Yes Java is available for FreeBSD natively, however it seems like you have to compile it. I am not going to waste however long it takes for that beast to compile. I have to use the Linux binary, which works on FreeBSD fine it would seem, thank goodness. i know this is not FreeBSDs fault. Its Suns fault. Why dont they (Sun) allow a release of a binary copy? And all of this patching is pretty crazy as well. Why does Sun make people jump through hoops to use their language, and furthermore, why do people use thier language when they treat users in such a hostile manner? Why not use Perl, Python. TCL, or one of the other high level language that do not require so much nonsense.
I think it is crucially important, for Open Source developers to design software that can easily be ported to multiple OSs as much as possible without refactoring the code for each OS. Native Windows APIs and closed source proprietary-API Windows only libraries must be avoided as much as possible. If Open Source software an be made portable and run on multiple OSs, this reduces duplication of work and allows attention to be devoted to creating one high quality implementation for all OSs rather than having to reimplement software for each OS it is needed. One solution, and would with the least duplication of work to allow most Linux apps to run on Windows, is to use a Unix compatability layer on Windows that provides a Unix-like environment, including X. This is the lowest level software API interface, below graphics and utility toolkits like Gtk, so all of the libraries like Gtk, and countless other libraries, do not even need to be ported to Windows. Cygwin has made progress but it is still a rather incomplete imitation of Unix. Providing a compatability layer at such a low level means less higher level libraries have to be refactored for Windows APIs. The code that must interact with the Windows APIs is centralised at one place, at the bottom of the dependancy tree of Unix applications, so only the compatability layer needs to be concerned with Windows' quirks, rather than refactoring hundreds of higher level utility libraries.
Also, Portable Open Source software is actually helpful to getting people to switch to open source OSs. If people can start using open source applications on Windows and gradually switch over to using them exclusively, when they move to Linux they will be able to use all of the same applications when they make the move. Part of what keeps people on Windows is the fact that most software they use can only run on Windows. Getting software to be able to run on any OSs is important to OS choice.
I have used all resolutions up to 1080p, and I can definitely notice a very clear difference between 1080p and 720. The 1080 in my eyes looks much clearer and well defined over 720.
As far as analog, no thanks. I believe digital if I am not mistaken allows for eisier archival and copying via computer (which is your right to do to make backup copies for your private use), and I believe higher picture quality (via use of compression and such to allow more data to be compacted onto the disk). DVD isnt perfect, it does have some loss and some artifacts, but nothing like VHS and SVHS with repeated use. Properly cared for and manufactured, DVDs will last longer than VHS with repeated use, since there are no moving parts and tape winding. Sure you can ruin a DVD if you scratch it up and play frisbee with it, but properly handled it should last a long while, if it was properly manufactured as well. Disk rot occurs I believe because the coating on the disk was not properly applied allowing for air pockets under the surface and oxidation of the metal film the data is written upon.
Power line technology has various problems, such as the fact that power lines were not designed to carry high frequency transmissions and tend to turn into giant antenna systems when they are used in such a manner, disrupting accessibility to radio services such as shortwave radio and amatuer radio. It also turns out to be a rather expensive technology to implement as well as being problematic, since transformers tend to absorb and block RF signals on the power lines, requiring expensive solutions to bypass them. It is to the point where it will be so expensive to implement that it would be better to just implement a fiber optic network, which would provide better bandwidth anyway.
I think a much better and more effective, as well as higher quality solution for both bandwidth broadband avialability and the choice, and for maximum capacity, is to construct a shared fiber optic networks which could be used to carry telephone, cable tv, and internet. These systems should be owned and operated by local governments (who could contract out maintanence and construction to independant contractors if they wish) who would charge an access fee to fund the operatation the networks, and which would be open to all information service providers to provide their information services over them, such as multiple cable tv, phone and internent providers, giving people perhaps dozens more choices, assuring competition and choice for the consumer. This also would seperate the operation of the physical infrastructure from the information services, so one entity isnt controlling both the information services and infrastructure, which allows that entity to have a monopoly over the information services provided over the physical infrastructure. Instead access to the physical infrastructure would be avialable to all information services, like phone, internet and cable, and all of the information services and consumers would pool their resources to build one communications system which tends to be more efficient than every information service having to have its own information service, and it would make it eisier for smaller companies to enter the market and provide additional choices for the consumer since they do not have to fund the construction of another communications system for their exclusive use.
I meant to say this in the parent article (I should have used preview:-\ ):
A common mistake people who are attempting to make Linux programs eisier to use is forgetting that we can make Linux programs both expert friendly, where the expert can have absolute control over exactly how a system works and can configure everything, while at the same time make it easy to use for novice users, where nothing has to be configured.
I certianly agree that the directory structures are far better than Windows. Windows directory structures are a constant sources of problems. The default behaviour in many Windows programs is to store its data files in the same parent directory as its program executable files. Unix type systems instead store executables, libraries, and data files each in their own directory tree. When it comes time to backeup your data, you just have to back up/home, as a result, and/etc if you want to get your system configuration info.
I believe we can simultaneously make Linux programs easy to use for the novice user but also friendly the expert user where everything can be configured and tweaked. As far as useability of Linux programs go, I think the solution to that is not to make Linux applications so idiot proof that only an idiot can use them, but rather, allowing as much as practically possible in a program to be configured by the user, but place the advanced configuration settings in advanced configuration tabs or screens, so they are there is an advanced user wants to change them, but so they do not confuse average users.Only the more commonly used options would be placed on the main configuration screens, witha button or tab to access the advanced configuration. Another thing is to allow a program to be used with requiring the user to set as few configuration options as possible. The program should be able to default to reasonable deafults for values the user has not configured. All of the configuration settings should be avialable of course, but the user should only have to configure as many of them as possible if they want to.
I a common mistake people who are attempting to make Linux programs eisier to use is that we can make Linux programs both expert friendly, where the expert can have absolute control over exactly how a system works and can configure everything, while at the same time make it easy to use for novice users, where nothing has to be configured. this can be done as described as above. We can also allow expert users to manually configure everything from the command line and from configuration files, while providing the novice user with a GUI configuration program that can do everything automatically if they wish. The key to doing this is to design things in layers. GUI installer programs for instance could simply be front ends to configuration files and command line programs and run and configure them automatically. Furthermore, everything that can be done from a GUI should be able to be accomplished from the command line and configuration files, and vice versa. Command line programs and GUI programs can both share the same underlying infrastructure. For instance if you are writing an installer program, one could first implement the main functionaily in a library or command line program, and the GUI program could simply be a front end to that, so the same functionality does not have to be implemented in the CLI and GUI programs. This allows us to have the best of both worlds. Expert users can configure everything manually with command line programs and configuration files, while novice users can use a simple GUI program, because they both interface with and use the same core functionality. The user should be given the freedom to configure as little or as much as they want, and to use the software how they want rather than how the software designers think they should use it. Software should not restrict users and enforce rules and limitations on them if they do not want this. Too many Linux programs seem to be highly inflexible and provides few options which actually harms useability for more advanced users rather than realising as above we can make programs simultaneously both user and expert friendly, to allow everything to be configured or nothing to be configured based on what the user wants.
Another major useability issue with Linux is the issue of hardware support. I think it is great to have as many open source drivers as possible avialab
This is ridiculous. If we can obtian the features we need by adding an extension, why break protocol compatability and create a huge mess of non-interoperability by creating a non-compatable protocol?
People assume that just because something is "old" must mean it is bad. The logic behind this is flawed. Something being old does not make it bad. Only being poorly designed does. X11 is probably one of the best designed protocols, as far as functionality, extensibility, and workability goes. Whatever needs to be accomplished can be achieved with an extension, while maintaining forward and backwards compatability.
Also, regarding the handling of mouse, keyboard, and video device drivers, these are all in the realm of the device dependant areas of the system which should be considered seperate from the X protocol. The mouse, keyboard, and video support could be done in the kernel and then an X11 Server could simply interface to that support using APIs. The X11 Server would still handle windowing, management, etc, but it would interface with the kernel which could handle talking ot the video driver. This DOES NOT require a redesign of X11 Protocol, since the X11 protocol is seperate from the device dependant layer which interacts with hardware. This how we can have X servers which are also VNC server and display the desktop to VNC clients.
I believe most of the problem with having to spend large amounts of time to port an app to different OSs could be fixed by each OS supporting the same basic API calls and command line and compile tools, as defined in things such as POSIX and Single Unix Spec, in order to assure source compatability, the ability to recompile an app on any OS with no modifications. It seems one of the biggest offenders in making portability difficult is Windows, and a lot of time has to be wasted to port to that platform.
Many OSs, as well, support each others binary formats as well. FreeBSD can run compiled Linux applications, which is possible since Linuxs libraries, etc, are open source and can be used to support such binary compatability on other OSs. FreeBSDs Linuix support is pretty fast too, with little or no performance hit.
A large number of browser exploits seemed to be based on buffer overflow issues, which is a result of manual memory allocation in lower-high-level languages such as C/C++. Perhaps if a web browser would be written in a language with automatic memory allocation and management, like Java, Perl, Tcl, and the like, we would see fewer security problems. C/C++ is good for systems programming, like low level graphics and OS libraries, but I dont think it is the ideal choice in many cases for applications.
Few have ever used a truly relational database, since SQL does such a terrible job it cannot be called truly relational. The relational theory I believe when properly used leads to well designed, easy to maintain, and well performing databases. I dont really think OO databases or some other design will change things much, since the same issues remain as to actually how to physically store and retrieve data. The perfomance issues we see today I believe are the result of bad implementation, in the engine itself, in such a poor adaption of relational concepts as SQL, or in the design of the application, not the relational theory itself which is solid.
If someone wanted, as well, OO based layers can be done on top of relational databases, using the same engine. Then one can choose which perspective to utilise. There are good methods to store hierarchical data on a relational database if thats needed.
I think KDE should be a state of the art sort of thing. Keeping memory usage reasonable is important, however, it should not interfere with functionality and capability. There are many other environments that exist for those who are using 486s and who just need a simple text editor and a low memory usage window manager.
Also, I hope they make sure these desktop environments completely themeable, personally I have distaste of the aqua and brushed metal themes , and would like to be able to completely change the look and feel of the UI. Customisability is important but it doesnt have to get in the way of useability, the key is to put more advanced settings in a seperate dialog way rather than all in one dialog, and come with a reasonable set of default settings.
One thing I dont like about most desktop environments is they are hard to configure and work the way I want them to. For instance, I would rather have the panel in a regular, floating resizable window rather than locked to one side of the screen, but I cant seem to change its behaviour. For those people who just want the preconfigured environment and dont want to customise anything, they can have that, while those who want to customise can do so in "expert configuration" screens or whatever, where they can configure and control everything to their hearts content.
Perhaps a node/client setup could be provided, where clients could connect to a shared Freenet node, while keeping what the client has requested secret, by encrypting the connection from the client to the node. Of course, anyone who wants to would still be able to use their own Freenet node as well.
In my opnion, making Freenet into an invitation only network will raise the barrier to entry so much that only the most elite nerds will go through the trouble of using it. I sure cant find any number of people who use Freenet already. I also do not think that this will improve the reliability of the network much which at this time is a major problem.
I can understand the reasoning behind wanting to keep nodes anonymous by not giving out the fact that a computer is connected to Freenet, but making it an invitation only network I believe is the totally wrng way to do it, and there are much better ways to keep certian nodes anonymous, while providing a list of nodes to initially connect to for the end user, perhaps by simply allowing a node to opt out of being included on any node lists given to new users. This would seem to allow nodes who are concerned about being detected as much as possible, while still allowing users without invitations to connect to the network.
I don't think rewriting code already written in Java would be a productive and worthwhile use of time, however, I do think something needs to be done to address the need for an open source Java implementation that can be used by projects. Suns Java already is sources avialable for free, but still under a non-open source licence. Hopefully Sun will make the thing open source, that would be the best. Free software/open source, is pretty important to assure someone cant pull a "Bitkeeper" sort of thing on us and suddenly withdawing software from avialability, and to assure peoples freedoms to edit and modify the software and distribute the changes.
Computer games where you interact with the game via a screen and a joystick is not what I would consider VR. I would consider VR to be something that totally immerses all of the senses of touch, sight and sound into an artificial world, including some kind of feedback technology to produce touch sensations, so it seems you really are inside of that world, not just watching it on a screen.
There are cases where AOP can play a useful and beneficial role and is a good feature to have. It is really up to the programmer to decide whether or not a feature would be useful or not in a particular setting. Programming languages should give users the freedom to make such determinations, since each application is different, each will have different needs. A set of language features that works great for one app may be insufficient for another. Some people criticise certian features because for the things that they have done they might not be useful, but in that case they shouldnt be obliged to use them. But the feature might be essential in another program, and if the feature is not avialable because someone else decided it didnt fit their needs for what they were doing and thus made the assumption that therefore no one else should be allowed to use it, the programmers task as been made more complex and barriers placed in their way and their job more difficult. The program language should give the programmer powerful tools to use so that if the programmers needs those tools they are avialable.
I am surprised they are planning on switching around the dates instead of just getting rid of DST altogether. It is such a nuiscance having to switch the schedule around twice every year, For this most recent change many of us now have to go to work an hour earlier. Arizona and Indiana are forward thinking on this manner and have already ended DST. I wish it would happen in the rest of country.
I looked at the links listed in the article but it seems like all of the instances of downloadable texts therein come from other websites and have nothing to do with Google or its project. Many literary works already have been converted into electronic form, including 20,000 Leagues Under the Sea and others by Project Gutenberg etc. and have been avialable online for sometime.
There should be no problem with being able to change the ABI while also keeping backwards compatability as well, you simply include previous copies of glibc, which older binaries are linked to, and link all new binaries to the new version of glibc. The new version doesnt have to keep backwards compatability with the old versions, just keep the old copies of glibc on the system. This is probably a better idea than trying to include backwards compatability in every glibc library. This is what FreeBSD does. You can run FreeBSD binaries back to 3.0 at least, and as well you can run Linux binaries on FreeBSD.
I dont use Macs. I dislike the default look and feel of the mac UI intensely, although I do not know whether it could be completely customised and changed to my liking. I dont mind people having the Mac or Aqua look, but why not let people choose and create their own themes using feature rich GUI tools, so people can have what they want.
I was dissappointed by the movies. First because of the drastic changes in the plot of the movie. I know omissions are necessary or else the movie would never fit inside of 3 hours. But they made changes to the movie that were rather unnecessary for time purposes, one example was Frodo being carried to the ford in the first movie by the elf, who was also changed from a man to a woman in the movie. Such changes are inexcusable for time purposes.
The other thing was the sets, and landscapes. They seemed way off from the books. The dead marshes and mordor seemed way too bright, I always imagined them as being darker, nearly night, and the marshes has being foggy and hazy. But it seemed almost as bright as day in both. Places like Lorien, which I had imagined as being magical and having a brilliant luminance about them appeared to dark and ordinary in my opinion.Hobbiton didnt appear "grown in" enough, not enough trees, the place looked like it was set in the middle of a field, rather than admist woods and rich gardens and shrubbery as I imagined. There was far too little in diversity of landscapes, too little contrast between Mordor and Lorien. The magic of the books is lost as a result in the movies.
The acting in the movies was simply brilliant, if only the plot and the landscapes had been better.
Beauty is highly subjective and in the eye of the beholder. Every person has different tastes, your tastes and what you think might be pretty is probably quite different from what others might think is pretty. I like the look and feel of Gaim and dislike intensely the "shiny aqua" look of many user interfaces. Clearly, the user should be able to define the exact themes of their user interface, not the programmer. Does GTK a have a theme system with a GUI configuration tool that allows every aspect of the user interface to be configured by the user and allow the user to create new themes?
Being a heavy perl programmer and someone who has used perl extensively for a long time, I disagree completely that Perl code is more difficult to read than other languages on average, or that Perl encourages people to write bad code. I have found that writing good, easy to read code is as natural as any other language. If people are going to write bad code, I believe they will write bad code in any language they use. Perl's TIMTOWTDI is actually a key feature in keeping Perl readable and avoiding ugly hacks since the programmer is able to choose the feature that most clearly and best expresses what they are trying to do rather than try to devise ugly hacks to try to circumvent ridiculous limitations of a language. Of all of the languages for which I have read code, this includes Java, and Python, Perl code for me is the clearest and the most easily understood on average, because it does not place a lot of limitations on what can be done and make certian things far more difficult than they need to be through such limitations, and gives you the choice and freedom of many ways to do something.
Furthermore, I believe Perl should remain Perl and give programmers complete freedom to do what they need to do and allow them many ways to do it. If you want a language that places restrictions on you and decides what features and things you should be allowed to use, where the language designers impose their opinions and tastes on you through placing arbitrary restrictions on how the language can be used, please feel free to use the language of your choice that does that. There are plenty of them out there.
Microsoft maintains its monopoly through its proprietary Application ABI/APIs and the proprietary hardware driver ABI/APIs, which are inheritently more difficult to support by other OSs due to their closed source nature. Microsoft is one of the least interoperable parties, by refusing to release any source code which would help Wine better support Windows applications, and its lack of even a nearly standard system API and GUI APIs.
People should have the freedom to choose what OS to use. The goal of Linux should be to give people another choice, encourage OS freedom not by becoming the next MS but by encouraging source compatability standards so that code is compatible on all operating systems.
I am tired of hearing people say that the lack of a "consistant UI" is what is holding Unix back. I think this is nonsense and does wonders to distract attention from the real issues, the problems of driver and application API incompatabilities hindering software and driver portability. Every WIndows application I have used has a different look and feel. I dont think it really sets users back much. What really sets users back is when they cannot use their camera or their printer on Linux without having to troubleshoot it.
What people really want is to be able to use their hardware on whichever OS they choose and as well have a reasonably straighforward installation and configuration process. However, this also must be done without taking away freedom and a high degree of configurability in software use and choice. Their is a misconception that in order for software to be user friendly it has to have a barren sparse user interface and virtually no advanced configurability at all, basically an inflexible unconfigurable piece of monolithic software, that works one way, the way the desiners think you should be allowed to use it. But software can be simulataneously user friendly and expert friendly, simultaneously simple to configurable and configurable to the hilt. Simply put the most frequently and commonly used options most prominently placed and the more complex options in expert user screens for those that want them, and allow the software to use a set of reasonable defaults which can be changed but do not have to be set for the user to use the software. This goes for many things, including system configuration. Also, just because we need plug and play device detection and GUI tools, doesnt mean we have to make things unconfigurable for the expert. The GUI tools and plug and play can simply be front ends for underlying command line tools configuration files, so the expert user still has all of the customisability and fine level control they had before. Just make GUI tools a front end for, or work alongside as an optional component, a rich and complete command line environment, and we can have the best of both worlds.
The key is to make things user friendly and expert friendly simultaneously.
I think it also important for their to be a diversity of software, including allowing the user to choose their OS, their window system, word processor, etc, etc. Agian, I dont thing the "look and feel" thing is what is really holding back Unix, rather its he hardware compatability and software compatability issues with Windows, which needs to be addressed not by adopting Windows' ways in system design, but through compatability layers like Wine. Perhaps there should also be a hardware driver compatability layer for Windows drivers.
An important aspect to OS choice is allowing an application and hardware to be used on any OS, and the key to obtaining that is to encourage standard system level APIs, POSIX and single unix spec, and a standard low level windowing system, X11, in order to obtain source compatability, the abaility to recompile software on any compliant OS without modification. Compatability layers, such as Wine, are also important, in the case of Windows. Microsoft already has the benefit of having the allegiance of most hardware and software companies, so all they have to do is make
Slashdot Mirror
Unix Domain Sockets use shared memory to transfer data between applications. How does this compare to other shared memory methods in performance?
Yes Java is available for FreeBSD natively, however it seems like you have to compile it. I am not going to waste however long it takes for that beast to compile. I have to use the Linux binary, which works on FreeBSD fine it would seem, thank goodness. i know this is not FreeBSDs fault. Its Suns fault. Why dont they (Sun) allow a release of a binary copy? And all of this patching is pretty crazy as well. Why does Sun make people jump through hoops to use their language, and furthermore, why do people use thier language when they treat users in such a hostile manner? Why not use Perl, Python. TCL, or one of the other high level language that do not require so much nonsense.
I think it is crucially important, for Open Source developers to design software that can easily be ported to multiple OSs as much as possible without refactoring the code for each OS. Native Windows APIs and closed source proprietary-API Windows only libraries must be avoided as much as possible. If Open Source software an be made portable and run on multiple OSs, this reduces duplication of work and allows attention to be devoted to creating one high quality implementation for all OSs rather than having to reimplement software for each OS it is needed. One solution, and would with the least duplication of work to allow most Linux apps to run on Windows, is to use a Unix compatability layer on Windows that provides a Unix-like environment, including X. This is the lowest level software API interface, below graphics and utility toolkits like Gtk, so all of the libraries like Gtk, and countless other libraries, do not even need to be ported to Windows. Cygwin has made progress but it is still a rather incomplete imitation of Unix. Providing a compatability layer at such a low level means less higher level libraries have to be refactored for Windows APIs. The code that must interact with the Windows APIs is centralised at one place, at the bottom of the dependancy tree of Unix applications, so only the compatability layer needs to be concerned with Windows' quirks, rather than refactoring hundreds of higher level utility libraries.
Also, Portable Open Source software is actually helpful to getting people to switch to open source OSs. If people can start using open source applications on Windows and gradually switch over to using them exclusively, when they move to Linux they will be able to use all of the same applications when they make the move. Part of what keeps people on Windows is the fact that most software they use can only run on Windows. Getting software to be able to run on any OSs is important to OS choice.
I have used all resolutions up to 1080p, and I can definitely notice a very clear difference between 1080p and 720. The 1080 in my eyes looks much clearer and well defined over 720.
As far as analog, no thanks. I believe digital if I am not mistaken allows for eisier archival and copying via computer (which is your right to do to make backup copies for your private use), and I believe higher picture quality (via use of compression and such to allow more data to be compacted onto the disk). DVD isnt perfect, it does have some loss and some artifacts, but nothing like VHS and SVHS with repeated use. Properly cared for and manufactured, DVDs will last longer than VHS with repeated use, since there are no moving parts and tape winding. Sure you can ruin a DVD if you scratch it up and play frisbee with it, but properly handled it should last a long while, if it was properly manufactured as well. Disk rot occurs I believe because the coating on the disk was not properly applied allowing for air pockets under the surface and oxidation of the metal film the data is written upon.
Power line technology has various problems, such as the fact that power lines were not designed to carry high frequency transmissions and tend to turn into giant antenna systems when they are used in such a manner, disrupting accessibility to radio services such as shortwave radio and amatuer radio. It also turns out to be a rather expensive technology to implement as well as being problematic, since transformers tend to absorb and block RF signals on the power lines, requiring expensive solutions to bypass them. It is to the point where it will be so expensive to implement that it would be better to just implement a fiber optic network, which would provide better bandwidth anyway.
I think a much better and more effective, as well as higher quality solution for both bandwidth broadband avialability and the choice, and for maximum capacity, is to construct a shared fiber optic networks which could be used to carry telephone, cable tv, and internet. These systems should be owned and operated by local governments (who could contract out maintanence and construction to independant contractors if they wish) who would charge an access fee to fund the operatation the networks, and which would be open to all information service providers to provide their information services over them, such as multiple cable tv, phone and internent providers, giving people perhaps dozens more choices, assuring competition and choice for the consumer. This also would seperate the operation of the physical infrastructure from the information services, so one entity isnt controlling both the information services and infrastructure, which allows that entity to have a monopoly over the information services provided over the physical infrastructure. Instead access to the physical infrastructure would be avialable to all information services, like phone, internet and cable, and all of the information services and consumers would pool their resources to build one communications system which tends to be more efficient than every information service having to have its own information service, and it would make it eisier for smaller companies to enter the market and provide additional choices for the consumer since they do not have to fund the construction of another communications system for their exclusive use.
I meant to say this in the parent article (I should have used preview :-\ ):
A common mistake people who are attempting to make Linux programs eisier to use is forgetting that we can make Linux programs both expert friendly, where the expert can have absolute control over exactly how a system works and can configure everything, while at the same time make it easy to use for novice users, where nothing has to be configured.
I certianly agree that the directory structures are far better than Windows. Windows directory structures are a constant sources of problems. The default behaviour in many Windows programs is to store its data files in the same parent directory as its program executable files. Unix type systems instead store executables, libraries, and data files each in their own directory tree. When it comes time to backeup your data, you just have to back up /home, as a result, and /etc if you want to get your system configuration info.
I believe we can simultaneously make Linux programs easy to use for the novice user but also friendly the expert user where everything can be configured and tweaked.
As far as useability of Linux programs go, I think the solution to that is not to make Linux applications so idiot proof that only an idiot can use them, but rather, allowing as much as practically possible in a program to be configured by the user, but place the advanced configuration settings in advanced configuration tabs or screens, so they are there is an advanced user wants to change them, but so they do not confuse average users.Only the more commonly used options would be placed on the main configuration screens, witha button or tab to access the advanced configuration. Another thing is to allow a program to be used with requiring the user to set as few configuration options as possible. The program should be able to default to reasonable deafults for values the user has not configured. All of the configuration settings should be avialable of course, but the user should only have to configure as many of them as possible if they want to.
I a common mistake people who are attempting to make Linux programs eisier to use is that we can make Linux programs both expert friendly, where the expert can have absolute control over exactly how a system works and can configure everything, while at the same time make it easy to use for novice users, where nothing has to be configured. this can be done as described as above. We can also allow expert users to manually configure everything from the command line and from configuration files, while providing the novice user with a GUI configuration program that can do everything automatically if they wish. The key to doing this is to design things in layers. GUI installer programs for instance could simply be front ends to configuration files and command line programs and run and configure them automatically. Furthermore, everything that can be done from a GUI should be able to be accomplished from the command line and configuration files, and vice versa. Command line programs and GUI programs can both share the same underlying infrastructure. For instance if you are writing an installer program, one could first implement the main functionaily in a library or command line program, and the GUI program could simply be a front end to that, so the same functionality does not have to be implemented in the CLI and GUI programs. This allows us to have the best of both worlds. Expert users can configure everything manually with command line programs and configuration files, while novice users can use a simple GUI program, because they both interface with and use the same core functionality. The user should be given the freedom to configure as little or as much as they want, and to use the software how they want rather than how the software designers think they should use it. Software should not restrict users and enforce rules and limitations on them if they do not want this. Too many Linux programs seem to be highly inflexible and provides few options which actually harms useability for more advanced users rather than realising as above we can make programs simultaneously both user and expert friendly, to allow everything to be configured or nothing to be configured based on what the user wants.
Another major useability issue with Linux is the issue of hardware support. I think it is great to have as many open source drivers as possible avialab
This is ridiculous. If we can obtian the features we need by adding an extension, why break protocol compatability and create a huge mess of non-interoperability by creating a non-compatable protocol?
People assume that just because something is "old" must mean it is bad. The logic behind this is flawed. Something being old does not make it bad. Only being poorly designed does. X11 is probably one of the best designed protocols, as far as functionality, extensibility, and workability goes. Whatever needs to be accomplished can be achieved with an extension, while maintaining forward and backwards compatability.
Also, regarding the handling of mouse, keyboard, and video device drivers, these are all in the realm of the device dependant areas of the system which should be considered seperate from the X protocol. The mouse, keyboard, and video support could be done in the kernel and then an X11 Server could simply interface to that support using APIs. The X11 Server would still handle windowing, management, etc, but it would interface with the kernel which could handle talking ot the video driver. This DOES NOT require a redesign of X11 Protocol, since the X11 protocol is seperate from the device dependant layer which interacts with hardware. This how we can have X servers which are also VNC server and display the desktop to VNC clients.
I believe most of the problem with having to spend large amounts of time to port an app to different OSs could be fixed by each OS supporting the same basic API calls and command line and compile tools, as defined in things such as POSIX and Single Unix Spec, in order to assure source compatability, the ability to recompile an app on any OS with no modifications. It seems one of the biggest offenders in making portability difficult is Windows, and a lot of time has to be wasted to port to that platform.
Many OSs, as well, support each others binary formats as well. FreeBSD can run compiled Linux applications, which is possible since Linuxs libraries, etc, are open source and can be used to support such binary compatability on other OSs. FreeBSDs Linuix support is pretty fast too, with little or no performance hit.
A large number of browser exploits seemed to be based on buffer overflow issues, which is a result of manual memory allocation in lower-high-level languages such as C/C++. Perhaps if a web browser would be written in a language with automatic memory allocation and management, like Java, Perl, Tcl, and the like, we would see fewer security problems. C/C++ is good for systems programming, like low level graphics and OS libraries, but I dont think it is the ideal choice in many cases for applications.
Few have ever used a truly relational database, since SQL does such a terrible job it cannot be called truly relational. The relational theory I believe when properly used leads to well designed, easy to maintain, and well performing databases. I dont really think OO databases or some other design will change things much, since the same issues remain as to actually how to physically store and retrieve data. The perfomance issues we see today I believe are the result of bad implementation, in the engine itself, in such a poor adaption of relational concepts as SQL, or in the design of the application, not the relational theory itself which is solid.
If someone wanted, as well, OO based layers can be done on top of relational databases, using the same engine. Then one can choose which perspective to utilise. There are good methods to store hierarchical data on a relational database if thats needed.
I think KDE should be a state of the art sort of thing. Keeping memory usage reasonable is important, however, it should not interfere with functionality and capability. There are many other environments that exist for those who are using 486s and who just need a simple text editor and a low memory usage window manager.
Also, I hope they make sure these desktop environments completely themeable, personally I have distaste of the aqua and brushed metal themes , and would like to be able to completely change the look and feel of the UI. Customisability is important but it doesnt have to get in the way of useability, the key is to put more advanced settings in a seperate dialog way rather than all in one dialog, and come with a reasonable set of default settings.
One thing I dont like about most desktop environments is they are hard to configure and work the way I want them to. For instance, I would rather have the panel in a regular, floating resizable window rather than locked to one side of the screen, but I cant seem to change its behaviour. For those people who just want the preconfigured environment and dont want to customise anything, they can have that, while those who want to customise can do so in "expert configuration" screens or whatever, where they can configure and control everything to their hearts content.
Perhaps a node/client setup could be provided, where clients could connect to a shared Freenet node, while keeping what the client has requested secret, by encrypting the connection from the client to the node. Of course, anyone who wants to would still be able to use their own Freenet node as well.
In my opnion, making Freenet into an invitation only network will raise the barrier to entry so much that only the most elite nerds will go through the trouble of using it. I sure cant find any number of people who use Freenet already. I also do not think that this will improve the reliability of the network much which at this time is a major problem.
I can understand the reasoning behind wanting to keep nodes anonymous by not giving out the fact that a computer is connected to Freenet, but making it an invitation only network I believe is the totally wrng way to do it, and there are much better ways to keep certian nodes anonymous, while providing a list of nodes to initially connect to for the end user, perhaps by simply allowing a node to opt out of being included on any node lists given to new users. This would seem to allow nodes who are concerned about being detected as much as possible, while still allowing users without invitations to connect to the network.
I don't think rewriting code already written in Java would be a productive and worthwhile use of time, however, I do think something needs to be done to address the need for an open source Java implementation that can be used by projects. Suns Java already is sources avialable for free, but still under a non-open source licence. Hopefully Sun will make the thing open source, that would be the best. Free software/open source, is pretty important to assure someone cant pull a "Bitkeeper" sort of thing on us and suddenly withdawing software from avialability, and to assure peoples freedoms to edit and modify the software and distribute the changes.
Computer games where you interact with the game via a screen and a joystick is not what I would consider VR. I would consider VR to be something that totally immerses all of the senses of touch, sight and sound into an artificial world, including some kind of feedback technology to produce touch sensations, so it seems you really are inside of that world, not just watching it on a screen.
There are cases where AOP can play a useful and beneficial role and is a good feature to have. It is really up to the programmer to decide whether or not a feature would be useful or not in a particular setting. Programming languages should give users the freedom to make such determinations, since each application is different, each will have different needs. A set of language features that works great for one app may be insufficient for another. Some people criticise certian features because for the things that they have done they might not be useful, but in that case they shouldnt be obliged to use them. But the feature might be essential in another program, and if the feature is not avialable because someone else decided it didnt fit their needs for what they were doing and thus made the assumption that therefore no one else should be allowed to use it, the programmers task as been made more complex and barriers placed in their way and their job more difficult. The program language should give the programmer powerful tools to use so that if the programmers needs those tools they are avialable.
I am surprised they are planning on switching around the dates instead of just getting rid of DST altogether. It is such a nuiscance having to switch the schedule around twice every year, For this most recent change many of us now have to go to work an hour earlier. Arizona and Indiana are forward thinking on this manner and have already ended DST. I wish it would happen in the rest of country.
I looked at the links listed in the article but it seems like all of the instances of downloadable texts therein come from other websites and have nothing to do with Google or its project. Many literary works already have been converted into electronic form, including 20,000 Leagues Under the Sea and others by Project Gutenberg etc. and have been avialable online for sometime.
There should be no problem with being able to change the ABI while also keeping backwards compatability as well, you simply include previous copies of glibc, which older binaries are linked to, and link all new binaries to the new version of glibc. The new version doesnt have to keep backwards compatability with the old versions, just keep the old copies of glibc on the system. This is probably a better idea than trying to include backwards compatability in every glibc library. This is what FreeBSD does. You can run FreeBSD binaries back to 3.0 at least, and as well you can run Linux binaries on FreeBSD.
I dont use Macs. I dislike the default look and feel of the mac UI intensely, although I do not know whether it could be completely customised and changed to my liking. I dont mind people having the Mac or Aqua look, but why not let people choose and create their own themes using feature rich GUI tools, so people can have what they want.
I was dissappointed by the movies. First because of the drastic changes in the plot of the movie. I know omissions are necessary or else the movie would never fit inside of 3 hours. But they made changes to the movie that were rather unnecessary for time purposes, one example was Frodo being carried to the ford in the first movie by the elf, who was also changed from a man to a woman in the movie. Such changes are inexcusable for time purposes.
The other thing was the sets, and landscapes. They seemed way off from the books. The dead marshes and mordor seemed way too bright, I always imagined them as being darker, nearly night, and the marshes has being foggy and hazy. But it seemed almost as bright as day in both. Places like Lorien, which I had imagined as being magical and having a brilliant luminance about them appeared to dark and ordinary in my opinion.Hobbiton didnt appear "grown in" enough, not enough trees, the place looked like it was set in the middle of a field, rather than admist woods and rich gardens and shrubbery as I imagined. There was far too little in diversity of landscapes, too little contrast between Mordor and Lorien. The magic of the books is lost as a result in the movies.
The acting in the movies was simply brilliant, if only the plot and the landscapes had been better.
Beauty is highly subjective and in the eye of the beholder. Every person has different tastes, your tastes and what you think might be pretty is probably quite different from what others might think is pretty. I like the look and feel of Gaim and dislike intensely the "shiny aqua" look of many user interfaces. Clearly, the user should be able to define the exact themes of their user interface, not the programmer. Does GTK a have a theme system with a GUI configuration tool that allows every aspect of the user interface to be configured by the user and allow the user to create new themes?
Being a heavy perl programmer and someone who has used perl extensively for a long time, I disagree completely that Perl code is more difficult to read than other languages on average, or that Perl encourages people to write bad code. I have found that writing good, easy to read code is as natural as any other language. If people are going to write bad code, I believe they will write bad code in any language they use. Perl's TIMTOWTDI is actually a key feature in keeping Perl readable and avoiding ugly hacks since the programmer is able to choose the feature that most clearly and best expresses what they are trying to do rather than try to devise ugly hacks to try to circumvent ridiculous limitations of a language. Of all of the languages for which I have read code, this includes Java, and Python, Perl code for me is the clearest and the most easily understood on average, because it does not place a lot of limitations on what can be done and make certian things far more difficult than they need to be through such limitations, and gives you the choice and freedom of many ways to do something.
Furthermore, I believe Perl should remain Perl and give programmers complete freedom to do what they need to do and allow them many ways to do it. If you want a language that places restrictions on you and decides what features and things you should be allowed to use, where the language designers impose their opinions and tastes on you through placing arbitrary restrictions on how the language can be used, please feel free to use the language of your choice that does that. There are plenty of them out there.
Microsoft maintains its monopoly through its proprietary Application ABI/APIs and the proprietary hardware driver ABI/APIs, which are inheritently more difficult to support by other OSs due to their closed source nature. Microsoft is one of the least interoperable parties, by refusing to release any source code which would help Wine better support Windows applications, and its lack of even a nearly standard system API and GUI APIs.
People should have the freedom to choose what OS to use. The goal of Linux should be to give people another choice, encourage OS freedom not by becoming the next MS but by encouraging source compatability standards so that code is compatible on all operating systems.
I am tired of hearing people say that the lack of a "consistant UI" is what is holding Unix back. I think this is nonsense and does wonders to distract attention from the real issues, the problems of driver and application API incompatabilities hindering software and driver portability. Every WIndows application I have used has a different look and feel. I dont think it really sets users back much. What really sets users back is when they cannot use their camera or their printer on Linux without having to troubleshoot it.
What people really want is to be able to use their hardware on whichever OS they choose and as well have a reasonably straighforward installation and configuration process. However, this also must be done without taking away freedom and a high degree of configurability in software use and choice. Their is a misconception that in order for software to be user friendly it has to have a barren sparse user interface and virtually no advanced configurability at all, basically an inflexible unconfigurable piece of monolithic software, that works one way, the way the desiners think you should be allowed to use it. But software can be simulataneously user friendly and expert friendly, simultaneously simple to configurable and configurable to the hilt. Simply put the most frequently and commonly used options most prominently placed and the more complex options in expert user screens for those that want them, and allow the software to use a set of reasonable defaults which can be changed but do not have to be set for the user to use the software. This goes for many things, including system configuration. Also, just because we need plug and play device detection and GUI tools, doesnt mean we have to make things unconfigurable for the expert. The GUI tools and plug and play can simply be front ends for underlying command line tools configuration files, so the expert user still has all of the customisability and fine level control they had before. Just make GUI tools a front end for, or work alongside as an optional component, a rich and complete command line environment, and we can have the best of both worlds.
The key is to make things user friendly and expert friendly simultaneously.
I think it also important for their to be a diversity of software, including allowing the user to choose their OS, their window system, word processor, etc, etc. Agian, I dont thing the "look and feel" thing is what is really holding back Unix, rather its he hardware compatability and software compatability issues with Windows, which needs to be addressed not by adopting Windows' ways in system design, but through compatability layers like Wine. Perhaps there should also be a hardware driver compatability layer for Windows drivers.
An important aspect to OS choice is allowing an application and hardware to be used on any OS, and the key to obtaining that is to encourage standard system level APIs, POSIX and single unix spec, and a standard low level windowing system, X11, in order to obtain source compatability, the abaility to recompile software on any compliant OS without modification. Compatability layers, such as Wine, are also important, in the case of Windows. Microsoft already has the benefit of having the allegiance of most hardware and software companies, so all they have to do is make