or DVD, or book, etc.), I am allowed to resell it. Copyright law does not give the copyright holder the right to restrict or suppress the secondary market. This is known as the "first sale doctrine".
When I buy a CD or DVD, I am not signing a license agreement, or even agreeing to one via "shrink-wrap" or "click-wrap". I have purchased the physical medium, and a substantial set of legal rights to use, lend, or sell that medium.
Despite the supposed convenience of buying music online, I find it to have substantially less value than I get when buying a CD. I wind up being subject to a restrictive license, and the quality of the music sometimes suffers as a result of compression. Yet I am typically expected to pay as much or more money.
Vote with your wallet.
I should note that there are some musicians, bands, and even publishers that recognize that they can make money online without resulting to the typical draconian tactics of the RIAA cabal. It's my understanding that Magnatune "gets it". I haven't bought any music from them yet, but I think I'll give it a try.
Everyone has reacted to this as if he has built something with a high explosive payload and aimed it at a city; nothing could be further from the truth. Get some sense of proportion, please.
I think nc can be used like distcc by redefining CC="nc gcc". However, more commonly it is done by putting $(NC) at the beginning of the build rules. Then you can use nc for any build rules, not just C compiles.
In addition to use with make, nc works well with SCons.
Out of more than a dozen tech jobs I've held, I only ever got ONE though job listings, and that was because I was living in a backwater place at the time and the company had few applicants. All the rest were by knowing someone at the company (directly or indirectly). You don't necessarily have to know them well; a casual acquaintance is enough to get your foot in the door.
The companies DON'T CARE whether you can find a job or not. There are too many IT people on the market, so they can afford to only hire people with a lot of experience even for an entry level position. They believe (correctly or not) that if they get someone with less experience it will cost them more money.
If you really want to stick with this career path, you need to find a company through friends, friends of friends, etc., that needs someone, possibly part time or as a consultant, and almost certainly for substandard pay. Work up from there.
They might have also had Wifi, but there was no sign indicating such, and I lost my 802.11g card a week earlier so I couldn't check for it. Anyhow, we were in a hurry so I didn't try the wired Ethernet either.
I don't mind that the reigstration requires cookies, but they should explicitly state that, especically if you try to submit a registration and the cookie is not present. Instead, they say something about the verification code not matching, and "Are you a robot?". Very unhelpful.
I just passed through Mojave very early Friday morning; I wish I could have stayed to watch the launch, but I had to get back to San Jose.
Hewlett-Packard used to put a huge amount of effort into making sure that there were no obvious bugs. And when a nasty bug did slip through, they usually would provide ROM upgrades on request. But Hewlett-Packard no longer exists; it's been replaced by the new HP, which has abandoned the "HP Way" for what is apparently the Compaq Way. There's no word of any ROM upgrade for the HP 12C Platinum.
The existence of serious bugs in something as simple as the GTO instruction makes me rather skeptical of the accuracy of the financial calculations. Some of the financial calculations such as solving for interest rates are actually quite challenging to do accurately, and Hewlett-Packard spent a lot of time on numerical analysis to make sure that they did a good job of it. I rather doubt that the new HP has done nearly as much homework on the mathematical algorithms the HP 12C Platinum.
The original HP-12C also has a better keyboard. Hewlett-Packard invested more effort into making the calculators ergonomic. When you buy a $100 calculator, you should be able to get reliable keys with good tactile feedback and double-shot injection molding so that the legends don't wear off. But HP doesn't care about this. What they've forgotten is that by cutting corners and making the products as cheap as possible (but not any more inexpensive), they are not only failing to cultivate new customer loyalty, but are losing the loyalty of their existing customer base.
I have yet to meet someone that actually needs a faster 12C, or one with more memory. But I don't doubt that there are a few people somewhere that do. For everyone else, I strongly recommend the HP-12C over the HP 12C Platinum.