Well if you actually RTFA, Red Hat wants to hack the JVM so that it supports real-time features.
What's wrong with Red Hat wanting to add real-time features?
So in other words, they want their own Red Hat Realtime Java fork.
So?
If the hypothetical Red Hat version passed the Java conformance testing, they could call it Java. If it didn't, they'd have to call it something else. Either way it doesn't hurt Sun.
It's entirely possible that Red Hat might do a good job of it, and Sun might choose to include Red Hat's work in the official Java. That would be a win for everyone, Sun included.
It's easy to throw around words like "hack" and "fork" and make them sound scary. It's much more difficult to come up with a rational explanation of how open-sourcing Java would actually harm Sun.
Easy, they still hold the exclusive license this way
If they were to GPL it, they would STILL hold the exclusive license. GPLing it doesn't give away the ownership, and it doesn't prevent the owner from also licensing it under other terms.
The same is true of various other open source licenses.
And in any case, that doesn't answer my question as to how it would hurt their bottom line.
For the last time, Java has the source available. It just isn't "Free Software"
Yes, so please explain again how it will hurt Sun's bottom line to go the extra step of making it open source. They've gone more than halfway there, but they're not going to get the real benefits of open source without finishing the job.
in the sense that you can't release your own binaries or fork the code.
In other words, it's not even CLOSE to being open source. Being able to release your own binaries is one of the strengths of open source (or free software). If that were allowed, we might actually have a working Java plugin for Firefox and Mozilla for 64-bit Linux.
"Almost open source" isn't good enough, when what's missing is the very thing that makes open source useful and worthwhile.
Jeezus, people, get over this self-righteous trip about source code. As an SE, I appreciate having source code available, but will in general just run what works.
That's EXACTLY why it would be better for it to be open source (or free software). Then we could actually hack together a working Java plugin for Linux on 64-bit Athlon64/Opteron platforms, since Sun is apparently unwilling to do so. That's just ONE example of how the current closed-source JDK is deficcient but could be fixed if it was open source.
Redhat isn't trying to make money on alternative Java implementations. They're doing it because the Sun implementation isn't available to them under licensing terms they are willing to accept. If Sun's Java license were open source (or better yet, free software), Redhat would likely not be investing nearly as much money and effort to develop an alternative.
If Sun fully opensources Java like Redhat wants them to it will result in Sun losing millions per year in license fees from IBM and others.
I don't think so. The reason that IBM and others pay license fees is that they want something more from Sun than just to download the binaries and sources, which Sun lets them do without paying any fees.
I think it would be fantastic if they opened up for everyone. But they have to look out for ol' number one first, or there won't be any Sun at all.
OK, so explain how making the binary available as a no-charge download contributes more money to their bottom line than releasing it as open source or free software would.
This license isn't even remotely "GNU/Linux friendly". It's neither "open source" nor "free software". It's "The Emperor's New License".
Sun wants to have their cake and eat it too; they want the benefits of open source without actually opening the source.
How long until fusion power can be put into production?
About 50 years.
And I'm not just being flippant, though the answer has been 50 years for the last 30 years or more.
ITER isn't going to be operational until 2016 at the earliest, and it's an experimental reactor not expected to be a net energy producer. Based on operational experience with ITER and IFMIF (for which construction has not even started), another experimental reactor will be designed and constructed with the goal of net energy production. Perhaps that might be operational by 2035. And if it works well enough, it's *remotely* possible that a commercial reactor could be designed and constructed, and be operational by 2055.
When all is said and done, fusion recactors are expected to produce *slightly* less expensive electricity than fission.
The big win with fusion will require a major theoretical breakthrough rather than simply carrying the current plans to their logical conclusion.
Also, if we do get large scale fusion, is it really going to be cleaner and safer than modern fission plants?
In general it's reasonable to expect that they'll be cleaner and safer. There is no possibility of a runaway chain reaction; the reactor only contains enough fuel at any given time to operate for a fraction of a second, vs. months or years for a fission reactor. If the fusion reaction containment fails, the reaction quickly stops, without serious damage to the reactor and without any abnormal leakage of radioactive material. A fusion reactor can't "melt down".
A fusion reactor will produce a greater quantity of radioactive waste (crumbling radioactive shielding and structural materials after years of exposure to high neutron flux), but fortunately the waste will have a very short half-life so it won't be dangerous for too many decades, and will thus be easier to store. No need to worry about safety over geological time scales, or about whether our descendents will be able to read warning signs printed in 21st century languages.
Yes, the 11/44 is a great machine. The last of the all-TTL PDP-11s. It was even designed to support multiprocessing, though such a configuration was never actually offered.
Unfortunately the H7140 power supply is very prone to failures. And it uses 35V AC fans for which replacements are not readily available.
Terry Kennedy reported back in 1996 that he'd found a bug in the CPU that allowed unprivileged users to crash the machine, but I haven't been able to track down the details.
My point is that without an free market for something, there is no established objective value. If someone is giving something away, there are two ways to interpret its value:
There is a market, and the market value is zero
There is effectively no market, so there is no established value
Note that neither case results in the amount $3000 being a value that can reasonably be claimed for the item given away. The $3000 is just an arbitrary number that some marketing person pulled out of his or her nether region.
I don't have any problem with people throwing in extra items with a purchase in order to add incentive for that purchase, but I don't like it when they insult my intelligence with a ridiculous made-up value for those extra items.
For that matter, I don't like being told that if I buy item A for $39.95, I get item B for "free". If item B is free, I'll take one, but you can keep the item A, and I'm not paying for it. The reality is that item B isn't free, and there's no way you can get it for free. The word "free" is instead used to mean that the quoted price is the price for both item A and item B together. Another "marketing breakthrough" which is used because of the psychological appeal of the notion of getting something "free".
I'm always disgusted when I see the ridiculous "valued at" statements in advertising. I publish some relatively obscure free (GPL'd) software, so I suppose I may as well assert that it's "valued at $10,000,000".
I'd like it to work approximately as well as a cordless phone used on a land line. No dropped calls, no momentary dropouts and very little garbling or static during calls.
Fundamentally what this requires is greater density of cell sites, so it's more a problem with infrastructure than with the phones. That's why I said that the Treo works about as poorly (or well) as any other cellphone.
I'm looking for a cell phone that will do three things well: make calls, [...]
You can find any other combination of features you want as long as you're willing to give up that first requirement. AFAICT, there are no phones that do that well. That's probably why they compete on how many other features they can cram in.
I expect eventually to see a product announcement for an amazing new "cellular phone" that has an incredible set of features, but doesn't actually place or receive phoen calls.
Anyhow, I'm pretty happy with the Treo 650. It makes calls about as poorly (or well, depending on your point of view) as any other cell phone, it syncs with my desktop, and it will act as a modem. Though it doesn't do EDGE or UMTS/HSPDA. I'm hoping that their next GSM Treo that runs PalmOS will do HSPDA.
I was an Apple employee, but left before the campus on Infinite Loop opened. When I have visitors from out of town, I like to take them for a ride around that campus, so that they can say that they've travelled the entire length of an infinite loop, from one endpoint to the other.
Infinite Loop isn't actually a loop; it has two endpoints on Mariani Drive.
for a cryptographer to use it they must make it very, very likely that this source generates true random numbers.
If a source is unpredictable, by definition it generates "true random numbers".
There are no physical processes which are unpredictable today AND ALSO that we can be certain will still be unpredictable tomorrow. Currently it is generally believed that the collapse of quantum wave functions is unpredictable, but this has not been proven.
AFAICT, there's no reason to think that the quasar as a random number source is any more or less random than the thermal noise in a diode. The latter is much more commonly used as a random number source, however. We can easily manufacture more diodes, but manufacturing more quasars is somewhat problematic at the present time.
That's not randomness at all. It only seems random
An interesting assertion, but without any backing evidence.
they're confusing randomness with unpredictability
There isn't any particularly better definition of randomness than "unpredicability". Some things are more unpredicable than others. Some things can even be proven to be unpredictable; for instance, the Blum-Blum-Shub PRNG has been proven to be unpredictable if you don't have a copy of its internal state, because it is mathematically intractable to derive the state from the output.
It seems unlikely that it will become possible to predict the behavior of quasars as you suggest; we can't even accurately predict the weather on earth, which is a much smaller system than a quasar. For that matter, we can't predict the detailed behavior of a lava lamp, making that a reasonable source of random numbers (but patented!).
anyone renewing a designated document (e.g. passport) will be able to opt-out of getting a card until 2010, but will still have their details put on the National ID Register immediately.
What a brilliant plan! Get most of the drawbacks of a National ID card, without the card itself! You've heard of "buy now, pay later"? This is the opposite!
Maybe the MPs can get reelected by fooling the voters into thinking that somehow this plan doesn't harm their privacy nor move the UK ever closer to being a police state.
Since there's no practical way to obtain rocket fuel from the moon itself
What makes you think that? The lunar crust contains plenty of aluminum and oxygen, which can be used as rocket fuel. You just set up solar panels to generate electricity to run a refining operation. What could be simpler?:-)
The Centauri won't give us jumpgate technology until 2155.
Yeah, and probably only because they want to enlist our aid in keeping those darn pesky Narns under control.
RIP Andreas Katsulas, May 18 1946 - Feb 13 2006
"I believe that when we leave a place a part of it goes with us and part of us remains. Go anywhere in these halls, when it is quiet, and just listen. After a while you will hear the echoes of all of our conversations, every thought and word we've exchanged. Long after we are gone, our voice will linger in these walls for as long as this place remains. But I will admit that the part of me that is going will very much miss the part of you that is staying." -- G'Kar
Michael Crichton's essay in the New York Times attempts a thoughtful summary of Metabolite's primary assertion: they not only own the connection between homocysteine levels in the blood and vitamin B12 deficiency, but also any thought connecting the two
Great. So MC has not only infringed the Metabolite patent by thinking about the connection between homocysteine levels and vitamin B12 deficiency, but by publishing the article he has incited all the readers to also infringe the patent. And now I'm probably inciting more people to infringe.
Slashdot Mirror
If the hypothetical Red Hat version passed the Java conformance testing, they could call it Java. If it didn't, they'd have to call it something else. Either way it doesn't hurt Sun.
It's entirely possible that Red Hat might do a good job of it, and Sun might choose to include Red Hat's work in the official Java. That would be a win for everyone, Sun included.
It's easy to throw around words like "hack" and "fork" and make them sound scary. It's much more difficult to come up with a rational explanation of how open-sourcing Java would actually harm Sun.
The same is true of various other open source licenses.
And in any case, that doesn't answer my question as to how it would hurt their bottom line.
"Almost open source" isn't good enough, when what's missing is the very thing that makes open source useful and worthwhile.
Redhat isn't trying to make money on alternative Java implementations. They're doing it because the Sun implementation isn't available to them under licensing terms they are willing to accept. If Sun's Java license were open source (or better yet, free software), Redhat would likely not be investing nearly as much money and effort to develop an alternative.
This license isn't even remotely "GNU/Linux friendly". It's neither "open source" nor "free software". It's "The Emperor's New License". Sun wants to have their cake and eat it too; they want the benefits of open source without actually opening the source.
And I'm not just being flippant, though the answer has been 50 years for the last 30 years or more.
ITER isn't going to be operational until 2016 at the earliest, and it's an experimental reactor not expected to be a net energy producer. Based on operational experience with ITER and IFMIF (for which construction has not even started), another experimental reactor will be designed and constructed with the goal of net energy production. Perhaps that might be operational by 2035. And if it works well enough, it's *remotely* possible that a commercial reactor could be designed and constructed, and be operational by 2055.
When all is said and done, fusion recactors are expected to produce *slightly* less expensive electricity than fission.
The big win with fusion will require a major theoretical breakthrough rather than simply carrying the current plans to their logical conclusion.
In general it's reasonable to expect that they'll be cleaner and safer. There is no possibility of a runaway chain reaction; the reactor only contains enough fuel at any given time to operate for a fraction of a second, vs. months or years for a fission reactor. If the fusion reaction containment fails, the reaction quickly stops, without serious damage to the reactor and without any abnormal leakage of radioactive material. A fusion reactor can't "melt down".A fusion reactor will produce a greater quantity of radioactive waste (crumbling radioactive shielding and structural materials after years of exposure to high neutron flux), but fortunately the waste will have a very short half-life so it won't be dangerous for too many decades, and will thus be easier to store. No need to worry about safety over geological time scales, or about whether our descendents will be able to read warning signs printed in 21st century languages.
Unfortunately the H7140 power supply is very prone to failures. And it uses 35V AC fans for which replacements are not readily available.
Terry Kennedy reported back in 1996 that he'd found a bug in the CPU that allowed unprivileged users to crash the machine, but I haven't been able to track down the details.
The PDP-11/73 does NOT use Unibus. It uses Qbus. And there weren't any PDP-8 computers as ARPANET hosts.
"So raise your hand if you think that was a Russian water tentacle."
Maybe it's more like "prophet". You know, like putting a statue of the Budda on a lazy susan so you can turn a prophet.
- There is a market, and the market value is zero
- There is effectively no market, so there is no established value
Note that neither case results in the amount $3000 being a value that can reasonably be claimed for the item given away. The $3000 is just an arbitrary number that some marketing person pulled out of his or her nether region.I don't have any problem with people throwing in extra items with a purchase in order to add incentive for that purchase, but I don't like it when they insult my intelligence with a ridiculous made-up value for those extra items.
For that matter, I don't like being told that if I buy item A for $39.95, I get item B for "free". If item B is free, I'll take one, but you can keep the item A, and I'm not paying for it. The reality is that item B isn't free, and there's no way you can get it for free. The word "free" is instead used to mean that the quoted price is the price for both item A and item B together. Another "marketing breakthrough" which is used because of the psychological appeal of the notion of getting something "free".
I'm always disgusted when I see the ridiculous "valued at" statements in advertising. I publish some relatively obscure free (GPL'd) software, so I suppose I may as well assert that it's "valued at $10,000,000".
Fundamentally what this requires is greater density of cell sites, so it's more a problem with infrastructure than with the phones. That's why I said that the Treo works about as poorly (or well) as any other cellphone.
I expect eventually to see a product announcement for an amazing new "cellular phone" that has an incredible set of features, but doesn't actually place or receive phoen calls.
Anyhow, I'm pretty happy with the Treo 650. It makes calls about as poorly (or well, depending on your point of view) as any other cell phone, it syncs with my desktop, and it will act as a modem. Though it doesn't do EDGE or UMTS/HSPDA. I'm hoping that their next GSM Treo that runs PalmOS will do HSPDA.
Infinite Loop isn't actually a loop; it has two endpoints on Mariani Drive.
There are no physical processes which are unpredictable today AND ALSO that we can be certain will still be unpredictable tomorrow. Currently it is generally believed that the collapse of quantum wave functions is unpredictable, but this has not been proven.
AFAICT, there's no reason to think that the quasar as a random number source is any more or less random than the thermal noise in a diode. The latter is much more commonly used as a random number source, however. We can easily manufacture more diodes, but manufacturing more quasars is somewhat problematic at the present time.
It seems unlikely that it will become possible to predict the behavior of quasars as you suggest; we can't even accurately predict the weather on earth, which is a much smaller system than a quasar. For that matter, we can't predict the detailed behavior of a lava lamp, making that a reasonable source of random numbers (but patented!).
Maybe the MPs can get reelected by fooling the voters into thinking that somehow this plan doesn't harm their privacy nor move the UK ever closer to being a police state.
I expect that the US will follow suit. :-(
RIP Andreas Katsulas, May 18 1946 - Feb 13 2006
"I believe that when we leave a place a part of it goes with us and part of us remains. Go anywhere in these halls, when it is quiet, and just listen. After a while you will hear the echoes of all of our conversations, every thought and word we've exchanged. Long after we are gone, our voice will linger in these walls for as long as this place remains. But I will admit that the part of me that is going will very much miss the part of you that is staying." -- G'Kar