What's more, that was five years, starting with Netscape's codebase. What should they start with now, OpenOffice? If so, why are they more qualified to handle it than the groups currently working on OpenOffice/LibreOffice?
If they're truly static, HTML+CSS is good, though I find I prefer Haml syntax to HTML, or just a template in general. Even "static" pages, and even with CSS, having the ability to loop or refactor a repetitive bit of markup into a partial or method makes it massively better than straight HTML.
Of course, if they're truly static, I'd cache them...
But I think my point stands -- passing on something merely because it has the words "Web app" and "Java" in the job description doesn't make a whole lot of sense. I would avoid anything with "Java" in the name, but that's only because I can think of other languages I'd much rather be using -- Ruby, Python, Lisp, or at the other end, C or assembly. If the problem is the number of technologies combined, it seems you'd avoid web apps in any language, not just java. I don't see anything particular about the combination of "Java" and "Web app" that makes it worse than either by itself.
I wouldn't buy 8 gigs of RAM, so I could spend that money on something else, and, as you suggested, because I have a laptop -- 8 gigs would've been significant at the time I bought this, and 4 gigs was the most this model could take.
I don't use hibernate much, I mostly suspend to RAM. I have a separate server to seed if I really want it, but I find torrents and such work much better if I'm on a private tracker. I find it's much more convenient to have at least a notebook and a server, and I'm thinking I may get a desktop eventually.
But even that server, hibernate isn't out of the question. Could be triggered by a UPS, or it could just be that I want to unplug it and move the power cable. If I hibernate and restore quickly enough, connections might not even drop.
So you can use that 8 gigs for something else, or not buy 8 gigs in the first place. In particular, so when a program is using large amounts of memory for no good reason, it can be swapped out, maybe even just for disk cache.
I think that's a legitimate response -- you asked for "playable games", and there are many, many examples of games that either work well under Wine, or have native ports. Maybe I'm weird, but I actually have at least a few good, quality, DRM-free Indie games I've bought for Linux, and simply haven't had the time to play.
I keep a Windows install around for the exceptions, but they really are the exceptions these days, and I avoid buying new games that don't work on Linux.
An interactive information environment for accessing, controlling, and using information. Using a computer, available sources of information are accessed, and components are extracted, labeled, and formed into discrete units called contexts. A user selects and rearranges context labels and their associated contents. Contexts are selected and combined into new information structures called alternates, which are combinable with contexts into preferred situations. The preferred situations in turn are combinable with the foregoing components into meta-situations. All components have labels; labels and their associated contents are interchangeably movable and copyable at the levels of these information structures, whether they are located locally or remotely, and the information structures are combinable. While a label is invoked and manipulated, its contents or description is simultaneously displayed. Each information structure can be rearranged into one or more models which can be displayed by user selection, and models can be displayed at varying levels of detail. With built-in copyright accounting, commercial control remains with information owners, while operational use is centralized in each user.
WTF does this mean, and WTF does it have to do with rollovers?
Depending on how you define "free will", i think it would be possible for a program to have free will.
Unless you are assuming an unreliable computer, or some other source of entropy, or some bizarre definition of free will which is compatible with determinism, I don't see how a program can have free will. Even under the first two cases, it wouldn't be the program which potentially has free will, but the source of entropy.
the randomness/chaos at the quantum level could very well be a source of free will,
Except we don't know it's random, first of all.
Regardless, it does nothing for the philosophical problems with free will, which essentially boils down to: Is a "free" decision by an agent determined by the state? In other words, are your decisions determined by the situation and your circumstances? If not, you're a lunatic, and you'd act randomly. If so, it's not free will, it's by definition determinism. If it's some combination, it seems the result is simply that your decision-making process is inaccurate, and sometimes leads to decisions it shouldn't.
Also, quantum is pretty irrelevant, such that philosophers now talk about "weak determinism", the view that human actions are completely determined by past events and the laws of nature. If you plug random quantum events into, say, an image generator, you're going to get static noise, and the structure of the human brain is such that they can hardly have an effect here, either.
Indeed, if randomness is the source of free will, then you should be able to write a program that appears to be human and appears to have free will, and is hooked into a random number generator at some point -- perhaps/dev/random. But you're assuming/dev/random is actually random -- in fact, unless you explicitly add some sources of entropy (like some sort of quantum sensor), even/dev/random is pseudorandom. Given two computers in the same state, without an external source of entropy, and/dev/random will produce exactly the same data in each. Its apparent randomness relies on external entropy and the assumption that modern computers are large and complex enough that no one can completely predict their internal state, especially when you take networks and other humans into account, but it's still deterministic.
And if you can make two programs, one which uses a pseudorandom generator, and one which uses the real thing, and you can't tell the difference, is it important that one has truly "free" will? That would imply that free will has pretty much nothing to do with our experience of consciousness, freedom, choice, etc.
but i don't really want my creation to have free will, i want it to be an extension of my own.
Which just pushes the problem back a step.
if we do have free will and i don't program my program to have it's own free will, then as i view it, it is still an extension of my will, and i can still have it accomplish tasks after i am gone. if i do have free will then it is my choice to write the program, and thus i am indeed giving myself a measure of immortality.
Interesting. I think I made some assumptions about what you meant by immortality.
But this kind of immortality -- very similar to the kind Beowulf was after (fame) -- doesn't rely on free will of any kind. After all, whatever you are, whether "free" or determined, it is still your creation which lives on. But then, I'm not sure I see that code is more a form of immortality than children, unless you plan to upload yourself eventually.
I certainly don't think the discussion is pointless if there is no such thing as free will, or if there is no such thing as a god. Without a god or free will, your immortality still works (to the extent it does at all), and it has very little impact on our lives here. (The belief in God and free will can have a profound impact, but the reality seems to do very little.)
And my point was, if you don't have a "pure Java" application, and if you are making assumptions about both the platform you're running on and the JVM you're running on, it actually seems like an improvement to switch to JNI, which only makes assumptions about the platform.
After all, C does have pretty consistent signal handling where that concept makes any sense.
If your "programs" are just acting out your own will, then, by your metaphor, if there is a god, we're merely acting out his will -- in other words, we don't have our own free will. How can we then be held accountable for anything we do? When a program doesn't work, we don't blame the program, we blame the programmer.
And while we may not be able to prove that something has free will, since free will is so ill-defined in the first place, we can certainly make some educated guesses.
That seems to be specifically about taking the name of God "in vain", which is loosely defined at best. Then there's "swearing", so don't swear false oaths. Finally, the stuff about "cursing", in context, seems to be talking about the kind of curse that's meant by "a pox be upon you" -- it's not about censoring certain words, it's about what they mean.
Yes, most overtly Christian people are opposed to foul language. Most overtly Christian people are also opposed to stoning and genocide, both of which are not only encouraged but sanctioned by God in the Bible. Basically, they take whatever they want to believe, and build a rationalization which includes some Bible verse or other -- just like the site you linked to does. Look up Jesus and the fig tree -- WWJD? He'd curse a fucking fig tree, that's what.
Keep in mind, people on both sides of the slavery debate used Bible verses to justify their positions. If the Bible has anything to say about slavery, it's that slavery is OK, just don't beat your slaves to death (but it's ok if they survive a few days and then die).
You've just stated Pascal's Wager. A trivial refutation: If you wait to find out on your own, it may be too late -- if you've chosen the right god. Suppose there's a different "programmer", or a team of programmers? Wouldn't they be more offended by you worshipping the wrong god than waiting to see which is right?
And I don't know about you, but I don't consider my programs to have free will, so I don't see how, if we're God's "programs", we have any free will of our own. And what if God is himself a program?
So now I'm curious -- you've just pointed out a potential problem and a solution which didn't require going beyond the standard API. It also seems like unless your library is incredibly poorly written, this would be a trivial change.
I suppose if I really, really needed signal handling, I'd do JNI, which would be incredibly ironic -- but really, if you're deliberately using features you know aren't supported on all Java platforms, you shouldn't be surprised when they don't work on all JVMs.
I think Firefox is altogether an excellent browser, but I think they've made some remarkably stupid decisions with regards to HTML5. That doesn't make all of Firefox "stupid software."
Similarly, I would imagine Tomcat is mostly well put together, but has made some stupid assumptions about the JDK it's running on. I'm also curious if it still has those issues...
The Microsoft tactic was to embrace, extend, and extinguish Java.
Google wants to use some of the Java stuff, build on it, and adapt it to their own platform.
Do you see the difference? Google isn't trying to kill portable Java, and isn't claiming that Android is a portable Java. Microsoft was pretty much deliberately trying to head off "compile once, run anywhere" by letting people develop what they thought was portable Java, discover it would only work on Windows, and then shrug and avoid other platforms.
There's an important legal distinction, also -- the Microsoft thing is a trademark dispute (they were actually pretending it was Java), while the Google thing is a patent dispute, which would be as if they sued Microsoft for.NET.
Oh, and by the way, I can do portable software in HTML5.
If I understand you, I could wait for a WM7 phone which has everything you mention, or I could just get it now in an Android phone.
So why should I wait, and why should I trust Microsoft over Google? And, for that matter, why would Apple be afraid of Microsoft's vaporware, when they can be afraid of Google's reality?
It already is a drop-in replacement, unless you're dealing with software that makes remarkably stupid assumptions about the JDK it's running on.
Unfortunately, that may be a lot of software -- I know Oracle's own JDeveloper uses some internal Sun JDK stuff, when there's no reason they couldn't use the standard public API for the same thing which OpenJDK also supports.
Still, if it's in your power to do so, fix the app. If OpenJDK breaks it, chances are, a future Sun JDK will break it, too.
I don't see the economic incentive for Oracle to keep this project,
Maybe because Oracle, being enterprise-y, has an absurd number of applications which run on Java? Improving Java performance means nearly all Oracle applications run faster. Making Java more flexible as a language and as a VM means they have more powerful tools and better techniques going forward, which they can use for developing things which plug directly into all that legacy Java code they've got.
And while Oracle certainly has the rights to close as much of it as they like, hopefully even they realize it's in their best interests to collaborate with the community (including IBM), rather than trying to go it alone.
I'm guessing the bulk of the Dev work is transitioning to IBM.
And why do you think IBM has a better incentive than Oracle?
Slashdot Mirror
What's more, that was five years, starting with Netscape's codebase. What should they start with now, OpenOffice? If so, why are they more qualified to handle it than the groups currently working on OpenOffice/LibreOffice?
...provided the textbooks are provided to me in un-DRM'd PDFs, or something similar.
Try to foist some "e-reader format" on me, or force me to use Windows or OS X to read them, and I'll find another university.
If they're truly static, HTML+CSS is good, though I find I prefer Haml syntax to HTML, or just a template in general. Even "static" pages, and even with CSS, having the ability to loop or refactor a repetitive bit of markup into a partial or method makes it massively better than straight HTML.
Of course, if they're truly static, I'd cache them...
But I think my point stands -- passing on something merely because it has the words "Web app" and "Java" in the job description doesn't make a whole lot of sense. I would avoid anything with "Java" in the name, but that's only because I can think of other languages I'd much rather be using -- Ruby, Python, Lisp, or at the other end, C or assembly. If the problem is the number of technologies combined, it seems you'd avoid web apps in any language, not just java. I don't see anything particular about the combination of "Java" and "Web app" that makes it worse than either by itself.
Out of curiosity, what would you use to write a web app in?
I wouldn't buy 8 gigs of RAM, so I could spend that money on something else, and, as you suggested, because I have a laptop -- 8 gigs would've been significant at the time I bought this, and 4 gigs was the most this model could take.
I don't use hibernate much, I mostly suspend to RAM. I have a separate server to seed if I really want it, but I find torrents and such work much better if I'm on a private tracker. I find it's much more convenient to have at least a notebook and a server, and I'm thinking I may get a desktop eventually.
But even that server, hibernate isn't out of the question. Could be triggered by a UPS, or it could just be that I want to unplug it and move the power cable. If I hibernate and restore quickly enough, connections might not even drop.
I have 8GB of RAM, why do I need swap?
So you can use that 8 gigs for something else, or not buy 8 gigs in the first place. In particular, so when a program is using large amounts of memory for no good reason, it can be swapped out, maybe even just for disk cache.
Also, hibernate.
What part of Android doesn't "just work"?
Puritanical, then. The "no porn on the App store" certainly isn't a Buddhist ideal.
I think that's a legitimate response -- you asked for "playable games", and there are many, many examples of games that either work well under Wine, or have native ports. Maybe I'm weird, but I actually have at least a few good, quality, DRM-free Indie games I've bought for Linux, and simply haven't had the time to play.
I keep a Windows install around for the exceptions, but they really are the exceptions these days, and I avoid buying new games that don't work on Linux.
If that was too long, here's the same thing in comic form.
Help me out...
An interactive information environment for accessing, controlling, and using information. Using a computer, available sources of information are accessed, and components are extracted, labeled, and formed into discrete units called contexts. A user selects and rearranges context labels and their associated contents. Contexts are selected and combined into new information structures called alternates, which are combinable with contexts into preferred situations. The preferred situations in turn are combinable with the foregoing components into meta-situations. All components have labels; labels and their associated contents are interchangeably movable and copyable at the levels of these information structures, whether they are located locally or remotely, and the information structures are combinable. While a label is invoked and manipulated, its contents or description is simultaneously displayed. Each information structure can be rearranged into one or more models which can be displayed by user selection, and models can be displayed at varying levels of detail. With built-in copyright accounting, commercial control remains with information owners, while operational use is centralized in each user.
WTF does this mean, and WTF does it have to do with rollovers?
Depending on how you define "free will", i think it would be possible for a program to have free will.
Unless you are assuming an unreliable computer, or some other source of entropy, or some bizarre definition of free will which is compatible with determinism, I don't see how a program can have free will. Even under the first two cases, it wouldn't be the program which potentially has free will, but the source of entropy.
the randomness/chaos at the quantum level could very well be a source of free will,
Except we don't know it's random, first of all.
Regardless, it does nothing for the philosophical problems with free will, which essentially boils down to: Is a "free" decision by an agent determined by the state? In other words, are your decisions determined by the situation and your circumstances? If not, you're a lunatic, and you'd act randomly. If so, it's not free will, it's by definition determinism. If it's some combination, it seems the result is simply that your decision-making process is inaccurate, and sometimes leads to decisions it shouldn't.
Also, quantum is pretty irrelevant, such that philosophers now talk about "weak determinism", the view that human actions are completely determined by past events and the laws of nature. If you plug random quantum events into, say, an image generator, you're going to get static noise, and the structure of the human brain is such that they can hardly have an effect here, either.
Indeed, if randomness is the source of free will, then you should be able to write a program that appears to be human and appears to have free will, and is hooked into a random number generator at some point -- perhaps /dev/random. But you're assuming /dev/random is actually random -- in fact, unless you explicitly add some sources of entropy (like some sort of quantum sensor), even /dev/random is pseudorandom. Given two computers in the same state, without an external source of entropy, and /dev/random will produce exactly the same data in each. Its apparent randomness relies on external entropy and the assumption that modern computers are large and complex enough that no one can completely predict their internal state, especially when you take networks and other humans into account, but it's still deterministic.
And if you can make two programs, one which uses a pseudorandom generator, and one which uses the real thing, and you can't tell the difference, is it important that one has truly "free" will? That would imply that free will has pretty much nothing to do with our experience of consciousness, freedom, choice, etc.
but i don't really want my creation to have free will, i want it to be an extension of my own.
Which just pushes the problem back a step.
if we do have free will and i don't program my program to have it's own free will, then as i view it, it is still an extension of my will, and i can still have it accomplish tasks after i am gone. if i do have free will then it is my choice to write the program, and thus i am indeed giving myself a measure of immortality.
Interesting. I think I made some assumptions about what you meant by immortality.
But this kind of immortality -- very similar to the kind Beowulf was after (fame) -- doesn't rely on free will of any kind. After all, whatever you are, whether "free" or determined, it is still your creation which lives on. But then, I'm not sure I see that code is more a form of immortality than children, unless you plan to upload yourself eventually.
I certainly don't think the discussion is pointless if there is no such thing as free will, or if there is no such thing as a god. Without a god or free will, your immortality still works (to the extent it does at all), and it has very little impact on our lives here. (The belief in God and free will can have a profound impact, but the reality seems to do very little.)
And my point was, if you don't have a "pure Java" application, and if you are making assumptions about both the platform you're running on and the JVM you're running on, it actually seems like an improvement to switch to JNI, which only makes assumptions about the platform.
After all, C does have pretty consistent signal handling where that concept makes any sense.
But which behavior? Actual, specified behavior, or private interfaces no one had any business using?
What difference does what make?
If your "programs" are just acting out your own will, then, by your metaphor, if there is a god, we're merely acting out his will -- in other words, we don't have our own free will. How can we then be held accountable for anything we do? When a program doesn't work, we don't blame the program, we blame the programmer.
And while we may not be able to prove that something has free will, since free will is so ill-defined in the first place, we can certainly make some educated guesses.
That seems to be specifically about taking the name of God "in vain", which is loosely defined at best. Then there's "swearing", so don't swear false oaths. Finally, the stuff about "cursing", in context, seems to be talking about the kind of curse that's meant by "a pox be upon you" -- it's not about censoring certain words, it's about what they mean.
Yes, most overtly Christian people are opposed to foul language. Most overtly Christian people are also opposed to stoning and genocide, both of which are not only encouraged but sanctioned by God in the Bible. Basically, they take whatever they want to believe, and build a rationalization which includes some Bible verse or other -- just like the site you linked to does. Look up Jesus and the fig tree -- WWJD? He'd curse a fucking fig tree, that's what.
Keep in mind, people on both sides of the slavery debate used Bible verses to justify their positions. If the Bible has anything to say about slavery, it's that slavery is OK, just don't beat your slaves to death (but it's ok if they survive a few days and then die).
You've just stated Pascal's Wager. A trivial refutation: If you wait to find out on your own, it may be too late -- if you've chosen the right god. Suppose there's a different "programmer", or a team of programmers? Wouldn't they be more offended by you worshipping the wrong god than waiting to see which is right?
And I don't know about you, but I don't consider my programs to have free will, so I don't see how, if we're God's "programs", we have any free will of our own. And what if God is himself a program?
So now I'm curious -- you've just pointed out a potential problem and a solution which didn't require going beyond the standard API. It also seems like unless your library is incredibly poorly written, this would be a trivial change.
I suppose if I really, really needed signal handling, I'd do JNI, which would be incredibly ironic -- but really, if you're deliberately using features you know aren't supported on all Java platforms, you shouldn't be surprised when they don't work on all JVMs.
That's not what I said.
I think Firefox is altogether an excellent browser, but I think they've made some remarkably stupid decisions with regards to HTML5. That doesn't make all of Firefox "stupid software."
Similarly, I would imagine Tomcat is mostly well put together, but has made some stupid assumptions about the JDK it's running on. I'm also curious if it still has those issues...
In other words, IBM has the same incentive as Oracle, so my point stands.
The Microsoft tactic was to embrace, extend, and extinguish Java.
Google wants to use some of the Java stuff, build on it, and adapt it to their own platform.
Do you see the difference? Google isn't trying to kill portable Java, and isn't claiming that Android is a portable Java. Microsoft was pretty much deliberately trying to head off "compile once, run anywhere" by letting people develop what they thought was portable Java, discover it would only work on Windows, and then shrug and avoid other platforms.
There's an important legal distinction, also -- the Microsoft thing is a trademark dispute (they were actually pretending it was Java), while the Google thing is a patent dispute, which would be as if they sued Microsoft for .NET.
Oh, and by the way, I can do portable software in HTML5.
Any examples you can share?
If I understand you, I could wait for a WM7 phone which has everything you mention, or I could just get it now in an Android phone.
So why should I wait, and why should I trust Microsoft over Google? And, for that matter, why would Apple be afraid of Microsoft's vaporware, when they can be afraid of Google's reality?
It already is a drop-in replacement, unless you're dealing with software that makes remarkably stupid assumptions about the JDK it's running on.
Unfortunately, that may be a lot of software -- I know Oracle's own JDeveloper uses some internal Sun JDK stuff, when there's no reason they couldn't use the standard public API for the same thing which OpenJDK also supports.
Still, if it's in your power to do so, fix the app. If OpenJDK breaks it, chances are, a future Sun JDK will break it, too.
I don't see the economic incentive for Oracle to keep this project,
Maybe because Oracle, being enterprise-y, has an absurd number of applications which run on Java? Improving Java performance means nearly all Oracle applications run faster. Making Java more flexible as a language and as a VM means they have more powerful tools and better techniques going forward, which they can use for developing things which plug directly into all that legacy Java code they've got.
And while Oracle certainly has the rights to close as much of it as they like, hopefully even they realize it's in their best interests to collaborate with the community (including IBM), rather than trying to go it alone.
I'm guessing the bulk of the Dev work is transitioning to IBM.
And why do you think IBM has a better incentive than Oracle?