OOo for Mac require X11, looks like crap, acts like crap, and can't properly size a window to save its life.
Yes, and you know why that is? Because Apple's X11 server works like crap: it's dog slow, doesn't integrate well with the desktop, and isn't even started up automatically when needed.
And why does X11 on OS X work like crap? Because Apple has no interest in fixing it: their goal is to tie developers to their proprietary window system, and they are going to do whatever it takes to achieve that.
OOo is up to the task, the problem is Apple and their X11 server. And rather than being Apple's patsy and porting every open source application to Apple's proprietary platform, people should stop investing time in open source for Macintosh until Apple properly supports X11.
I have been using my first mac (powerbook) for almost a month now, and i can say that NeoOffice/J does a much better job for a user in mac os x than the X11 version of OpenOffice.
That's Apple's fault: they are putting roadblocks in the way of people trying to do a better job with X11 integration on Macintosh. The OOo developers got so annoyed with Apple's behavior that they stopped working on Macintosh integration.
There is no technical reason why X11 couldn't be as smoothly integrated into OS X as Carbon and Cocoa are: X11 should be preinstalled and run automatically on every Macintosh, and its window management should be tightly integrated with the Macintosh desktop.
The fact that it isn't (and that X11 is dog slow on Macintosh) is Apple politics: Apple doesn't want X11 to run too well on Macintosh--they probably are afraid that if X11 becomes well enough integrated so that people can write applications with a native L&F, it would become the predominant API on OS X. To prevent that, Apple wants X11 to run just well enough so that people can use workstation applications on Macintosh if they have to, but so that X11 applications continue to look foreign and don't integrate very well.
The school district should have given the laptop to students as a one-time grant and let it be their own responsibility. If they lose it, they have to worry about getting another one. If they mess up the software installation and can't get schoolwork done, they have to reinstall the software.
For the district to own the laptops and then use the criminal justice system to try to tell students what to do with the laptops just sets all the wrong incentives. Under those conditions, I would have just locked up the school-provided laptop in a safe place and brought my own.
See, SCO's argument was that the GPL is invalid, unenforceable, illegal and unconstitutional. So, given that, it would be wrong and unfair to say "Sorry, you can't use this license to distribute your software, so all of you have to stop" to all of those open source developers that want so badly to distribute their work to the world.
I don't think SCO has standing to ask the court to decide what should happen to other licensees of other people's copyrighted software. Furthermore, the existence of dual-licensed GPL software also makes that a hard argument to make.
Except maybe in the case of Linux, which belongs mostly to SCO anyway.
And that's why SCO really has to argue that the GPL is not valid: if it is valid, then SCO has already given all the people they ever distributed Linux to an implicit license to all their applicable intellectual property and their case falls apart. (Of course, if it isn't valid, they have committed, and are continuing to commit, massive copyright violations.)
how, in fact, do so many of these comments saying "if sco says the licence is invalid, then they cant use it..."
My posting didn't say that; you are simply a sloppy reader.
that still doesnt mean that they cant abide by it, until they do
Quite right: as long as SCO complies with the terms of the GPL, they can keep shipping GPL'ed software. I was merely pointing out that their position is inconsistent: on the one hand, they claim the GPL is invalid, on the other hand, they are relying on the validity of the GPL to ship software.
If SCO actually were to prevail in court with their argument, then they would have to stop shipping GPL'ed software, and they might actually be guilty of wilfull copyright infringement for the GPL'ed software they have shipped in the past.
Note that SCO is making these hare-brained arguments against the GPL simply because without them, their case would collapse before it even started: under the GPL, by shipping Linux, SCO has already given everybody a license to their intellectual property in the Linux kernel.
Once upon a time people wouldnt just mod crap up cause it agreed with their "SCO SUCKS" attitude, but because it was actually intelligent. Well, I dont read slashdot so much anymore.
Actually, what sucks is people like you. Don't let the door hit you on the way out.
All the source code on Community Codebase is available under OSI approved licenses. Additionally, they're "standard" licenses, such as the MIT, BSD, and PHP licenses. They're not under a special eBay corporate license.
Thanks for the clarification. You may want to make sure that that comes across more clearly in future interviews.
First of all, there are several incompatible versions of RSS already, so one more won't hurt.
Secondly, what is the point of the Safari comment? Safari has a tiny market share and it wasn't the first browser with RSS features either. Is there some kind of competition going on between Microsoft and Apple who can copy other people's features faster? Why not limit mentions of Apple to those areas where they actually came up with something for themselves?
A deeper problem with SCO's position is the following. SCO seems to assume that if the GPL is invalid, they can do with the software whatever they want.
But that's not the way it works. The software is copyrighted, GPL or no GPL. The GPL is the agreement that permits people to copy the software under certain limited conditions. If the GPL isn't valid, it just means that everything returns to the situation without the GPL and SCO can't ship any GPL'ed software at all.
By analogy, assume you pay for a license to Microsoft Windows with a check. Then, your check bounces and your license becomes invalid. Does that mean that Microsoft Windows is all of a sudden public domain? No, it means that you can't use it.
SCO can argue that the GPL is "unenforceable" and "voidable" if they like. So, what remains if the GPL is "unenforcable" and "voidable"? A large piece of copyrighted software to which they have no license at all. Meaning, that if their assertion is true, they are wilfully violating multiple copyrights.
The precise look and layout is "original", in the same way that, say, iPhoto is "original" or next week's cover of Business Week is "original". The technology, of course, is not original.
When Apple abused the patent system in this fashion against Microsoft, they were soundly defeated.
Apple continues to abuse the patent system, filing patents on things they didn't invent and filing patents on trivialities. One grave recent abuse has been their desktop search patent. Other serious problems for FOSS projects have been their springloaded folders and FreeType-related patents. Then there are plenty of design patents on the appearance of OS X. And those are just the tip of the iceberg.
Apple isn't guilty of abusing their patents NOW
You are so wrong. Go search the USPTO for what Apple has filed over the last few years.
Furthermore, I don't see Apple calling for patent reform. The fact is that they like the current system. The company hasn't changed since their Microsoft lawsuits, they have only gotten less powerful and less important.
That depends on the open source license. If they distribute the software under the GPL or LGPL, they may implicitly be licensing all necessary patents along with the software. That's another reason why the GPL and LGPL are good for you.
he problem is that JavaScript dialog boxes do not display or include their origin,
I can assure you that even if they did contain their origin, it would still not make much of a difference--most users wouldn't bother to look.
Maybe what we need is a secure web standard, something that runs only over https, uses strict XHTML, dispenses with JavaScript, pop-ups, frames, and popups, and is used for banking and similar applications. Preferably, that should be a separate browser.
The Community Codebase is free for all members of eBay's Developers Program and PayPal Developer Network
If you have to be a member of some "developer program" or "developer network", then the code isn't open source. The term "Community Codebase" also suggests that it isn't open source, but that it is a program like Microsoft's "Shared Source" or Sun's "Community Source". Let's hope that eBay didn't mislabel a non-open source license as open source.
Even if the eBay code turns out to be truly open source, it is still closely tied to eBay's service. On balance, it's still better if such code is open sourced, but the decision for eBay to open source a bit of client code is a completely different business decision than, say, IBM or Intel open sourcing a compiler.
Companies like Sun, Microsoft, and (possibly) eBay seem to be trying hard to redefine the term open source for their own marketing purposes. Don't let them get away with it. A company that misapplies the term open source is being deceptive and should be condemned strongly.
(I suppose it isn't surprising that someone like McVoy would go along with this sort of thing; he has, after all, been trying to take advantage of the open source community as well.)
MIT isn't just computer geeks, you know. People in mechanical engineering, civil engineering, chemical engineering, and related disciplines tend to be much less geeky. And even among computer geeks, there is a significant population that hangs out at the gym in their spare time.
Science likes to tantalize you with incredible possibilities that float just outside your reach;)
I think that's just whining. The amount of progress made over the last 30 years alone is astounding: enormous chip densities, DLP chips, single atom imagery, high temperature superconductors, the human genome, amorphous metals, etc.
I fully expect superconductivity, superfluidity, and super-strong materials to become more mainstream over the next decade--that has become more engineering than science at this point. There even is a possibility that some form of desktop fusion will become a reality for energy generation, although that will still require some fundamental insights.
read it again. they get 200% improvement over central coding in a clustered environment. they get 300% improvement over no coding in a clustered environment.
Come on, you are trying to change the subject by looking at a different claim. You yourself wrote:
the big improvement they talk about. the flash crowd problem is one of the main problems they are trying to tackle with networking coding and they anticipate a 40-200% improvement over central coding or unencoded transfer.
I responded directly to that statement:
Specifically, they get a 40% improvement over central coding and a 200% improvement over unencoded transfers
this is exactly the big improvement they talk about. the flash crowd problem is one of the main problems they are trying to tackle with networking coding and they anticipate a 40-200% improvement over central coding or unencoded transfer.
Specifically, they get a 40% improvement over central coding and a 200% improvement over unencoded transfers. Hence, the big improvement is from going to unencoded to central coding, and the small improvement is going from central coding to network coding, even in their own (flawed) experiments.
but the paper clearly describes the superiority of network coding over erasure/online codes performed centrally as far as their model / simulations are concerned.
Star Trek also clearly describes the superiority of warp drive over impulse drive, and it clearly is superior in the simulated Star Trek universe, but until someone actually builds one in the real world, that remains fiction, too.
Whether I understand the "concepts" or not has nothing to do with the interpretation of the performance claims the authors make; those are summarized on page 2:
2) We provide experimental evidence that, in many situations of practical interest, network coding performs better than transmitting unencoded blocks, or using techniques that are based on erasure codes, which can be thought as coding but only at the server. Network coding performs better by almost a factor of two compared to performing encoding at the server and by a factor of three compared to not coding at all when the topology is clustered. Similarly, network coding improves the download rates by almost 20% compared to source coding and by more than 30% compared to no coding in an heterogeneous network. During the early stages of a ash crowd, network coding outperforms source coding and no coding by 40% and 200% respectively. Even when the system reaches a steady-state, network coding still provides signicant benets compared to using other techniques. Moreover, when tit-for- tat incentives are used the performance of network coding is barely impacted, while, other schemes suffer signicantly.
You can dig deeper into what they actually did and draw your own conclusion. My interpretation is that they have not demonstrated a big improvement of their method over known FEC methods, but perhaps your notions of "big" differ from mine.
If link prefetching is going to make a difference at all, it's only going to be in Google search results; other search engines aren't big enough, and the statistics of non-search engine sites are not going to be affected by prefetching (since market share is computed by sessions, not pages). But it is totally unclear that prefetching would make any significant difference even for Google search results; in order to alter results, prefetching would have to occur for sites that are actually used to determine market share, and those sites would have to be not visited by the user explicitly. What fraction of Google search results are that? I suspect it's not very significant.
As for a "more efficient caching mechanism", sorry, I don't believe it. Again, it's not number of pages fetched but sessions that count. It is unlikely that a correct implementation of HTTP caching would permit Opera to eliminate entire sessions, simply because web sites generally configure some content as uncacheable in order to track their users or show ads. If Opera's caching mechanism eliminates those accesses, then Opera isn't "more efficient", it's simply broken.
Market share statistics based on browser log files are tricky and probably not all that accurate. But problems stem mostly from sample bias, not technology. Technologically, I suspect that the privacy features (disabling of cookies, adblock, etc.) in Firefox, if anything, cause it to be undercounted, rather than overcounted.
As for Opera, the people seem to be getting desparate. I'm sorry that there isn't much a market for their product anymore, but that's the way it is. Maybe it's time for them to look for a different market niche.
Computer criminals and black-hat-hackers are as much a fact of life as rain showers in Seattle, earthquakes in California, flus in winter, and accidents on highways.
Security isn't an accidental byproduct of software, it is one of its primary functions; if software doesn't provide security, then it is defective. That's just like if you buy a padlock, you have an expectation that it actually works as a lock. The padlock manufacturer can't say "oh, well, our padlock doesn't work, but that's really the criminal's fault".
Any vendor that puts out software that contains easily avoidable security holes (like buffer overflows, backdoors,...) is very much to blame. In fact, it should be possible to hold liable for negligence.
This guy's looking to cash in on the success of another company. It's parasitic, childish, and it clogs the court system that is already full of garbage as it is...
Well, and how would that be different from Apple's past patents and lawsuits?
Oh, actually, I suppose there is one big difference: this guy at least came up with something original (even if it should not be patentable), while Apple just keeps ripping off other people's ideas and even attempts to patent them and misrepresent them as their own.
the Patent system needs fixing.
I quite agree. But this case is a bad example for the excesses of the patent system, both because this patent is far from the worst, and because the target of the lawsuit is a brazen patent system abuser themselves.
That said... when someone uses the patent system as their personal checkbook
Well, I hope you start by boycotting Apple, because they have been a big and blatant abuser of the patent system themselves.
Solar sails are not intended to be propelled by solar wind, they are propelled by light. For interstellar voyages, you'd propel the solar sail actively (with a laser). That technology will also be tested as part of this experiment.
No. Rockets don't accidentally escape into space--it takes way too much fuel. If it were that easy, the sail would have been sent on a trajectory away from earth.
Solar sails aren't driven by solar wind or protons, they are driven by light (photons).
An interstellar voyage might be possible, but would probably require a laser or microwave system aimed at the sail for much of its journey (a brief "push" like that is also being tested as part of this experiment).
Slashdot Mirror
OOo for Mac require X11, looks like crap, acts like crap, and can't properly size a window to save its life.
Yes, and you know why that is? Because Apple's X11 server works like crap: it's dog slow, doesn't integrate well with the desktop, and isn't even started up automatically when needed.
And why does X11 on OS X work like crap? Because Apple has no interest in fixing it: their goal is to tie developers to their proprietary window system, and they are going to do whatever it takes to achieve that.
OOo is up to the task, the problem is Apple and their X11 server. And rather than being Apple's patsy and porting every open source application to Apple's proprietary platform, people should stop investing time in open source for Macintosh until Apple properly supports X11.
I have been using my first mac (powerbook) for almost a month now, and i can say that NeoOffice/J does a much better job for a user in mac os x than the X11 version of OpenOffice.
That's Apple's fault: they are putting roadblocks in the way of people trying to do a better job with X11 integration on Macintosh. The OOo developers got so annoyed with Apple's behavior that they stopped working on Macintosh integration.
There is no technical reason why X11 couldn't be as smoothly integrated into OS X as Carbon and Cocoa are: X11 should be preinstalled and run automatically on every Macintosh, and its window management should be tightly integrated with the Macintosh desktop.
The fact that it isn't (and that X11 is dog slow on Macintosh) is Apple politics: Apple doesn't want X11 to run too well on Macintosh--they probably are afraid that if X11 becomes well enough integrated so that people can write applications with a native L&F, it would become the predominant API on OS X. To prevent that, Apple wants X11 to run just well enough so that people can use workstation applications on Macintosh if they have to, but so that X11 applications continue to look foreign and don't integrate very well.
The school district should have given the laptop to students as a one-time grant and let it be their own responsibility. If they lose it, they have to worry about getting another one. If they mess up the software installation and can't get schoolwork done, they have to reinstall the software.
For the district to own the laptops and then use the criminal justice system to try to tell students what to do with the laptops just sets all the wrong incentives. Under those conditions, I would have just locked up the school-provided laptop in a safe place and brought my own.
See, SCO's argument was that the GPL is invalid, unenforceable, illegal and unconstitutional. So, given that, it would be wrong and unfair to say "Sorry, you can't use this license to distribute your software, so all of you have to stop" to all of those open source developers that want so badly to distribute their work to the world.
I don't think SCO has standing to ask the court to decide what should happen to other licensees of other people's copyrighted software. Furthermore, the existence of dual-licensed GPL software also makes that a hard argument to make.
Except maybe in the case of Linux, which belongs mostly to SCO anyway.
And that's why SCO really has to argue that the GPL is not valid: if it is valid, then SCO has already given all the people they ever distributed Linux to an implicit license to all their applicable intellectual property and their case falls apart. (Of course, if it isn't valid, they have committed, and are continuing to commit, massive copyright violations.)
how, in fact, do so many of these comments saying "if sco says the licence is invalid, then they cant use it..."
My posting didn't say that; you are simply a sloppy reader.
that still doesnt mean that they cant abide by it, until they do
Quite right: as long as SCO complies with the terms of the GPL, they can keep shipping GPL'ed software. I was merely pointing out that their position is inconsistent: on the one hand, they claim the GPL is invalid, on the other hand, they are relying on the validity of the GPL to ship software.
If SCO actually were to prevail in court with their argument, then they would have to stop shipping GPL'ed software, and they might actually be guilty of wilfull copyright infringement for the GPL'ed software they have shipped in the past.
Note that SCO is making these hare-brained arguments against the GPL simply because without them, their case would collapse before it even started: under the GPL, by shipping Linux, SCO has already given everybody a license to their intellectual property in the Linux kernel.
Once upon a time people wouldnt just mod crap up cause it agreed with their "SCO SUCKS" attitude, but because it was actually intelligent. Well, I dont read slashdot so much anymore.
Actually, what sucks is people like you. Don't let the door hit you on the way out.
All the source code on Community Codebase is available under OSI approved licenses. Additionally, they're "standard" licenses, such as the MIT, BSD, and PHP licenses. They're not under a special eBay corporate license.
Thanks for the clarification. You may want to make sure that that comes across more clearly in future interviews.
First of all, there are several incompatible versions of RSS already, so one more won't hurt.
Secondly, what is the point of the Safari comment? Safari has a tiny market share and it wasn't the first browser with RSS features either. Is there some kind of competition going on between Microsoft and Apple who can copy other people's features faster? Why not limit mentions of Apple to those areas where they actually came up with something for themselves?
A deeper problem with SCO's position is the following. SCO seems to assume that if the GPL is invalid, they can do with the software whatever they want.
But that's not the way it works. The software is copyrighted, GPL or no GPL. The GPL is the agreement that permits people to copy the software under certain limited conditions. If the GPL isn't valid, it just means that everything returns to the situation without the GPL and SCO can't ship any GPL'ed software at all.
By analogy, assume you pay for a license to Microsoft Windows with a check. Then, your check bounces and your license becomes invalid. Does that mean that Microsoft Windows is all of a sudden public domain? No, it means that you can't use it.
SCO can argue that the GPL is "unenforceable" and "voidable" if they like. So, what remains if the GPL is "unenforcable" and "voidable"? A large piece of copyrighted software to which they have no license at all. Meaning, that if their assertion is true, they are wilfully violating multiple copyrights.
Explain to me how this is original?
The precise look and layout is "original", in the same way that, say, iPhoto is "original" or next week's cover of Business Week is "original". The technology, of course, is not original.
When Apple abused the patent system in this fashion against Microsoft, they were soundly defeated.
Apple continues to abuse the patent system, filing patents on things they didn't invent and filing patents on trivialities. One grave recent abuse has been their desktop search patent. Other serious problems for FOSS projects have been their springloaded folders and FreeType-related patents. Then there are plenty of design patents on the appearance of OS X. And those are just the tip of the iceberg.
Apple isn't guilty of abusing their patents NOW
You are so wrong. Go search the USPTO for what Apple has filed over the last few years.
Furthermore, I don't see Apple calling for patent reform. The fact is that they like the current system. The company hasn't changed since their Microsoft lawsuits, they have only gotten less powerful and less important.
That depends on the open source license. If they distribute the software under the GPL or LGPL, they may implicitly be licensing all necessary patents along with the software. That's another reason why the GPL and LGPL are good for you.
he problem is that JavaScript dialog boxes do not display or include their origin,
I can assure you that even if they did contain their origin, it would still not make much of a difference--most users wouldn't bother to look.
Maybe what we need is a secure web standard, something that runs only over https, uses strict XHTML, dispenses with JavaScript, pop-ups, frames, and popups, and is used for banking and similar applications. Preferably, that should be a separate browser.
The Community Codebase is free for all members of eBay's Developers Program and PayPal Developer Network
If you have to be a member of some "developer program" or "developer network", then the code isn't open source. The term "Community Codebase" also suggests that it isn't open source, but that it is a program like Microsoft's "Shared Source" or Sun's "Community Source". Let's hope that eBay didn't mislabel a non-open source license as open source.
Even if the eBay code turns out to be truly open source, it is still closely tied to eBay's service. On balance, it's still better if such code is open sourced, but the decision for eBay to open source a bit of client code is a completely different business decision than, say, IBM or Intel open sourcing a compiler.
Companies like Sun, Microsoft, and (possibly) eBay seem to be trying hard to redefine the term open source for their own marketing purposes. Don't let them get away with it. A company that misapplies the term open source is being deceptive and should be condemned strongly.
(I suppose it isn't surprising that someone like McVoy would go along with this sort of thing; he has, after all, been trying to take advantage of the open source community as well.)
"Refute" means to demonstrate conclusively that something is false. Sony hasn't done that. They have merely disputed the claim.
MIT isn't just computer geeks, you know. People in mechanical engineering, civil engineering, chemical engineering, and related disciplines tend to be much less geeky. And even among computer geeks, there is a significant population that hangs out at the gym in their spare time.
Science likes to tantalize you with incredible possibilities that float just outside your reach ;)
I think that's just whining. The amount of progress made over the last 30 years alone is astounding: enormous chip densities, DLP chips, single atom imagery, high temperature superconductors, the human genome, amorphous metals, etc.
I fully expect superconductivity, superfluidity, and super-strong materials to become more mainstream over the next decade--that has become more engineering than science at this point. There even is a possibility that some form of desktop fusion will become a reality for energy generation, although that will still require some fundamental insights.
Come on, you are trying to change the subject by looking at a different claim. You yourself wrote:
I responded directly to that statement:
this is exactly the big improvement they talk about. the flash crowd problem is one of the main problems they are trying to tackle with networking coding and they anticipate a 40-200% improvement over central coding or unencoded transfer.
Specifically, they get a 40% improvement over central coding and a 200% improvement over unencoded transfers. Hence, the big improvement is from going to unencoded to central coding, and the small improvement is going from central coding to network coding, even in their own (flawed) experiments.
but the paper clearly describes the superiority of network coding over erasure/online codes performed centrally as far as their model / simulations are concerned.
Star Trek also clearly describes the superiority of warp drive over impulse drive, and it clearly is superior in the simulated Star Trek universe, but until someone actually builds one in the real world, that remains fiction, too.
You can dig deeper into what they actually did and draw your own conclusion. My interpretation is that they have not demonstrated a big improvement of their method over known FEC methods, but perhaps your notions of "big" differ from mine.
I find those claims rather dubious.
If link prefetching is going to make a difference at all, it's only going to be in Google search results; other search engines aren't big enough, and the statistics of non-search engine sites are not going to be affected by prefetching (since market share is computed by sessions, not pages). But it is totally unclear that prefetching would make any significant difference even for Google search results; in order to alter results, prefetching would have to occur for sites that are actually used to determine market share, and those sites would have to be not visited by the user explicitly. What fraction of Google search results are that? I suspect it's not very significant.
As for a "more efficient caching mechanism", sorry, I don't believe it. Again, it's not number of pages fetched but sessions that count. It is unlikely that a correct implementation of HTTP caching would permit Opera to eliminate entire sessions, simply because web sites generally configure some content as uncacheable in order to track their users or show ads. If Opera's caching mechanism eliminates those accesses, then Opera isn't "more efficient", it's simply broken.
Market share statistics based on browser log files are tricky and probably not all that accurate. But problems stem mostly from sample bias, not technology. Technologically, I suspect that the privacy features (disabling of cookies, adblock, etc.) in Firefox, if anything, cause it to be undercounted, rather than overcounted.
As for Opera, the people seem to be getting desparate. I'm sorry that there isn't much a market for their product anymore, but that's the way it is. Maybe it's time for them to look for a different market niche.
Computer criminals and black-hat-hackers are as much a fact of life as rain showers in Seattle, earthquakes in California, flus in winter, and accidents on highways.
...) is very much to blame. In fact, it should be possible to hold liable for negligence.
Security isn't an accidental byproduct of software, it is one of its primary functions; if software doesn't provide security, then it is defective. That's just like if you buy a padlock, you have an expectation that it actually works as a lock. The padlock manufacturer can't say "oh, well, our padlock doesn't work, but that's really the criminal's fault".
Any vendor that puts out software that contains easily avoidable security holes (like buffer overflows, backdoors,
This guy's looking to cash in on the success of another company. It's parasitic, childish, and it clogs the court system that is already full of garbage as it is...
Well, and how would that be different from Apple's past patents and lawsuits?
Oh, actually, I suppose there is one big difference: this guy at least came up with something original (even if it should not be patentable), while Apple just keeps ripping off other people's ideas and even attempts to patent them and misrepresent them as their own.
the Patent system needs fixing.
I quite agree. But this case is a bad example for the excesses of the patent system, both because this patent is far from the worst, and because the target of the lawsuit is a brazen patent system abuser themselves.
That said... when someone uses the patent system as their personal checkbook
Well, I hope you start by boycotting Apple, because they have been a big and blatant abuser of the patent system themselves.
Solar sails are not intended to be propelled by solar wind, they are propelled by light. For interstellar voyages, you'd propel the solar sail actively (with a laser). That technology will also be tested as part of this experiment.
No. Rockets don't accidentally escape into space--it takes way too much fuel. If it were that easy, the sail would have been sent on a trajectory away from earth.
Solar sails aren't driven by solar wind or protons, they are driven by light (photons).
An interstellar voyage might be possible, but would probably require a laser or microwave system aimed at the sail for much of its journey (a brief "push" like that is also being tested as part of this experiment).