You might want to try out the Unix (Linux) command-line. Piping together small independent tool seems to be a concept you already discovered yourself and the Unix commandline is the best place to use this concept.
In my experience 99 out of 100 apps are run with $ java -jar appname.jar from the command line, or else they come with an installer that sets up a launch script for you.
The last one that tried the script-setup-thing was a major pain. It used InstallNoWhere (InstallAnyWhere it is called by its makers) which needed Java 1.3 to run and brought its own JRE 1.3 with it. At first it did not run at all because the script that called it (the Java-Part of the Installer) had a workaround for some Linux-Glibc-Bug some major versions before mine and was incompatible with the fix. After googling a bit and removing the lines in the script with the workaround I tried to run it again. Now it did not run because of another issue. The JRE it insisted an using which was packaged with the Installer had not been compiled statically but dynamically with glibc (the glibc from the release of the installer) and did not run with the newer version.
So I had to unpack the Installer-File by Hand and search for the Class to call to start the Installer. When I found that one the Installation ran almost fine except for one little problem. It insisted on using an 1.3 JRE for the Installed Software so I had to find an JRE 1.3 somewhere. After installing that I could then change the Start-Script for the installed Java-App to use my Systems Standard-JRE with which it run not perfect but crashed only approximately once per hour.
Now tell me how someone with normal (no) experience with such issues should install such a Java-App
And that would harm you how?
Most Programs install Links on the Desktop AND in the Start Menu so it is just a matter of copying a few links from the Start Menu to the Desktop
Good Text Adventures have at least some of the qualities of a good book although most Text Adventures are more like Short Stories or very thin books in length.
Wait till you have seen all of Gameplay a hundred times, only then will you be able to really appreciate a good storyline because story is different in any game, gameplay is more or less the same in any game of the same genre, there will be a point when you grow tired of gameplay but you will never grow tired of good stories.
The Problem simply is that for every real choice you double the number of possible story lines so if you get a choice betwenn two of these every 5 minutes you have 4096 distinct storylines to build after an hour of gameplay which is quite impossible at least for a commercial game with limited ressources.
If that is the way you say why did some Java-program (It was Borland Together I think) I had to use a few months ago for my University insist on using a JRE 1.3 instead of using my 1.4 JRE and crashed constantly when I changed the start-script to use 1.4?
If you want to write code that doesn't do the job, doesn't have usability, doesn't have compatibility, doesn't integrate, and doesn't consider business use then I don't know go make the next pong or something.
Can you think of anything major that he has promised to do?
I don't know about the US but in most so-called Democracies the Politicians do everything but the things they promised, after all its hard to find good promises for the next election, so why not keep the old problems unfixed to have one less thing for the party to worry about.
The sad thing about this is the use of the term P2P for all this. If they had used filesharing or some term like that there might still be hope for Bittorrent-like P2P replacing HTTP-Downloads for Big files everywhere and Browsers having internal Bittorrent Clients like they have download managers now but this whole story is killing these legal uses as well.
But with speech recognition there are things better technology can not overcome, speed and parallel use of the voice for other things like telephone or RL-talking being the of the biggest concerns that are simply not solvable by better speech recognition technology
You can use the exact same argument the other way round. Speech can not be done while doing something else. I can not use the telephone or speak to someone in RL while at the same time using my voice to command my computer while keyboard and mouse work just fine.
I imagine telephone support to be an even greater mess than it is now when speech recognition is used by the client (the person in need for help).
The point is that you can not use Java Library from any other Programming Language than Java so to use this Library you have to write your App in Java as well where it suffers from this problems
Nice idea. Perhaps this speech recognition is good for something useful. Neither dictating text nor commanding my PC via voice seem to be very useful to me (except for people missing the arms/hands to do it the traditional way). Both are slower than typing/mouse and more errorprone. Even if they could develop a 110% accurate Speech Recognition I wouldn't use it to replace typing because of the slowness of speech.
Java might be as fast as C in Code Execution but if you want to build a library that Open Source Applications outside the Java-Developer-niche use you have to write it in C. C is still THE No. 1 language for libraries for use in programs written in lots of different programming languages.
The only thing that really sucks with KMail is that you have to install half of KDE to use it. I am sure I am not the only one who does not use KDE as Desktop (I use XFCE4) and wants to use this. So please KDE-Developers get a clue and use a new packaging scheme with one package per KDE-Application.
Slashdot Mirror
You might want to try out the Unix (Linux) command-line. Piping together small independent tool seems to be a concept you already discovered yourself and the Unix commandline is the best place to use this concept.
You could hack the Installer of every Game you use to not show you the EULA in case you want to make mods or Server-Emus later.
Why should Gentoo aim for the "easy-install"-Users when - as you say yourself - there are already countless other distributions to target them?
Now tell me how someone with normal (no) experience with such issues should install such a Java-App
And that would harm you how?
Most Programs install Links on the Desktop AND in the Start Menu so it is just a matter of copying a few links from the Start Menu to the Desktop
Sadly, most Movie Producers don't realize or ignore this and go on ruining good books with bad movies that claim to be the same as the book.
Good Text Adventures have at least some of the qualities of a good book although most Text Adventures are more like Short Stories or very thin books in length.
Wait till you have seen all of Gameplay a hundred times, only then will you be able to really appreciate a good storyline because story is different in any game, gameplay is more or less the same in any game of the same genre, there will be a point when you grow tired of gameplay but you will never grow tired of good stories.
The Problem simply is that for every real choice you double the number of possible story lines so if you get a choice betwenn two of these every 5 minutes you have 4096 distinct storylines to build after an hour of gameplay which is quite impossible at least for a commercial game with limited ressources.
If that is the way you say why did some Java-program (It was Borland Together I think) I had to use a few months ago for my University insist on using a JRE 1.3 instead of using my 1.4 JRE and crashed constantly when I changed the start-script to use 1.4?
Ever heard about this thing called "morale". Some people think happy employees are more productive (I've heard).
The sad thing about this is the use of the term P2P for all this. If they had used filesharing or some term like that there might still be hope for Bittorrent-like P2P replacing HTTP-Downloads for Big files everywhere and Browsers having internal Bittorrent Clients like they have download managers now but this whole story is killing these legal uses as well.
But with speech recognition there are things better technology can not overcome, speed and parallel use of the voice for other things like telephone or RL-talking being the of the biggest concerns that are simply not solvable by better speech recognition technology
Funny you mention that.
You can use the exact same argument the other way round. Speech can not be done while doing something else. I can not use the telephone or speak to someone in RL while at the same time using my voice to command my computer while keyboard and mouse work just fine.
I imagine telephone support to be an even greater mess than it is now when speech recognition is used by the client (the person in need for help).
And how do you guarantee there are no buffer overflows in the massive amounts of C Code in you JVM and Libraries?
In related news: There are Programming Languages without the Overhead of the JVM and without Pointers.
The point is that you can not use Java Library from any other Programming Language than Java so to use this Library you have to write your App in Java as well where it suffers from this problems
Not to mention that it is worthless to have exactly one responsive Java App which is itself a Java Development Tool.
*searches for the +1 Scary Mod-Option (again)*
Nice idea. Perhaps this speech recognition is good for something useful. Neither dictating text nor commanding my PC via voice seem to be very useful to me (except for people missing the arms/hands to do it the traditional way). Both are slower than typing/mouse and more errorprone. Even if they could develop a 110% accurate Speech Recognition I wouldn't use it to replace typing because of the slowness of speech.
Java might be as fast as C in Code Execution but if you want to build a library that Open Source Applications outside the Java-Developer-niche use you have to write it in C. C is still THE No. 1 language for libraries for use in programs written in lots of different programming languages.
The only thing that really sucks with KMail is that you have to install half of KDE to use it. I am sure I am not the only one who does not use KDE as Desktop (I use XFCE4) and wants to use this. So please KDE-Developers get a clue and use a new packaging scheme with one package per KDE-Application.