+30% of your profits are eaten by Apple. Lets say I create dual release of iPhone and PSP game. Here are my costs / profits:
Costs
PC $400
PSP devkit $1500 Profits
$20 per application ($40 minus manufacturing costs, and distribution/retail commission)
Costs
MAC $600
iPhone Store Minimum $100 Profits
$28 per application == $40 retail * 0.7
Assuming I'm selling the same application through retail channels and iPhone store, I'd need to sell the application through iPhone store for around 28.57 to break even vs. retail. The costs of starting a business in PSP vs. iPhone in your example is around $1200. That equates to 60 sales. In terms of 'mass' release gaming titles, this is a negligible cost. What this whole thing does mean is exactly what we've seen: AAA titles where developers invest a fair amount of money to make a 'really good game' are going to stay on the devices because of the fact that the systems are more customized to their needs, and that the distribution channel is larger (every PSP/DS is a potential sale vs. iPhones/iPodTouch where there will be a lot of people that would never consider playing/buying any game). The iPhone will be popular for toy and independent gaming studios that don't sell in the economies of scale that portable consoles require for profitability.
That said, USB and PCI were developed/approved by a consortium. Firewire was proprietary until Apple licensed it to outside parties. Mini-disk may not be considered proprietary if it had taken off, though tape, DVD, Blu-Ray, HD-DVD could all be considered in a grey area in terms of openness. Here's one scale of openness (I'm assuming said third party never uses the words "Display Port" or "Apple" in their marketing as this would violate their trademarks which I don't address in this example)
1. Can anyone develop DisplayPort enabled devices without reading the published specs or licensing the apple product (legal reverse engineering): Maybe, as long as said innovations aren't patentable in the country of creation / sale 2. Can anyone develop DisplayPort enabled devices based on their public specifications without being actionable by Apple: Probably not 3. Can anyone develop DisplayPort enabled devices with a signed licensee agreement: Probably, though the contract explicitly excludes any applicable patents 4. Can anyone develop DisplayPort enabled devices with a signed licensee agreement & patent license: Yes
PS: I could care less about this, but you're blowing a lot of hot air over something that isn't necessarily true.
That's the same argument that I have about consoles in general vs. PC gaming. Suck it up. You can't have a happy medium between the best development platform vs. financially success. I know the Wii will never play games that are capable on PS3/360, just as I'd know that there are games that PC's games that will never play properly on PS3/360's. That's life.
I don't necessarily disagree with what you're trying to convey, but your approach needs some work:
"Microsoft would tend to disagree with you, I'm sure. They managed to corner the market with a free product, in the case of Internet Explorer. Before you also start bleating at me about how they were leveraging their monopoly, realise how well Google are doing in the market on search."
Netscape was an ok web browser and when IE made it to around 4 or 5, they became ok as well. The difference was that IE -was- installed by default on the majority of new desktops (mshtml cannot be uninstalled) and Netscape wasn't good enough to justify users going out to find a new one. This was during a period where more and more casual PC users were entering the PC market to use the Internet. Many of those users would care less what 'browser' they used.
"some of said companies will add value to it (and close it)" Lets say that Microsoft took vorbis, added some proprietary and hence incompatible hooks to the product then made it part of the default Windows installation, nobody would cry bloody murder? I think the most likely case would be that the 'market' would split with some using microsoft's default broken implementation while others would go with the free open version. Same things happened with web development and Java. In both cases Microsoft embraced and extended to the point of breaking inter-compatibility.
"Corporations are entirely free to make closed, modified derivatives of Apache; again, if Stallman is right, why haven't they been destroyed yet?" Apache is still alive because of the people contributing to it. If a closed source product (like MS) wanted to embrace and extend Apache, that's their decision. But apache is itself an implementation of a specification, not the specification itself. The only benefit someone has of taking Apache code would be to save development costs in rolling out their own Web server. I'm sure there are a ton of companies that have done just that. If your point is to develop code that people will use, mission accomplished. If your goal was to foster your personal philosophy of software without commercial value, you're probably on less steady grounds.
"how come any of the BSDs still exist at all?" BSD is a bad example, since its lost favor for the vast majority of developers. Linux could be considered the slightly younger, more attractive platform to work on, so people jumped ship to Linux, but nothing stops BSD from living on forever. That's not to say its a success story in modern times vs. Apache which still maintains a very high relevance.
"why didn't Microsoft destroy the WC3 after they acquired Internet Explorer?" They had a monopoly but not exclusive monopoly on the market. That's why they embraced and extended their browser to include proprietary hooks like ActiveX and many non-standard coding extensions. If given enough time, they would've completely tossed out WC3 from consideration. Lucky for us, Mozilla took over the sinking Netscape product line and saved our ability to choose web platforms.
"Internet's protocols still exist, rather than single, monopolised implementations?" IPv4, TCP, UDP, all De-facto standards that EVERYONE uses. Maybe you should pick a better example, because this one is 100% inaccurate. If you're trying to come up with 'proprietary' protocols, you would be best to look at Instant Messangers, some streaming media, some email provisions and SMB. Beyond that, I can't recall many protocols that don't have proprietary hooks that limit their use. The major reason why protocols aren't proprietary is because the protocols aren't copyrightable in many(all?) countries. If its not illegal to copy, someone could analyse and reproduce said function and publish specifications for how it works without legal reprocussions. Once that's done, the protocol can be implemented by any who care to use it.
Things like Cryptography and Compression which are process intensive are already implemented in native hooks in the base JVM. For instance, if you're using a GZIPOutputStream to GnuZip compress your data, the actually compression is done in a native block. Just because -a part- of your code requires high thoughtput, it doesn't mean that the entire piece of code needs to be developed in said language. Think of it like the distinction between C and ASM. You'd never write your entire code in assembly anymore, but there are some algorithms/loops that you'll do a better job of optimizing by using assembly instead of your general purpose C compiler.
I think you're looking at deterministic in a different way than I. I know that when there are no longer any references to an Object, that object is completely out of scope from my application. I don't care how the system I'm on cleans it up, the same way that 99% of the time, you don't care about what a free/delete/close really does behind the scenes. What does matter is that:
1. The process of cleaning up this memory doesn't negatively impact the application's runtime performance
2. 'Native'/OS resources outside of the code are released in a timely manner (Sockets, shared memory, etc..)
Point 1 is addressed in several different GC's in Java. Most happy cases where fine grained timing is not mandatory have very little issue with one's code unless you're creating/destroying a LOT of objects in which case, object reuse would be more meaningful. Although it may not be perfect, I think Java does a pretty good job of mitigating the GC's interruption of an application.
Point 2 is where the real problem of non-finalization occurs. If one doesn't explicitly release these resources when a programmer is done with them, you'll start getting error for too many open files, or this or that. In this case, it should be up to the developer to specifically deallocate these pieces of code. In any Input/Output steam for instance, a user must explicitly call close() to free up those OS resources that have been allocated for them (This is done on finalization as well if you don't explicitly call it). I don't care about the memory being used, since the GC magically takes care of that for me. What I do care about is that the files are released.
Java does allow one to free these resources when you tell it to, so even though I could say its a non-issue, this brings up an interesting dilemma as a Java developer. If I expect that the GC takes care of everything behind the scenes, I may be coaxed into forgetting about this behaviour when I'm developing my code. Generally, an issue like this is either caught right away which is fine, or its caught much later in the development when the programs are being tested excessively, or when performing load testing, etc.. This just re-enforces how important it is to profile your code with a good profiler to see what's staying in use and what gets freed up. Most Java 'memory leaks' I've seen fall into the annoying but not game-breaking category, but you'll eventually see degradation in performance the longer the issue is left undetected.
that isn't in XP, hence nobody cares. You'll have the what, 30% market segment with Vista, and maybe 10% that are regular gamers who will be using this.
This will just encourage the further brokenness that Windows is turning the PC gaming platform. Good Job!
PS: Before everyone jumps in to say that everyone will jump into Win7, I think you're mistaken. The only way Microsoft will kill XP for most existing users would be to introduce a critical bug that they choose not to fix. I played with Win7 for a few days and can safely say that it doesn't add anything that I've ever wanted to use that a trivial search for google wouldn't find an as-good or better alternative. And maybe its just me, but pretty much every single UI 'enhancement' since circa Win2k is always a step backwards in terms of -my- productivity.
Its lucky that I'm Linux competent since Fedora/Gnome makes practically everything I need easy and uncluttered. If the barrier for entry was a little lower, I could see mass exodus potential coming as XP users take an honest look at what they -really- want to update to.
yeah, they have the ability to dump full heap/stack dumps on really bad errors if you enable them, but typically Java tells you (generally if not specifically) where your errors occur unless you're doing something really funky, or your error happens to be OutOfMemory, in which case the memory 'leak' could be difficult to trace without a good profiler like jprofiler.
1. Peer association break simple reference counting. If you're doing non-trivial OO coding, you'll almost certainly have cyclic associations throughout 2. Unless I'm missing something, the counter itself needs concurrency locking which may eventually become a big deal if you're dealing with a large thread count applications of this object.
In practice, you (a standard developer) never need to see the raw memory byte level contents of an object. That said, debuggers and profilers can and do get raw access to objects, but they use the many debugging/profiling hooks (JVMTI, JVMDI, etc..) into the JVM to do so, not by using a standard top level core API.
All of what you said is possible and much more once you have native code interface access... but, one cannot run native code (that hasn't been signed by for the JVM itself) unless the security context has been set to allow basically all operations. Since the native hook (JNI) allows for any use of C libraries within the processes code space, an application with native hooks looses all sand-box restrictions. Of course thats why a java applet on the internet requires a popup in order to elevate sandbox privileges to that level.
I don't see how you could reasonably graft on moded text editor bindings onto a modeless IDE, but by all means, write a plugin if you like. What I do like (though rarely use) is that basically every operation you could want to use within the IDE can be bound to any set of key combinations. We're talking several hundred unique operations key bindable. Of course the number of actual commands depends on what features/plugins you've installed.
I never use it for development, just text editing when I'm on windows, but this is probably the best free text editor that I've seen in a long time. Props to them.
Since I don't really know the answer, I'll ask: Can you use VC++ without using the platform SDK? If so, then good for all. If not, then your argument is irrelevant. If you can't use VC++ without a product that contains a flawed stack, then both products are broken by association.
>>Does it refactor methods signatures? >How often do you do that, really? Hell, I have that in VS and I still never use it. A lot. When you work on projects that are less than perfectly coupled, this moves from handy to necessary.
>>Rename methods, classes, fields or local variables? >Again, how often do you really do this that a simple search/replace in the current file won't be sufficient? Once again, I have this feature in VS... I never use it. If I rename/refactor an interface, my work has interfaces that are linked to literally hundreds of files. Its not something you want to risk doing wrong, or by hand. If I wanted to rename a method through eclipse, I would:
- Select a method to rename
- Shift-Alt-R (or use the right click menu
- Type in a new name and hit enter. After a few seconds of crunching, all references have been renamed..
This will find all structural links to that method in all open project that reference my editing project. Each and every reference to the method will get the name change, and life moves on. Through the command line, you could be spending days search/replacing every possible reference to the method in 100 files.
"You mean you have to stop your program to fix the function you're in the middle of?"
When I started switched my Java IDE to Eclipse, this was one of the big OMG moments. This single feature saves me so much time, no other argument could sway me away from using it. Beyond that, there are so many features packed in there that I can't even imagine developing without it any more.
8. Watch the movie again. The Marcus unit couldn't beat a T-800 alone. Secondly, after the 'Marcus incident' Skynet probably mothballed the entire concept. Instead it simply used the skin tissue from humans (with some slow regenerative or preservative qualities) that had not higher order facilities.
9. Not so much a goof as it is a large leap of faith for the writers
10. It was a lie. Skynet planted the idea in the resistance's head which is how they ultimately found the command/control sub
11. This is a gaping plot hole. Marcus -turning off the defense perimeter- moment when connecting to the central computer just means that either Skynet is moronic for leaving them disabled, or its a big gaping plot hole.
All things said, I loved the movie. It kept me going the whole way, and much like Die hard 4, it wasn't like the originals, but it wasn't bad in its own unique way.
4. *shrugs* never really noticed the bikes power source. Maybe nuclear cores were a new feature for the T-800's. Since he knew about the cores from T3, it wouldn't even be a logic gap that he knew all about them before even meeting a T-800 in person. (The larger flaw would be that there weren't any other terminators to assist the lone T-800 in killing John/Kyle)
5. Her character, just like the TV series would've reached the point on, is the only salvation for humanity against Skynet is to embrace the machines as more than just enemies or tools, and to reach a mutually agreeable outcome with those that can reason. Moon's empathy for Marcus show us that there are those in the Terminator universe that could see the machines (Marcus at least) as more than just a machine, and to treat him like a person/sentient. Marcus' ultimate sacrifice would be quite flat without his humanization in the eyes of the resistance.
6. You could be right, but the ocean's a big place, and even focusing an effort around continental USA wouldn't be a trivial feat. Doubly so if the submarine was in any way modern in disguising its sonar signatures.. *Shrugs*
0. Getting their act together as per Terminator 3, many were able to start organizing -while- judgment day was happening. This may be a little far fetched, but still wasn't it set years after judgment day?
1. According to their explanation, Skynet has strong zones of influence in major cities (as per all previous movies), but when battling in open terrain, humans seem to have the upper hand, or at least they're fighting an effective guerrilla war.
2. Major plot hole. Even if they were informed through the original Terminators Chip (assuming the T1 ending and not the subsequent rewrite of history in T2) they still wouldn't know what he looks like. The fact that Cyberdyne (USAF bought them later) were the ones who developed Marcus Wright, it would allude to the fact that they were very much in business.
3. I thought that was pretty evident in two points:
a. They needed real tissue to wrap all but Marcus' infiltrator terminators which were portrayed in T1 as the Arnold unit. Secondly, the infiltrators didn't all look like Arnold, which is why they eventually needed dogs to sniff them out (sad there weren't any dogs in the movie)
b. They were baiting John to raid Skynet to free the prisoners before the joint strike. I think the Kyle Reese thing was probably changed in a later draft not realizing how stupid a slip-up it really is (This from the people that thought nobody would care if a Terminator could infect and control standard cars what have no useful CPU functions remotely...)
just has sour grapes because they couldn't figure Linux out on their first (or second, or third..) go around and now they're spewing why it doesn't count. That being said, 1% is a pretty pathetic number. Lets hope this new found high point only leads to better things in the future. Maybe they'll be the proverbial Mozilla of an IE / Netscape consumer market struggle. We all know how that played out now, don't we?
.. I know that aren't power users or in the computer field couldn't tell you the difference between their computer and the OS. Just like Mac's. You're buying a mac, not Windows computer.
They DO use ratings from torrents and DVR's for I believe two weeks after the original viewing of the show. On DVR's, If you watch it after two weeks, it doesn't register as a viewing. I think if you download the torrent within two weeks of the original showing, it is considered a viewing.
Unless there's a lot of prop reuse, I doubt it makes much financial sense to go down the road of mini-series. You might as well make a 'cheap' movie that's the same length because you'll at least get a few chumps/rubes/guinea-pigs to pay > $10 to see it.
You're right. Skynet's conciousness was created through a virus, or a random -whatever- online, and after opening the defence network's flood gates to secure the virus, the reverse happened, and the virus all of a sudden had real physical presence in the world. After the nukes, who knows what, but I'm hoping Salvation will enlighten us all *hoping its better than T3*.
PS: T3 was flawed like pretty much all of the 3's sequels the last few years. Their ambition to always out BIG an audience got in the way of telling a comprehensive story. The story in my eyes did have a good core, but it got muddled up with *insert some new antagonist* which was the worst part of the movie, and stepped awkwardly through the conner romance. All in all, it just didn't flow like it should have.
Slashdot Mirror
+30% of your profits are eaten by Apple. Lets say I create dual release of iPhone and PSP game. Here are my costs / profits:
Costs
PC $400
PSP devkit $1500
Profits
$20 per application ($40 minus manufacturing costs, and distribution/retail commission)
Costs
MAC $600
iPhone Store Minimum $100
Profits
$28 per application == $40 retail * 0.7
Assuming I'm selling the same application through retail channels and iPhone store, I'd need to sell the application through iPhone store for around 28.57 to break even vs. retail. The costs of starting a business in PSP vs. iPhone in your example is around $1200. That equates to 60 sales. In terms of 'mass' release gaming titles, this is a negligible cost. What this whole thing does mean is exactly what we've seen: AAA titles where developers invest a fair amount of money to make a 'really good game' are going to stay on the devices because of the fact that the systems are more customized to their needs, and that the distribution channel is larger (every PSP/DS is a potential sale vs. iPhones/iPodTouch where there will be a lot of people that would never consider playing/buying any game). The iPhone will be popular for toy and independent gaming studios that don't sell in the economies of scale that portable consoles require for profitability.
Sure looks proprietary to me:
http://devworld.apple.com/softwarelicensing/agreements/pdf/MiniDisplayPortImpLicense.pdf
http://www.appletweets.net/apple-granted-a-patent-for-its-mini-displayport/
That said, USB and PCI were developed/approved by a consortium. Firewire was proprietary until Apple licensed it to outside parties. Mini-disk may not be considered proprietary if it had taken off, though tape, DVD, Blu-Ray, HD-DVD could all be considered in a grey area in terms of openness. Here's one scale of openness (I'm assuming said third party never uses the words "Display Port" or "Apple" in their marketing as this would violate their trademarks which I don't address in this example)
1. Can anyone develop DisplayPort enabled devices without reading the published specs or licensing the apple product (legal reverse engineering): Maybe, as long as said innovations aren't patentable in the country of creation / sale
2. Can anyone develop DisplayPort enabled devices based on their public specifications without being actionable by Apple: Probably not
3. Can anyone develop DisplayPort enabled devices with a signed licensee agreement: Probably, though the contract explicitly excludes any applicable patents
4. Can anyone develop DisplayPort enabled devices with a signed licensee agreement & patent license: Yes
PS: I could care less about this, but you're blowing a lot of hot air over something that isn't necessarily true.
That's the same argument that I have about consoles in general vs. PC gaming. Suck it up. You can't have a happy medium between the best development platform vs. financially success. I know the Wii will never play games that are capable on PS3/360, just as I'd know that there are games that PC's games that will never play properly on PS3/360's. That's life.
I don't necessarily disagree with what you're trying to convey, but your approach needs some work:
"Microsoft would tend to disagree with you, I'm sure. They managed to corner the market with a free product, in the case of Internet
Explorer. Before you also start bleating at me about how they were leveraging their monopoly, realise how well Google are doing in the market on search."
Netscape was an ok web browser and when IE made it to around 4 or 5, they became ok as well. The difference was that IE -was- installed by default on the majority of new desktops (mshtml cannot be uninstalled) and Netscape wasn't good enough to justify users going out to find a new one. This was during a period where more and more casual PC users were entering the PC market to use the Internet. Many of those users would care less what 'browser' they used.
"some of said companies will add value to it (and close it)"
Lets say that Microsoft took vorbis, added some proprietary and hence incompatible hooks to the product then made it part of the default Windows installation, nobody would cry bloody murder? I think the most likely case would be that the 'market' would split with some using microsoft's default broken implementation while others would go with the free open version. Same things happened with web development and Java. In both cases Microsoft embraced and extended to the point of breaking inter-compatibility.
"Corporations are entirely free to make closed, modified derivatives of Apache; again, if Stallman is right, why haven't they been destroyed yet?"
Apache is still alive because of the people contributing to it. If a closed source product (like MS) wanted to embrace and extend Apache, that's their decision. But apache is itself an implementation of a specification, not the specification itself. The only benefit someone has of taking Apache code would be to save development costs in rolling out their own Web server. I'm sure there are a ton of companies that have done just that. If your point is to develop code that people will use, mission accomplished. If your goal was to foster your personal philosophy of software without commercial value, you're probably on less steady grounds.
"how come any of the BSDs still exist at all?"
BSD is a bad example, since its lost favor for the vast majority of developers. Linux could be considered the slightly younger, more attractive platform to work on, so people jumped ship to Linux, but nothing stops BSD from living on forever. That's not to say its a success story in modern times vs. Apache which still maintains a very high relevance.
"why didn't Microsoft destroy the WC3 after they acquired Internet Explorer?"
They had a monopoly but not exclusive monopoly on the market. That's why they embraced and extended their browser to include proprietary hooks like ActiveX and many non-standard coding extensions. If given enough time, they would've completely tossed out WC3 from consideration. Lucky for us, Mozilla took over the sinking Netscape product line and saved our ability to choose web platforms.
"Internet's protocols still exist, rather than single, monopolised implementations?"
IPv4, TCP, UDP, all De-facto standards that EVERYONE uses. Maybe you should pick a better example, because this one is 100% inaccurate. If you're trying to come up with 'proprietary' protocols, you would be best to look at Instant Messangers, some streaming media, some email provisions and SMB. Beyond that, I can't recall many protocols that don't have proprietary hooks that limit their use. The major reason why protocols aren't proprietary is because the protocols aren't copyrightable in many(all?) countries. If its not illegal to copy, someone could analyse and reproduce said function and publish specifications for how it works without legal reprocussions. Once that's done, the protocol can be implemented by any who care to use it.
Things like Cryptography and Compression which are process intensive are already implemented in native hooks in the base JVM. For instance, if you're using a GZIPOutputStream to GnuZip compress your data, the actually compression is done in a native block. Just because -a part- of your code requires high thoughtput, it doesn't mean that the entire piece of code needs to be developed in said language. Think of it like the distinction between C and ASM. You'd never write your entire code in assembly anymore, but there are some algorithms/loops that you'll do a better job of optimizing by using assembly instead of your general purpose C compiler.
I think you're looking at deterministic in a different way than I. I know that when there are no longer any references to an Object, that object is completely out of scope from my application. I don't care how the system I'm on cleans it up, the same way that 99% of the time, you don't care about what a free/delete/close really does behind the scenes. What does matter is that:
1. The process of cleaning up this memory doesn't negatively impact the application's runtime performance
2. 'Native'/OS resources outside of the code are released in a timely manner (Sockets, shared memory, etc..)
Point 1 is addressed in several different GC's in Java. Most happy cases where fine grained timing is not mandatory have very little issue with one's code unless you're creating/destroying a LOT of objects in which case, object reuse would be more meaningful. Although it may not be perfect, I think Java does a pretty good job of mitigating the GC's interruption of an application.
Point 2 is where the real problem of non-finalization occurs. If one doesn't explicitly release these resources when a programmer is done with them, you'll start getting error for too many open files, or this or that. In this case, it should be up to the developer to specifically deallocate these pieces of code. In any Input/Output steam for instance, a user must explicitly call close() to free up those OS resources that have been allocated for them (This is done on finalization as well if you don't explicitly call it). I don't care about the memory being used, since the GC magically takes care of that for me. What I do care about is that the files are released.
Java does allow one to free these resources when you tell it to, so even though I could say its a non-issue, this brings up an interesting dilemma as a Java developer. If I expect that the GC takes care of everything behind the scenes, I may be coaxed into forgetting about this behaviour when I'm developing my code. Generally, an issue like this is either caught right away which is fine, or its caught much later in the development when the programs are being tested excessively, or when performing load testing, etc.. This just re-enforces how important it is to profile your code with a good profiler to see what's staying in use and what gets freed up. Most Java 'memory leaks' I've seen fall into the annoying but not game-breaking category, but you'll eventually see degradation in performance the longer the issue is left undetected.
that isn't in XP, hence nobody cares. You'll have the what, 30% market segment with Vista, and maybe 10% that are regular gamers who will be using this.
This will just encourage the further brokenness that Windows is turning the PC gaming platform. Good Job!
PS: Before everyone jumps in to say that everyone will jump into Win7, I think you're mistaken. The only way Microsoft will kill XP for most existing users would be to introduce a critical bug that they choose not to fix. I played with Win7 for a few days and can safely say that it doesn't add anything that I've ever wanted to use that a trivial search for google wouldn't find an as-good or better alternative. And maybe its just me, but pretty much every single UI 'enhancement' since circa Win2k is always a step backwards in terms of -my- productivity.
Its lucky that I'm Linux competent since Fedora/Gnome makes practically everything I need easy and uncluttered. If the barrier for entry was a little lower, I could see mass exodus potential coming as XP users take an honest look at what they -really- want to update to.
yeah, they have the ability to dump full heap/stack dumps on really bad errors if you enable them, but typically Java tells you (generally if not specifically) where your errors occur unless you're doing something really funky, or your error happens to be OutOfMemory, in which case the memory 'leak' could be difficult to trace without a good profiler like jprofiler.
Two points:
1. Peer association break simple reference counting. If you're doing non-trivial OO coding, you'll almost certainly have cyclic associations throughout
2. Unless I'm missing something, the counter itself needs concurrency locking which may eventually become a big deal if you're dealing with a large thread count applications of this object.
Looking at the wikipedia article on GC's it mentions a blurb on ref counters that seems to echo some of this sentiment.
http://en.wikipedia.org/wiki/Garbage_collection_(computer_science)#Reference_counting
http://en.wikipedia.org/wiki/Reference_counting
In practice, you (a standard developer) never need to see the raw memory byte level contents of an object. That said, debuggers and profilers can and do get raw access to objects, but they use the many debugging/profiling hooks (JVMTI, JVMDI, etc..) into the JVM to do so, not by using a standard top level core API.
All of what you said is possible and much more once you have native code interface access... but, one cannot run native code (that hasn't been signed by for the JVM itself) unless the security context has been set to allow basically all operations. Since the native hook (JNI) allows for any use of C libraries within the processes code space, an application with native hooks looses all sand-box restrictions. Of course thats why a java applet on the internet requires a popup in order to elevate sandbox privileges to that level.
I don't see how you could reasonably graft on moded text editor bindings onto a modeless IDE, but by all means, write a plugin if you like. What I do like (though rarely use) is that basically every operation you could want to use within the IDE can be bound to any set of key combinations. We're talking several hundred unique operations key bindable. Of course the number of actual commands depends on what features/plugins you've installed.
I never use it for development, just text editing when I'm on windows, but this is probably the best free text editor that I've seen in a long time. Props to them.
Since I don't really know the answer, I'll ask: Can you use VC++ without using the platform SDK? If so, then good for all. If not, then your argument is irrelevant. If you can't use VC++ without a product that contains a flawed stack, then both products are broken by association.
>>Does it refactor methods signatures?
>How often do you do that, really? Hell, I have that in VS and I still never use it.
A lot. When you work on projects that are less than perfectly coupled, this moves from handy to necessary.
>>Rename methods, classes, fields or local variables?
>Again, how often do you really do this that a simple search/replace in the current file won't be sufficient? Once again, I have this feature in VS... I never use it.
If I rename/refactor an interface, my work has interfaces that are linked to literally hundreds of files. Its not something you want to risk doing wrong, or by hand. If I wanted to rename a method through eclipse, I would:
- Select a method to rename
- Shift-Alt-R (or use the right click menu
- Type in a new name and hit enter. After a few seconds of crunching, all references have been renamed..
This will find all structural links to that method in all open project that reference my editing project. Each and every reference to the method will get the name change, and life moves on. Through the command line, you could be spending days search/replacing every possible reference to the method in 100 files.
"You mean you have to stop your program to fix the function you're in the middle of?"
When I started switched my Java IDE to Eclipse, this was one of the big OMG moments. This single feature saves me so much time, no other argument could sway me away from using it. Beyond that, there are so many features packed in there that I can't even imagine developing without it any more.
7. Relays
8. Watch the movie again. The Marcus unit couldn't beat a T-800 alone. Secondly, after the 'Marcus incident' Skynet probably mothballed the entire concept. Instead it simply used the skin tissue from humans (with some slow regenerative or preservative qualities) that had not higher order facilities.
9. Not so much a goof as it is a large leap of faith for the writers
10. It was a lie. Skynet planted the idea in the resistance's head which is how they ultimately found the command/control sub
11. This is a gaping plot hole. Marcus -turning off the defense perimeter- moment when connecting to the central computer just means that either Skynet is moronic for leaving them disabled, or its a big gaping plot hole.
All things said, I loved the movie. It kept me going the whole way, and much like Die hard 4, it wasn't like the originals, but it wasn't bad in its own unique way.
4. *shrugs* never really noticed the bikes power source. Maybe nuclear cores were a new feature for the T-800's. Since he knew about the cores from T3, it wouldn't even be a logic gap that he knew all about them before even meeting a T-800 in person. (The larger flaw would be that there weren't any other terminators to assist the lone T-800 in killing John/Kyle)
5. Her character, just like the TV series would've reached the point on, is the only salvation for humanity against Skynet is to embrace the machines as more than just enemies or tools, and to reach a mutually agreeable outcome with those that can reason. Moon's empathy for Marcus show us that there are those in the Terminator universe that could see the machines (Marcus at least) as more than just a machine, and to treat him like a person/sentient. Marcus' ultimate sacrifice would be quite flat without his humanization in the eyes of the resistance.
6. You could be right, but the ocean's a big place, and even focusing an effort around continental USA wouldn't be a trivial feat. Doubly so if the submarine was in any way modern in disguising its sonar signatures.. *Shrugs*
0. Getting their act together as per Terminator 3, many were able to start organizing -while- judgment day was happening. This may be a little far fetched, but still wasn't it set years after judgment day?
1. According to their explanation, Skynet has strong zones of influence in major cities (as per all previous movies), but when battling in open terrain, humans seem to have the upper hand, or at least they're fighting an effective guerrilla war.
2. Major plot hole. Even if they were informed through the original Terminators Chip (assuming the T1 ending and not the subsequent rewrite of history in T2) they still wouldn't know what he looks like. The fact that Cyberdyne (USAF bought them later) were the ones who developed Marcus Wright, it would allude to the fact that they were very much in business.
3. I thought that was pretty evident in two points:
a. They needed real tissue to wrap all but Marcus' infiltrator terminators which were portrayed in T1 as the Arnold unit. Secondly, the infiltrators didn't all look like Arnold, which is why they eventually needed dogs to sniff them out (sad there weren't any dogs in the movie)
b. They were baiting John to raid Skynet to free the prisoners before the joint strike. I think the Kyle Reese thing was probably changed in a later draft not realizing how stupid a slip-up it really is (This from the people that thought nobody would care if a Terminator could infect and control standard cars what have no useful CPU functions remotely...)
just has sour grapes because they couldn't figure Linux out on their first (or second, or third..) go around and now they're spewing why it doesn't count. That being said, 1% is a pretty pathetic number. Lets hope this new found high point only leads to better things in the future. Maybe they'll be the proverbial Mozilla of an IE / Netscape consumer market struggle. We all know how that played out now, don't we?
haha just watch "The Corporation". It was a great movie, even if you love our corporate overlords =)
.. I know that aren't power users or in the computer field couldn't tell you the difference between their computer and the OS. Just like Mac's. You're buying a mac, not Windows computer.
They DO use ratings from torrents and DVR's for I believe two weeks after the original viewing of the show. On DVR's, If you watch it after two weeks, it doesn't register as a viewing. I think if you download the torrent within two weeks of the original showing, it is considered a viewing.
Unless there's a lot of prop reuse, I doubt it makes much financial sense to go down the road of mini-series. You might as well make a 'cheap' movie that's the same length because you'll at least get a few chumps/rubes/guinea-pigs to pay > $10 to see it.
You're right. Skynet's conciousness was created through a virus, or a random -whatever- online, and after opening the defence network's flood gates to secure the virus, the reverse happened, and the virus all of a sudden had real physical presence in the world. After the nukes, who knows what, but I'm hoping Salvation will enlighten us all *hoping its better than T3*.
PS: T3 was flawed like pretty much all of the 3's sequels the last few years. Their ambition to always out BIG an audience got in the way of telling a comprehensive story. The story in my eyes did have a good core, but it got muddled up with *insert some new antagonist* which was the worst part of the movie, and stepped awkwardly through the conner romance. All in all, it just didn't flow like it should have.