If you think a piece of music is ugly, chances are that you just don't get it, and didn't try very hard. If you say it isn't music, you're just wrong.
There is big difference between music that is incomprehensible and music that is ugly. My sister is an ethnomusicologist, and I have encountered many forms of music from Africa, South America, and Middle East that are strange and meaningless at first encounter. I have learned to enjoy microtone music. After listening for a while and learning something of the culture behind the music, it begins to make sense, and you begin to see the beauty.
I guess what you're having a problem with is that I do comprehend certain forms of American music, and largely because of what I know about the culture that produced it, I hear an ugly, distorted, maimed version of what it ought to be. Pain, suffering, and rage are a fact of life, but they are not the way things ought to be. Music that screams with rage and pain, is a distorted form of music that shouts with joy and exuberance.
In other words, it is not merely strange and incomprehensible, but conveys a message which I have understood, judged, and found wanting. How shockingly un-PC!
There are creatures that most people find ugly, and yet they breed successfully. Do boy cockroaches find girl cockroaches beautiful because of their corrupted nature?
Perhaps the real question is, do you believe that anything is truly ugly? Or are you convinced that everything is beautiful when properly viewed? Ugliness is beauty spoiled. So we probably agree that everything could have been or even can be beautiful, but I have noticed that many things have their beauty spoiled to the point that they are ugly. I can be sure that this is not due to my own warped perception when I have seen the beauty of its unfallen state.
For instance, suppose a beautiful woman is tragically maimed in a fire. There are still many beautify aspects about her - character, voice, personality, etc. But her burned and scarred face is now objectively ugly. We can still be attracted to her by looking past the ugly face to the more unchangeable qualities behind it. If the woman was burned in some noble sacrifice, like saving her child from the flames, the scars may even symbolize to us the beauty of that deed - but they are still ugly.
But I know that won't happen because people are too busy making yet another toolkit for X or another extension or another weird project with a weird name written all in lower-case on Sourceforge. Meanwhile, in August of 2005, Longhorn is due out, with hardware acceleration, vector-scaled widgets for resolution-independent resizing, a yet-to-be-revealed photorealistic user interface, and even the ability to add and remove RAM without rebooting.
Those are all (except for the last) things I have no use for. And the last is first and foremost a hardware feature. Linux support will come quickly if the hardware is documented. Those weird projects with names written all in lower case are what do useful things for me and my companies customers. We care about getting work done, not about ever more glitzy user interfaces that require expensive hardware upgrades just to make a workstation usable again.
Many of our customers do use Windoze on the desktop. And I do mean doze. I am astonished that they are willing wait 45 minutes (literally and very common) for a mail message to display. Of course, their machines were responsive when they fist got them, but with all the software upgrades, glitzy dodads, (and probably loads of spyware running in stealth mode) the machines gradually slow to a crawl. Then, the users figure the hardware must be "worn out" and buy an expensive new machine. Occasionally I can intercept the discard to use for another LTSP terminal.
Sorry, I don't buy the "beauty is in the eye of the beholder" argument. The orcs in Tolkien aren't beautiful. Presumably, they might be attractive to other orcs - but that would be a consequence of their corrupted nature. (And it seems as if orcs are produced by evil bio engineering rather than relying on mutual attraction.) Finding ugly things beautiful suggests that the beholder's ability to perceive beauty has been marred or bent.
On the other hand, we might simply be disagreeing over the definition of "music" (and other forms of art). If your definition encompasses any form of self-expression, then screaming certainly qualifies. And in fact, listening to such expressions of pain and anguish is something we do in our home in order to empathize with the sufferer. But not as "music". For instance, empathizing with pain and auguish does not help with studying. It's not that the screaming won't get into the home (in fact, we've been known to do it ourselves on occasion), it just won't get in as "music".
If you are truly saying that beauty is relative, then there is no true beauty in your world. Whatever you perceive as most beautiful can never be shared because others might perceive it differently. True beauty is there to be shared, independent of our human limitations. Though the perceiving of it becomes easier with practice.
She is too busy discovering exotic music. She likes Medieval, and we also like the Indian Cello and Sitar albums. All kinds of Music from all parts of the world is fine. However, screaming is not music. A valid expression of emotions of rage and futility, yes, but not music.
She'll have plenty of time to tinker with the philosophy of life we are passing on when she leaves home.
People who don't pass on any moral structure to their kids are not doing them a favor. Stretched analogy time: It is like giving someone a computer with no software at all, and telling them to build their own boot code, operating system, and applications from the raw machine code on up.
A much better way is to give them a Linux distribution. Even if they hate it, they have something that works (hey, it works for me) that they can personalize to their hearts content before passing is on to their own kids.
Like Open Source, philosophy needs to be open. No hidden secrets available only to the elite (Gnosticism = Microsoft). When you understand the logical and historical basis of your philosophy, you have the tools to correct it without starting from scratch.
There is no such thing as a self made man. Or if there is, he won't get very far. We stand on the shoulders of giants.
I found some medieval music I like, but am concerned about how they will handle my CC number. The "buy" link doesn't seem to use SSL.
I have a 14 year old daughter, and she likes modern music. I came up with an objective rule which she understands for what modern music is allowed in our home. Music where the singers scream or the instruments scream (distortion) is not allowed. Shoutin', on the other hand, is fine (Black Gospel).
This still has some false positives (e.g. Jesus Christ Superstar), but works fairly well.
Linux is about choice. People use many different email clients - Mozilla, Netscape, Sylpheed, pine, Evolution, mutt, firebird, etc. People use many different web browsers - Mozilla, Galeon, Phoenix, Opera, Netscape, lynx, links, etc. There are many desktop/window configurations, Gnome, sawfish, KDE, Winmaker, Blackbox, Ice, fvwn, etc.
This is a support nightmare if you are trying to do centralized support - but it is the heterogeneous environment the original poster was talking about. An exploit for Mozilla is not going to take out nearly the percentage of Linux users as an exploit for Outlook does for Windows.
While centralized support is a nightmare - it is unnecessary with open source. You can have many support companies specializing in various application and system choices. They can do this because the source code is not locked away by one corporation.
What I would like is an extremely efficient programming environment that compiles my six line x++ program down to a few hundred bytes total...that's in-memory while running. I want to use my RAM for data and number crunching, not unusable code.
Borland Pascal. And its open source cousin, FreePascal.
Borland Pascal is my low level language of Choice
for the Windows platform (and for DOS before that). When statically linking a library, it pulls out only and precisely only the code and data needed for that application. It worked great for DOS TSR programs (which worked best with minimal memory use).
The catch is, that this approach only works for static linking. If you want to share the library between multiple different applications, the shared part has to be constant. When you get to GUI controls, e.g. with Delphi, there is a large DLL (.so for the linux version). You can still use the smart static linker, but your hello world GUI app is about 100K on Windows, and goes up from there as you use more interface elements.
I don't want to get into the tradeoffs of static vs dynamic linking, but when lots of apps use the same DLL/so, it saves memory and disk space, plus you can upgrade/break all the applications simply by upgrading/breaking the DLL/so. When all apps use a rather disjoint set of the DLL, static linking is better.
On a related note, a lot of the GUI code is loaded with the OS in Windows. Java Swing ignores this and recreates it all in Java. This makes Swing programs much bigger and slower. There are other (non Sun) GUI APIs for Java that use the Windows API directly and are much smaller and faster. C#, naturally, doesn't make this mistake. Java should have stuck with AWT, enhancing its capabilities instead of reinventing the Windowing OS. A mostly Java implementation of AWT would suffice for platforms without a suitable native Windowing system.
in 1975, a fellow student wrote a program that retrieved the entire password list for the system, including admins. The system was an HP2000, and the flaw was that BASIC common memory was not initialized to zero by the system - it contained whatever was left over from another program. So if you ran a program with lots of common variables just after logging in and interpret them properly . . .
He presented the program and the password list to the school administrator. Instead of persecuting him, they contacted HP, who fixed the problem, and HP hired him with a college scholarship as soon as he graduated.
CAUCE is touting Trusted Email Open System as the solution to spam. I've been reading about it, and it sounds like if it becomes a reality, I'll have to buy a "trusted computer" running Windows if I want to send email to anyone using the system. Naturally, Bill Gates is all for it.
I am relying on sendmail to do the right thing protocol wise. If you are correct, then sendmail probably collects the remaining data before rejecting and discarding it - so I am not actually saving any bandwidth. Which means that I should probably scan the entire message, not just the headers. Arrgh.
SMTP was just not designed with spamming scum in mind.
I run a Python milter for sendmail with DSPAM header triage. It has been REJECTing messages at the rate of over 1000/hr for the last 2 days after receiving only the headers. The filter is automatically trained by the few that get through to my mailbox. I couldn't have survived the last few days (at least while getting any work done) without it.
Using the milter interface, messages can be rejected anytime. While rejecting extremely spammy mail is worthwhile, it is important to allow mail for which there is some doubt through in case of false positives. I suppose the milter could fully receive the mail before rejecting it, and *still* deliver it to the user for feedback.
I know of companies where the *only* available ISP (other than dialup) is spam-friendly. To run a business, you have to have broadband (such is the new millenium), and they have you over a barrel. For one company in Tulsa with this situation, we tried getting two 56K dialup accounts and trying to distribute the traffic. It wasn't very satisfactory. The dialup accounts disconnect every 6 hours or so regardless of traffic.
These ISPs are rightly listed by SPEWS - but have a monopoly. (No, satellite is not suitable.) To be able to send mail in these situations, we relay it through a machine on another ISP.
As the father of four, I am very pleased with Bob the Builder as the "Ideal Role Model". Bob occasionally makes technical mistakes (he can't hang wallpaper to save his life - but Wendy is a pro), but he never ever loses his temper or yells or gets discouraged. This is boring for adults, but a good model of the ideal for children - actually I watch it too. I'm just checking to make sure the ideals presented match mine:-)
Why does a sinless hero lose interest for adults? I think the main reason is that having screwed up a lot, we need someone to model how to patch things up again. For instance, when Bob spends hours creating a trumpet from spare pipe, and Rolly flattens it, Bob does not get mad. He just says, "It's OK, you didn't see it, Rolly." Not me, I'd be livid. And Bob is no help showing me how to patch up the relationship afterward.
The main disadvantage from my perspective is price. Second would be vendor lockin - except for the Java, anything you develop can't be easily used anywhere else. The platform is evolving. Perhaps it will evolve into a kick-ass networked enterprise Java with database box - where you don't know or care what OS runs under the covers. Just like you don't know what processor runs under the covers now.
Disclaimer - I've never actually played with an AS/400 due to the price:-) I just read about it and like many of the design points.
Give the guy a break. What he means is, "you can't do embedded with Java II enterprise edition". He just doesn't consider those realtime and subset specifications to be real "java". Computer science types like a consistent vocabulary. The language and bytecode might be the same, but the libraries and environment are very different between the Micro and Enterprise editions. So how can they both be "java"?
IBM's AS/400 JVM does exactly this. Java byte code is compiled the first time it is run and optimized for the exact processor and OS version. The binary is managed transparently by the JVM and updated when the bytecode changes. On AS400, this is all integrated with the filesystem (i.e., you can't screw it up by touching with a funny date).
You get all the advantages of Native compilation, and system specific JIT compilation combined - at the expense of more complexity and lots more disk space.
As others have mentioned, M$ is attempting to do a similar thing with.NET. Whether they can do as good a job without charging AS/400 prices remains to be seen.
As wonderful as all this is, I still like direct bytecode execution because it minimizes memory use and startup - which is more important than CPU for many applications (business logic and embedded). Native compiled code is quite a bit bigger than bytecode, uses more memory, and takes longer to load (though not as long as JITing the bytecode).
I have to pay 20% more at CVS because I refuse to sign up for their "bonus card". I'd have to pay more at Giant too, but we managed to get cards with bogus info.
It may be inevitable that "none shall buy and sell except they have the mark of the beast", but that doesn't mean we should welcome it with open arms.
Fortunately, there are still merchants like Shoppers Food Warehouse and Trader Joe's in our area that don't require cards. In fact, SFW even advertises "No Card Required!".
Assuming your post is serious, by recharging, I reuse the alkaline batteries a dozen or more times. This reduces the amount I throw out by at least a factor of ten. I do check on the charger when I walk by and stop the charging when I notice a leak beginning to appear, but I have a life and don't sit in front of the charger staring at it.
I don't like NiCad because the Cadmium is so nasty. I will look into the NiMh and other rechargeables. Presumably they are no more toxic than ordinary alkaline.
I bought some rechargeable alkaline batteries and a charger. They recharged a few times, then leaked. Since the whole reason to pay a premium for rechargeable alkaline was that they wouldn't leak like regular alkaline, I put some regular alkaline batteries in the charger. They might leak a little more often than officially rechargeable, but not much.
So now I just buy regular alkaline and recharge them until they start leaking, then throw them out. Recharging before deep discharge gives the longest battery life for alkaline. I had a set in my RC car that I recharged after every play session, and they lasted for two years (with several sessions a month).
If batteries leak badly, I clean up the discharge with white vinegar and baking soda.
Python lists have the functional equivalent of head and tail. What you probably mean is that you want to keep a reference to the tail and have the head get collected if not referenced elsewhere. Because Python lists are implemented as arrays, the whole array stays in memory as long as any part is referenced.
Python already has full garbage collection in both the C and Java implementations. The reference counting accelerates collection at the expense of runtime in C python. Since the reference counting overhead is fairly small compared to the overall interpreter overhead, it is a good tradeoff. Memory use is minimized, and circular references still get collected by the general collector. The Java implementation does not use reference counting, but relies solely on the Java runtime for GC.
Thanks to reference counting in the C implementation, you can write:
buf = open('infile','r').read()
and the file is closed as soon as it is read into buf. However, this is not recommended practice for maximum portability, since the file will not get closed right away in the Java implementation and future C Python version might drop reference counting.
While the python threading model does not easily take full advantage of multiple CPUs, it does make it easy to incorporate C code - even C code that is not written with threading in mind. Java supports full threading, but you have to be much more careful with JNI extensions. With python, all other threads are stopped when calling into C until you tell Python otherwise. So by default, non thread aware code runs just fine.
What the concept needs is a database by issue. For each issue, companies are given a score from 0.0 (demonic) to 1.0 (angelic). Issue databases would be maintained by different groups. Issues would have names like 'PETA' (no animals were harmed in the making of this product), 'kosher' (if animals were killed, it was done humanely, and you're not eating anything made from swine or snails), 'gambling', 'pro-life', 'pro-choice', etc. Users would subscribe to and weight issues of their choice.
In addition to the summary score, the device should display the top 10 or so issues contributing to the summary score.
Some issue sources would be meta-issue sources combining multiple issues into a pre-weighted ethical ranking, e.g. 'Vatican' or 'Steinem'.
BTW, how does one go about testing the rightness of an ethical standard? Or is it an arbitrary individual choice? If the latter, is there any good reason why anyone should regard Saddam or Bush or Clinton as 'immoral'?
When I buy a computer, I always go for ECC memory, cooler CPUs, disk mirroring, and as many other reliability features as I can afford. The only time I have ever had Linux crash on such machines is with those cheap 10G IDE tape drives on ServerWorks. And that is fixed by turning off DMA for the tape. We have dozens of Linux machines of various manufactures in the field, some as servers, and some as LTSP multiuser GUI systems.
However, I have convinced several family and friends to switch to Linux on their desktops purchased for Windows. Part of the motivation was that Windows crashed constantly. Well, guess what? Linux crashes too. They say it crashes less often, but I suspect that is subjective. These machines have non-parity memory, hot CPUs, even hotter graphics cards, and no backup or disk mirroring. These crash-prone machines are generally 3 times the price of a Dell business server with backup (probably due to the high end 3D graphics and sound missing from a business server).
The sample size is small (3 desktop users of Windows and Linux), but they consistently rank OS reliability (on the cruddy hardware just described) like this:
Windows ME - absolutely the worst. Don't plan to run for more than an hour.
Windows 9x - better
Windows XP - a big improvement
Windows 98 running under Win4Lin - even better
RedHat 7.3 - the best
As a result, they have all settled down to running mostly Linux and Win4Lin, and booting XP for the occasional Windows only multimedia app that won't run under Win4Lin.
Re:This is a lot of work - have you read the LGPL?
on
LGPL is Viral for Java
·
· Score: 1
In Java and other dynamic languages like Smalltalk or Python, it is actually not very complex. It is standard procedure. All major APIs use a plugin architecture to allow changing implementation via configuration or on the fly. This makes systems must more robust and upgradeable. JDBC, Javamail, Java Cryptograpy Extensions, Java Message Service, you name it - it uses a plugin API. You can plug in GPL, LGPL, Proprietary, Sun Reference Implementation - all without affecting the license of your application, whether opensource or proprietary.
Slashdot Mirror
There is big difference between music that is incomprehensible and music that is ugly. My sister is an ethnomusicologist, and I have encountered many forms of music from Africa, South America, and Middle East that are strange and meaningless at first encounter. I have learned to enjoy microtone music. After listening for a while and learning something of the culture behind the music, it begins to make sense, and you begin to see the beauty.
I guess what you're having a problem with is that I do comprehend certain forms of American music, and largely because of what I know about the culture that produced it, I hear an ugly, distorted, maimed version of what it ought to be. Pain, suffering, and rage are a fact of life, but they are not the way things ought to be. Music that screams with rage and pain, is a distorted form of music that shouts with joy and exuberance.
In other words, it is not merely strange and incomprehensible, but conveys a message which I have understood, judged, and found wanting. How shockingly un-PC!
Perhaps the real question is, do you believe that anything is truly ugly? Or are you convinced that everything is beautiful when properly viewed? Ugliness is beauty spoiled. So we probably agree that everything could have been or even can be beautiful, but I have noticed that many things have their beauty spoiled to the point that they are ugly. I can be sure that this is not due to my own warped perception when I have seen the beauty of its unfallen state.
For instance, suppose a beautiful woman is tragically maimed in a fire. There are still many beautify aspects about her - character, voice, personality, etc. But her burned and scarred face is now objectively ugly. We can still be attracted to her by looking past the ugly face to the more unchangeable qualities behind it. If the woman was burned in some noble sacrifice, like saving her child from the flames, the scars may even symbolize to us the beauty of that deed - but they are still ugly.
Those are all (except for the last) things I have no use for. And the last is first and foremost a hardware feature. Linux support will come quickly if the hardware is documented. Those weird projects with names written all in lower case are what do useful things for me and my companies customers. We care about getting work done, not about ever more glitzy user interfaces that require expensive hardware upgrades just to make a workstation usable again.
Many of our customers do use Windoze on the desktop. And I do mean doze. I am astonished that they are willing wait 45 minutes (literally and very common) for a mail message to display. Of course, their machines were responsive when they fist got them, but with all the software upgrades, glitzy dodads, (and probably loads of spyware running in stealth mode) the machines gradually slow to a crawl. Then, the users figure the hardware must be "worn out" and buy an expensive new machine. Occasionally I can intercept the discard to use for another LTSP terminal.
On the other hand, we might simply be disagreeing over the definition of "music" (and other forms of art). If your definition encompasses any form of self-expression, then screaming certainly qualifies. And in fact, listening to such expressions of pain and anguish is something we do in our home in order to empathize with the sufferer. But not as "music". For instance, empathizing with pain and auguish does not help with studying. It's not that the screaming won't get into the home (in fact, we've been known to do it ourselves on occasion), it just won't get in as "music".
If you are truly saying that beauty is relative, then there is no true beauty in your world. Whatever you perceive as most beautiful can never be shared because others might perceive it differently. True beauty is there to be shared, independent of our human limitations. Though the perceiving of it becomes easier with practice.
She'll have plenty of time to tinker with the philosophy of life we are passing on when she leaves home.
People who don't pass on any moral structure to their kids are not doing them a favor. Stretched analogy time: It is like giving someone a computer with no software at all, and telling them to build their own boot code, operating system, and applications from the raw machine code on up.
A much better way is to give them a Linux distribution. Even if they hate it, they have something that works (hey, it works for me) that they can personalize to their hearts content before passing is on to their own kids.
Like Open Source, philosophy needs to be open. No hidden secrets available only to the elite (Gnosticism = Microsoft). When you understand the logical and historical basis of your philosophy, you have the tools to correct it without starting from scratch.
There is no such thing as a self made man. Or if there is, he won't get very far. We stand on the shoulders of giants.
I have a 14 year old daughter, and she likes modern music. I came up with an objective rule which she understands for what modern music is allowed in our home. Music where the singers scream or the instruments scream (distortion) is not allowed. Shoutin', on the other hand, is fine (Black Gospel).
This still has some false positives (e.g. Jesus Christ Superstar), but works fairly well.
This is a support nightmare if you are trying to do centralized support - but it is the heterogeneous environment the original poster was talking about. An exploit for Mozilla is not going to take out nearly the percentage of Linux users as an exploit for Outlook does for Windows.
While centralized support is a nightmare - it is unnecessary with open source. You can have many support companies specializing in various application and system choices. They can do this because the source code is not locked away by one corporation.
Borland Pascal. And its open source cousin, FreePascal. Borland Pascal is my low level language of Choice for the Windows platform (and for DOS before that). When statically linking a library, it pulls out only and precisely only the code and data needed for that application. It worked great for DOS TSR programs (which worked best with minimal memory use).
The catch is, that this approach only works for static linking. If you want to share the library between multiple different applications, the shared part has to be constant. When you get to GUI controls, e.g. with Delphi, there is a large DLL (.so for the linux version). You can still use the smart static linker, but your hello world GUI app is about 100K on Windows, and goes up from there as you use more interface elements.
I don't want to get into the tradeoffs of static vs dynamic linking, but when lots of apps use the same DLL/so, it saves memory and disk space, plus you can upgrade/break all the applications simply by upgrading/breaking the DLL/so. When all apps use a rather disjoint set of the DLL, static linking is better.
On a related note, a lot of the GUI code is loaded with the OS in Windows. Java Swing ignores this and recreates it all in Java. This makes Swing programs much bigger and slower. There are other (non Sun) GUI APIs for Java that use the Windows API directly and are much smaller and faster. C#, naturally, doesn't make this mistake. Java should have stuck with AWT, enhancing its capabilities instead of reinventing the Windowing OS. A mostly Java implementation of AWT would suffice for platforms without a suitable native Windowing system.
He presented the program and the password list to the school administrator. Instead of persecuting him, they contacted HP, who fixed the problem, and HP hired him with a college scholarship as soon as he graduated.
How times change.
Am I missing something?
SMTP was just not designed with spamming scum in mind.
Using the milter interface, messages can be rejected anytime. While rejecting extremely spammy mail is worthwhile, it is important to allow mail for which there is some doubt through in case of false positives. I suppose the milter could fully receive the mail before rejecting it, and *still* deliver it to the user for feedback.
to see if they're still together before pronouncing the marriage "successful"?
These ISPs are rightly listed by SPEWS - but have a monopoly. (No, satellite is not suitable.) To be able to send mail in these situations, we relay it through a machine on another ISP.
Why does a sinless hero lose interest for adults? I think the main reason is that having screwed up a lot, we need someone to model how to patch things up again. For instance, when Bob spends hours creating a trumpet from spare pipe, and Rolly flattens it, Bob does not get mad. He just says, "It's OK, you didn't see it, Rolly." Not me, I'd be livid. And Bob is no help showing me how to patch up the relationship afterward.
Disclaimer - I've never actually played with an AS/400 due to the price :-) I just read about it and like many of the design points.
Give the guy a break. What he means is, "you can't do embedded with Java II enterprise edition". He just doesn't consider those realtime and subset specifications to be real "java". Computer science types like a consistent vocabulary. The language and bytecode might be the same, but the libraries and environment are very different between the Micro and Enterprise editions. So how can they both be "java"?
You get all the advantages of Native compilation, and system specific JIT compilation combined - at the expense of more complexity and lots more disk space.
As others have mentioned, M$ is attempting to do a similar thing with .NET. Whether they can do as good a job without charging AS/400 prices remains to be seen.
As wonderful as all this is, I still like direct bytecode execution because it minimizes memory use and startup - which is more important than CPU for many applications (business logic and embedded). Native compiled code is quite a bit bigger than bytecode, uses more memory, and takes longer to load (though not as long as JITing the bytecode).
It may be inevitable that "none shall buy and sell except they have the mark of the beast", but that doesn't mean we should welcome it with open arms.
Fortunately, there are still merchants like Shoppers Food Warehouse and Trader Joe's in our area that don't require cards. In fact, SFW even advertises "No Card Required!".
I don't like NiCad because the Cadmium is so nasty. I will look into the NiMh and other rechargeables. Presumably they are no more toxic than ordinary alkaline.
So now I just buy regular alkaline and recharge them until they start leaking, then throw them out. Recharging before deep discharge gives the longest battery life for alkaline. I had a set in my RC car that I recharged after every play session, and they lasted for two years (with several sessions a month).
If batteries leak badly, I clean up the discharge with white vinegar and baking soda.
Python already has full garbage collection in both the C and Java implementations. The reference counting accelerates collection at the expense of runtime in C python. Since the reference counting overhead is fairly small compared to the overall interpreter overhead, it is a good tradeoff. Memory use is minimized, and circular references still get collected by the general collector. The Java implementation does not use reference counting, but relies solely on the Java runtime for GC.
Thanks to reference counting in the C implementation, you can write:
and the file is closed as soon as it is read into buf. However, this is not recommended practice for maximum portability, since the file will not get closed right away in the Java implementation and future C Python version might drop reference counting.While the python threading model does not easily take full advantage of multiple CPUs, it does make it easy to incorporate C code - even C code that is not written with threading in mind. Java supports full threading, but you have to be much more careful with JNI extensions. With python, all other threads are stopped when calling into C until you tell Python otherwise. So by default, non thread aware code runs just fine.
In addition to the summary score, the device should display the top 10 or so issues contributing to the summary score.
Some issue sources would be meta-issue sources combining multiple issues into a pre-weighted ethical ranking, e.g. 'Vatican' or 'Steinem'.
BTW, how does one go about testing the rightness of an ethical standard? Or is it an arbitrary individual choice? If the latter, is there any good reason why anyone should regard Saddam or Bush or Clinton as 'immoral'?
However, I have convinced several family and friends to switch to Linux on their desktops purchased for Windows. Part of the motivation was that Windows crashed constantly. Well, guess what? Linux crashes too. They say it crashes less often, but I suspect that is subjective. These machines have non-parity memory, hot CPUs, even hotter graphics cards, and no backup or disk mirroring. These crash-prone machines are generally 3 times the price of a Dell business server with backup (probably due to the high end 3D graphics and sound missing from a business server).
The sample size is small (3 desktop users of Windows and Linux), but they consistently rank OS reliability (on the cruddy hardware just described) like this:
- Windows ME - absolutely the worst. Don't plan to run for more than an hour.
- Windows 9x - better
- Windows XP - a big improvement
- Windows 98 running under Win4Lin - even better
- RedHat 7.3 - the best
As a result, they have all settled down to running mostly Linux and Win4Lin, and booting XP for the occasional Windows only multimedia app that won't run under Win4Lin.In Java and other dynamic languages like Smalltalk or Python, it is actually not very complex. It is standard procedure. All major APIs use a plugin architecture to allow changing implementation via configuration or on the fly. This makes systems must more robust and upgradeable. JDBC, Javamail, Java Cryptograpy Extensions, Java Message Service, you name it - it uses a plugin API. You can plug in GPL, LGPL, Proprietary, Sun Reference Implementation - all without affecting the license of your application, whether opensource or proprietary.