If software patents are truly a problem for free software then they are a problem for all software since none actually protect you. They might as well give up and go back to using pen and paper if they are worried about software patents because using microsoft products or any other proprietary products will not protect them.
Any program of any size will infringe on dozens to many thousands of them. I don't think you can even write hello world in any language without infringing on at least a few patents. Stuff like java,python,perl,php, slashdot, mozilla, opera, kde, windows, probably violate thousands of patents and realistically you can be sued at any time for any of those since neither open or closed stuff gives you any kind of warranty from that kind of stuff and realistically they can't.
Software patents are a mine field. Anyone using//developing software is in a huge minefield and no matter where you step there is a landmine. The odds are when you put your foot down the mine will not explode but it is still a minefield. Once you are in the middle of the minefield it is too late to worry about violating software patents. Overall the only real option is just to ignore software patents and never look them up or learn much about them. At least then you can not be found to violate the patent on purpose which lessens the penalty and it makes it easier to disprove the patent.
Work needs to be done to throw out software patents but until then ignoring them seems to be best.
I disagree the problem is software patents. Patenting ideas is just too general purpose and the same idea tends to be come up with hundreds to hundreds of thousands of times all over the world without anyone knowing about the patent. Software patents just give a monopoly to the first person to get the idea patented no matter how many others come up with it.
Nobody comes up with ideas on their own since none of us is on an island completely seperated from the rest of humanity. You added your small piece to the idea of many others and that should not give you any right to control that.
Look at how many came up with stuff like read-copy-update stuff. Things are independently come up with so many times in so many places that software patents are just not a good idea. We all build our stuff on the backs of others and when we get to a certain level of knowledge as a whole the same idea will be come up with in many places.
Isn't some company going after some users of sql server over some patent violation? I am sure the exact name of the company will be covered somewhere in this discussion. However my point is that ALL software violates LOTS of patents. No matter what software you use someone can sue you at any time for patent violation no matter who you got the software from. Go read the license agreement for almost any commercial software and you will find out that you are left out to dry period.
With how loose software patents seem to be I expected nothing less then thousands in most places and in the usa I except it infringes on millions of patents. Heck I suspect hello world in any language infringes on at least a dozen patents. Stuff like python,perl,java,c# their runtimes etc probably infringes on tens to hundreds of thousands of patents.
So many stupid things have been patented that people seem to come up with hundreds to thousands of times in other places without ever hearing about the patent it just does not seem to matter very much. If you are writing softare you infringe on software patents and there is nothing you can do about it so why worry about it. Work needs to be done to change the patent system but worrying won't solve anything. I don't think it is possible to write software that does not infringe on various software patents and with that being that case it means that software patents are broken horribly. A patent it supposed to give someone a limited monopoly on an idea in exchange for sharing a unique idea with the rest of the human race. When people are coming up with them indepedently all over they are just not doing the job they should. Overall I don't even think patents should exist at all anymore. People do not come up with these ideas on their own, they get a lot of help from the society they are in and then they use that help and gouge the society with "their" invention of which likely less then 1% of the idea is actually theirs.
When running 1-2 apps at a time I find that windows 2000 and windows XP often are a little faster then kde 3.2 on debian sid. However under kde I rarely run less then 15 apps at a time and my normal usage is about 30 apps and zope(web app server) with about a 800M db loaded for devel and tests and I have found once I get more then 4-5 apps open at a time that my linux system performs far more smoothly then windows 2000 or xp does. Overall the more loaded the system the better linux does by comparison. The worse is when running db tests that keep the system pegged for hours but interactivity is still very good on the linux box. Overall my experience with windows 2000 and windows xp under extended high loads has been horrible. I find if something can keep disk io peaked with random access reads/writes, use all available ram and cpu and use just a little swap after a few hours any windows version will just lock up and die but I can leave my linux boxes doing that for months without any issues.
I suspect that even hello world infringes on several dozen patents. Something like python, perl, slashdot, gnome, kde etc are likely to violate thousands to tends of thousands. So much pointless stuff has software patents for it that there is no real point worrying. When you are in the middle of a minefield it is too late to worry about safety.
Overall I would just ignore all software patents. If you don't pay attention to any the odds are the penalties will be far far less. Also it makes it easier to invalidate a patent if you knew nothing about it when you infringed. Overall just try and write verty good software and get large businesses hooked on it. When it costs far less money to get the patent thrown out then it does to switch to some other system they will defend it.
Also remember that proprietary software offers no real advantages here. Any proprietary product you use could be nailed by this at any time also and it could put them out of business so it seems the risks are pretty close to me but free software is more likely to be defended by a larger number of people.
I suspect at some point free software is going to end up with some kind of get out of jail free type thing with resepect to patents like nasa has. Patents just won't apply to it since it hurts the society too much.
It would really help if people don't do this. No matter how often it is said to customers that people do this just based on the statistics we get told by customers to make the sites primarily IE compatible. If more browsers would report their true user agent it would make life a lot simpler.
It is pretty easy to make a site completely standards compliant and send it identical html. However to make it work well IE needs to get a different stylesheet from the standards compliant browsers which is basically everything else used. So if you have your browser lie as IE which it is not then the pages tend to render wrong since you where given CSS for a broken box model instead of a correct one.
Also if more browsers started showing higher percentages of usage it would be a lot easier to convince clients of the importance of paying to have more work done for other browsers.
One thing I would note is that when I look at statistics for usage I usually see around 90% for IE without cleaning however when I have checked carefully in the logs IE seems to have some strange flaw where it will sometimes request a page 3-5 times within a few ms I have even tested that locally and watched it with ethereal. I have no idea why and it is not from clicking the link more then once but it does screw the statistics up badly. When that is cleaned out I often see IE in the 60-80% range. Other browsers are being used out there by a large number of people but the log analyzers are broken pretty much and the other browsers have correct behavior.
The python standard library has tkinter. Currently it looks ugly but that is being worked on however it is easy to write for and it runs quickly and I have had almost no bugs working with it at various times.
Although based on looking at tkinter and some of the other toolkits it really doesn't look that bad it just does not look as good as the more modern things like qt and gtk both of which have good python bindings also.
An even better reason to use python is that you can target these various platforms. Currently you can just write python and it will work in cpython and jython. Fairly soon it will work on the CLR http://ironpython.com/ and at a pretty good speed also and it will likely get faster with time. From what I can see I can write my apps in python and will be able to deploy them on almost any kind of platform that is around in many ways.
Overall I see no reason to use java,c# etc but the java runtime and the.NET CLR could be useful and I don't have to change languages to use them or their libraries.
The problem with them being mostly the same but just a little different in a few places is overall I find that harder to work with. Especially when it changes with different patch versions.
I would rather deal with something that was broken in x way but it was consistently broken in that way then something that it sometimes broken one way and sometimes another. At least if it is consistent it can be worked around unlike where it is different in each one. However if you stay away from the quirks it works much more reliably in all of them.
At one point Mac IE was more standards compliant then Windows IE. Now they are both abysmal POS browsers and at least on the mac from what I have seen safari is going to wipe Mac IE out especially since Mac IE is dead as far as new versions, bug fixes etc go.
From what I see 95% of the people don't use IE. I tend to see closer to 85 - 80%. One of the things I have noticed is some sites are a self fulfilling prophecy they make it only work in their version of IE and so that is about the only kind of browser they see. The more browsers a site is compatible with the more I tend to see other browsers.
On the second point there is NO IE STANDARD. There NEVER was. Each version of IE has its own set of quirks and renders things differently. I have seen more then a few pages that where built for IE 6 on Windows XP that rendered differently on IE 6 on Windows 98 and IE 6 on Windows 2K. Overall I tend to get more consistent results of rendering in IE but using the subset of the W3C standards that IE actually supports. Anything else and you are using a quirk in the browser with only a few documented exceptions and using quirks just tends to cause more problems long term.
So to sum up for that there is the w3c standards and then there is writing for the IE that you have installed, there is no IE standard.
The current solution is done to work around bugs in current version of IE already. People figured out they could also run adds on the returned pages. Go look on microsoft.com and search on content-disposition. IE 5, 5.5 and 6.0 various patch versions have it broken in various ways and some versions actually work although many don't and they have broke it fairly often with patches. IE 6.0 with x patches works but with y ones doesn't for example.
Normally instead of that page you could return the data with content-disposition of attachment, provide a file name etc and other information. That works in EVERY other browser I have tested it with except IE which seems to be broken that way a lot.
So what it comes down to is that if you want it to work in all browsers you basically need to do it this way. If you don't care about IE compatibility you can do it the correct way.
Actually I use the rewrite module on all of my sites since I have apache running in front of zope in rewrite/proxy mode. It is pretty simple to do and the instructions are clear on how to do it.
What is easy/complex for different people is based on what they are used to. I can setup new sites, addition urls etc in about a minute or two which is how long it takes me to type the commands in apache.
Given enough time a C/C++ application will end up beating one in java/python/perl/c# etc but lets be honest in the real world there is not infinite time. It takes a LOT of work to make a fast C/C++ application that has had the memory leaks removed, bounds checked where they need to be etc. In the higher level languages you can implement a good enough solution that is fast enough quickly and it is simpler to debug and fix them.
I have seen a number of apps redone in java/python etc that have ended up a good deal faster then the c/c++ versions where. It also usually takes a month to do what it took 10-12 months to do in the c/c++ version.
For example in python you have dictionaries and because they are so simple to use people tend to use them as a base data structure all over the place and in python dictionaries are probably the most heavily optimized piece of code it has. So the most natural way to make a hashtable is just built in which tends to be a good deal harder in various lower level languages.
Also it is often easier to make a program written in a higher level language faster then it started out with basic profiling since the code is shorter and simpler to read. Overall from what I have seen programs in higher level langauges tend to be optimized more if it is important since it takes far less time to code them and the optimization is usually simpler to see and write.
The debian XFree 4.3 is also NOT a stock XFree 4.3. It includes many backported patches and patches that the XFree86 team would not accept for various reasons I never got. Overall XFree86 4.3 for debian is a good release and not anywhere as close to out of data as some would believe.
One of the things that I understand is that ATI tried to get some stuff fixed for radeon users for over a year in XFree86 and kept submitting patches that where never applied. The debian version has those patches applied and overall if you have a radeon card then the debian xfree release is probably the best release to use currently.
Actually you can make a pdf version of any document with kde. You can just print to a pdf file. Also if you have kde installed you can pretty much do it with any app on the system just have them use kprinter to print with.
I have used this feature from many times with koffice, konqueror, kmail etc since it gives your a format that pretty much everything on earth can read exactly as it was intended which docs fail to do far too often. I do like that openoffice can do this also but from what I understand KDE did have that first in the unix world as far as any app being able to use it.
Fontconfig not working for you in KDE sure is strange I have been using it for over a year now in debian sid.
In my/etc/X11/XF86Config-4 file there are NO fontpath entries and there have not been for over a year. Since its introduction and kde supporting it fontconfig has worked pretty much flawlessly for me. There where a few buggy packages in debian sid where the font cache did not get updated correctly and you had to run fc-cache but it has been months since that has been a problem.
I compile almost nothing on my systems except for the code that I am working on. I use a debian system because it has just worked for me however I have also used knoppix and recent versions of Mandrake and I have NEVER seen the issues you are talking about. Actually I have not seen library issues of any kind on linux sine redhat 5.2 or so many years ago and I have used many versions.
I have not seen libc issues in a long time on redhat, mandrake, suse, debian, knoppix, etc.
Overall it seems to be cool to knock linux and while it does have issues it would be better if you stuck to real issues that existed. If you are compiling all this software yourself then you should probably stop until you learn how to compile software correctly.
I have used Mac OS/X, Windows XP, OS/2, etc and I find KDE to be far more usable then any of them and I think it is more usable then they ever can be for some very simple reasons.
In KDE I configure spellcheck ONCE. The settings are system wide, the dictionary is shared and it works the same in EVERY KDE app. I don't have to deal with different applications having different ideas and having to set that up for each one.
In KDE I configure the proxy settings for my network ONCE. Every KDE app uses those settings from viewing webpages to using webdav to grab data remotely. I just set it once and don't worry about it again.
In KDE I configure my editor ONCE. I set what I want my default editor to be customize it to how I want to view text, work with it etc. I currently use the kate embedded component as my default editor and it works the same way showing embedded data in konqueror, in the kate application, in kwrite, in kdevelop 3 etc. I could change it to kvim in one place if I wanted and use it everywhere. I can set my editing component to ANY registered text editing component. I have seen NO equivalent functionality in any other environment.
KDE has io slaves. EVERY file save/load, the url bar in konqueror etc are url transparent. I just work with files and don't have to worry about what app I am using. I can just open a file in my editor, word processor, spreadsheet etc and not need a special sftp, ftp, webdav,imap etc client to grab the data. I just worry about what data I need to work with and where I need to save it to and not about what app I am using.
KDE saves me a large ammount of time getting work done and while some areas are still buggy and need work it is far more usable now then I see OS/X, XP, etc as EVER being. I don't think proprietary software can agree to the point that I can choose what my system wide editor is, what my spellcheck system is etc.
Mac OS/X looks prettier but I also find that for what I do it makes the job far slower and harder then it needs to be. It has lots of pretty graphics but things like io slaves just save too much time. I am sure it is great how easy spellcheck is to setup in various apps but I setup these things ONCE for the system and now I just get about doing my work. I have 3 monitors setup and I have 6 virtual desktops and on startup I tend to have about 50 gui apps running along with zope that I use to do my work.
You can have mac osx and all of these other environments. KDE is becoming more usable all the time and it is becoming more integrated and easier for more settings to be system wide.
Note: I know that GNOME has some of these same features and will likely have all of them in the future and that GNOME and KDE are working together on freedesktop standards etc. I just commented on the KDE side because that is what I use the most. I want both to continue to exist and grow stronger through competition and cooperation.
In the end I don't think multiple proprietary vendors can ever really work together and so I don't see it as possible to have an environment as integrated as KDE is already and will become more so in the future.
Python does compile to bytecode. It just does not require a separate step to compile to bytecode like java does. If the bytecode is out of date it will be recompiled and used automatically.
Konqueror supports proxy config like that (1 for all proxy types) and it also prompts for password on any connection type that returns some kind of denied thing I have tested it with ftp, sftp, webdav etc.
Konqueror in kde 3.2.1 is also a nice improvement it renders faster and more correctly then it did before. Overall in my experience it is a tossup as to which browser konqueror/mozilla is more standards compliant some stuff one does better and some stuff the other does better.
KDE has transparent webdav to every kde application through the io slaves. You could use konqueror like filemanager and do webdav://server or webdavs://server since it also has ssl webdav support but more useful is that you can use those urls from any kde application so in your word processor, code editor, sound recorder etc you can save to webdav just like you would your home dir. To make it easier to get back to it you can bookmark it.
I normally don't use the webdav functionality however stuff like sftp works the same way.
Konqueror does all of that because it is not a file manager, web browser etc. It is a kpart embedder. It embeds khtml to do html rendering and other stuff for other things. I can embed kate, kword, kpdf etc for other things. It can speak all of those protocols because every kde app can. You can use sftp, smb, imap, pop3 etc from ANY kde application. Although why you would want to open a pop3 connection from kword I am not sure but it does work. Any file dialog box in kde can use any of the io slaves.
That also means that you can paste any url that kio understand into a file upload box in a webpage and it will deal with it and have the webpage the content of that url. That saves such a large ammount of time doing web apps and lots of other things. Give an sftp url to some data and just have it load it up.
I have tried it at lots of dpi settings and every program in it scales correctly since there is a layout manager.
Shortcuts by default are the same in all kde applications for the same command. You can also change that system wide. If you change cut to be F12 it is F12 in every kde application.
However KDE also shows a large degree of caring with the kio-slaves. Can you tell me that windows or macosx have EVERY app url transparent to probably 50 or more different protocols. Do you have any idea how much time can be saved in a web app when the client can work with any of the urls transparently? Overall from what I have seen of users using kde they have found it far easier to use then windows and macosx and more productive. Also once you show them how the io slaves work they don't want to go back to anything else because it is so easy.
As a small example of what has saved customers and other developers a fair bit of time is that kio-slaves work in an html file form control. I have often seen it where you wanted a file on some remote webpage uploaded to a different one. All you do is take the url for the one you want and paste it into the file upload box and when you send the form kde will grab the remote file and send it as part of the form submit. No messing with downloading files locally and then uploading them. Everything just works transparently for you.
That kind of thinking is for the end user and it helps a lot and gnome is the only system other then kde that has these kinds of features although from what I understand it does not work as well yet but it is progressing rapidly and likely the kde and gnome ones will becoming unified at some point.
Slashdot Mirror
If software patents are truly a problem for free software then they are a problem for all software since none actually protect you. They might as well give up and go back to using pen and paper if they are worried about software patents because using microsoft products or any other proprietary products will not protect them.
Any program of any size will infringe on dozens to many thousands of them. I don't think you can even write hello world in any language without infringing on at least a few patents. Stuff like java,python,perl,php, slashdot, mozilla, opera, kde, windows, probably violate thousands of patents and realistically you can be sued at any time for any of those since neither open or closed stuff gives you any kind of warranty from that kind of stuff and realistically they can't.
Software patents are a mine field. Anyone using//developing software is in a huge minefield and no matter where you step there is a landmine. The odds are when you put your foot down the mine will not explode but it is still a minefield. Once you are in the middle of the minefield it is too late to worry about violating software patents. Overall the only real option is just to ignore software patents and never look them up or learn much about them. At least then you can not be found to violate the patent on purpose which lessens the penalty and it makes it easier to disprove the patent.
Work needs to be done to throw out software patents but until then ignoring them seems to be best.
I disagree the problem is software patents. Patenting ideas is just too general purpose and the same idea tends to be come up with hundreds to hundreds of thousands of times all over the world without anyone knowing about the patent. Software patents just give a monopoly to the first person to get the idea patented no matter how many others come up with it.
Nobody comes up with ideas on their own since none of us is on an island completely seperated from the rest of humanity. You added your small piece to the idea of many others and that should not give you any right to control that.
Look at how many came up with stuff like read-copy-update stuff. Things are independently come up with so many times in so many places that software patents are just not a good idea. We all build our stuff on the backs of others and when we get to a certain level of knowledge as a whole the same idea will be come up with in many places.
Isn't some company going after some users of sql server over some patent violation? I am sure the exact name of the company will be covered somewhere in this discussion. However my point is that ALL software violates LOTS of patents. No matter what software you use someone can sue you at any time for patent violation no matter who you got the software from. Go read the license agreement for almost any commercial software and you will find out that you are left out to dry period.
With how loose software patents seem to be I expected nothing less then thousands in most places and in the usa I except it infringes on millions of patents. Heck I suspect hello world in any language infringes on at least a dozen patents. Stuff like python,perl,java,c# their runtimes etc probably infringes on tens to hundreds of thousands of patents.
So many stupid things have been patented that people seem to come up with hundreds to thousands of times in other places without ever hearing about the patent it just does not seem to matter very much. If you are writing softare you infringe on software patents and there is nothing you can do about it so why worry about it. Work needs to be done to change the patent system but worrying won't solve anything. I don't think it is possible to write software that does not infringe on various software patents and with that being that case it means that software patents are broken horribly. A patent it supposed to give someone a limited monopoly on an idea in exchange for sharing a unique idea with the rest of the human race. When people are coming up with them indepedently all over they are just not doing the job they should. Overall I don't even think patents should exist at all anymore. People do not come up with these ideas on their own, they get a lot of help from the society they are in and then they use that help and gouge the society with "their" invention of which likely less then 1% of the idea is actually theirs.
When running 1-2 apps at a time I find that windows 2000 and windows XP often are a little faster then kde 3.2 on debian sid. However under kde I rarely run less then 15 apps at a time and my normal usage is about 30 apps and zope(web app server) with about a 800M db loaded for devel and tests and I have found once I get more then 4-5 apps open at a time that my linux system performs far more smoothly then windows 2000 or xp does. Overall the more loaded the system the better linux does by comparison. The worse is when running db tests that keep the system pegged for hours but interactivity is still very good on the linux box. Overall my experience with windows 2000 and windows xp under extended high loads has been horrible. I find if something can keep disk io peaked with random access reads/writes, use all available ram and cpu and use just a little swap after a few hours any windows version will just lock up and die but I can leave my linux boxes doing that for months without any issues.
I suspect that even hello world infringes on several dozen patents. Something like python, perl, slashdot, gnome, kde etc are likely to violate thousands to tends of thousands. So much pointless stuff has software patents for it that there is no real point worrying. When you are in the middle of a minefield it is too late to worry about safety.
Overall I would just ignore all software patents. If you don't pay attention to any the odds are the penalties will be far far less. Also it makes it easier to invalidate a patent if you knew nothing about it when you infringed. Overall just try and write verty good software and get large businesses hooked on it. When it costs far less money to get the patent thrown out then it does to switch to some other system they will defend it.
Also remember that proprietary software offers no real advantages here. Any proprietary product you use could be nailed by this at any time also and it could put them out of business so it seems the risks are pretty close to me but free software is more likely to be defended by a larger number of people.
I suspect at some point free software is going to end up with some kind of get out of jail free type thing with resepect to patents like nasa has. Patents just won't apply to it since it hurts the society too much.
It would really help if people don't do this. No matter how often it is said to customers that people do this just based on the statistics we get told by customers to make the sites primarily IE compatible. If more browsers would report their true user agent it would make life a lot simpler.
It is pretty easy to make a site completely standards compliant and send it identical html. However to make it work well IE needs to get a different stylesheet from the standards compliant browsers which is basically everything else used. So if you have your browser lie as IE which it is not then the pages tend to render wrong since you where given CSS for a broken box model instead of a correct one.
Also if more browsers started showing higher percentages of usage it would be a lot easier to convince clients of the importance of paying to have more work done for other browsers.
One thing I would note is that when I look at statistics for usage I usually see around 90% for IE without cleaning however when I have checked carefully in the logs IE seems to have some strange flaw where it will sometimes request a page 3-5 times within a few ms I have even tested that locally and watched it with ethereal. I have no idea why and it is not from clicking the link more then once but it does screw the statistics up badly. When that is cleaned out I often see IE in the 60-80% range. Other browsers are being used out there by a large number of people but the log analyzers are broken pretty much and the other browsers have correct behavior.
The python standard library has tkinter. Currently it looks ugly but that is being worked on however it is easy to write for and it runs quickly and I have had almost no bugs working with it at various times.
Although based on looking at tkinter and some of the other toolkits it really doesn't look that bad it just does not look as good as the more modern things like qt and gtk both of which have good python bindings also.
An even better reason to use python is that you can target these various platforms. Currently you can just write python and it will work in cpython and jython. Fairly soon it will work on the CLR http://ironpython.com/ and at a pretty good speed also and it will likely get faster with time. From what I can see I can write my apps in python and will be able to deploy them on almost any kind of platform that is around in many ways.
.NET CLR could be useful and I don't have to change languages to use them or their libraries.
Overall I see no reason to use java,c# etc but the java runtime and the
The problem with them being mostly the same but just a little different in a few places is overall I find that harder to work with. Especially when it changes with different patch versions.
I would rather deal with something that was broken in x way but it was consistently broken in that way then something that it sometimes broken one way and sometimes another. At least if it is consistent it can be worked around unlike where it is different in each one. However if you stay away from the quirks it works much more reliably in all of them.
At one point Mac IE was more standards compliant then Windows IE. Now they are both abysmal POS browsers and at least on the mac from what I have seen safari is going to wipe Mac IE out especially since Mac IE is dead as far as new versions, bug fixes etc go.
From what I see 95% of the people don't use IE. I tend to see closer to 85 - 80%. One of the things I have noticed is some sites are a self fulfilling prophecy they make it only work in their version of IE and so that is about the only kind of browser they see. The more browsers a site is compatible with the more I tend to see other browsers.
On the second point there is NO IE STANDARD. There NEVER was. Each version of IE has its own set of quirks and renders things differently. I have seen more then a few pages that where built for IE 6 on Windows XP that rendered differently on IE 6 on Windows 98 and IE 6 on Windows 2K. Overall I tend to get more consistent results of rendering in IE but using the subset of the W3C standards that IE actually supports. Anything else and you are using a quirk in the browser with only a few documented exceptions and using quirks just tends to cause more problems long term.
So to sum up for that there is the w3c standards and then there is writing for the IE that you have installed, there is no IE standard.
The current solution is done to work around bugs in current version of IE already. People figured out they could also run adds on the returned pages. Go look on microsoft.com and search on content-disposition. IE 5, 5.5 and 6.0 various patch versions have it broken in various ways and some versions actually work although many don't and they have broke it fairly often with patches. IE 6.0 with x patches works but with y ones doesn't for example.
Normally instead of that page you could return the data with content-disposition of attachment, provide a file name etc and other information. That works in EVERY other browser I have tested it with except IE which seems to be broken that way a lot.
So what it comes down to is that if you want it to work in all browsers you basically need to do it this way. If you don't care about IE compatibility you can do it the correct way.
Actually I use the rewrite module on all of my sites since I have apache running in front of zope in rewrite/proxy mode. It is pretty simple to do and the instructions are clear on how to do it.
What is easy/complex for different people is based on what they are used to. I can setup new sites, addition urls etc in about a minute or two which is how long it takes me to type the commands in apache.
Given enough time a C/C++ application will end up beating one in java/python/perl/c# etc but lets be honest in the real world there is not infinite time. It takes a LOT of work to make a fast C/C++ application that has had the memory leaks removed, bounds checked where they need to be etc. In the higher level languages you can implement a good enough solution that is fast enough quickly and it is simpler to debug and fix them.
I have seen a number of apps redone in java/python etc that have ended up a good deal faster then the c/c++ versions where. It also usually takes a month to do what it took 10-12 months to do in the c/c++ version.
For example in python you have dictionaries and because they are so simple to use people tend to use them as a base data structure all over the place and in python dictionaries are probably the most heavily optimized piece of code it has. So the most natural way to make a hashtable is just built in which tends to be a good deal harder in various lower level languages.
Also it is often easier to make a program written in a higher level language faster then it started out with basic profiling since the code is shorter and simpler to read. Overall from what I have seen programs in higher level langauges tend to be optimized more if it is important since it takes far less time to code them and the optimization is usually simpler to see and write.
The debian XFree 4.3 is also NOT a stock XFree 4.3. It includes many backported patches and patches that the XFree86 team would not accept for various reasons I never got. Overall XFree86 4.3 for debian is a good release and not anywhere as close to out of data as some would believe.
One of the things that I understand is that ATI tried to get some stuff fixed for radeon users for over a year in XFree86 and kept submitting patches that where never applied. The debian version has those patches applied and overall if you have a radeon card then the debian xfree release is probably the best release to use currently.
Actually you can make a pdf version of any document with kde. You can just print to a pdf file. Also if you have kde installed you can pretty much do it with any app on the system just have them use kprinter to print with.
I have used this feature from many times with koffice, konqueror, kmail etc since it gives your a format that pretty much everything on earth can read exactly as it was intended which docs fail to do far too often. I do like that openoffice can do this also but from what I understand KDE did have that first in the unix world as far as any app being able to use it.
Fontconfig not working for you in KDE sure is strange I have been using it for over a year now in debian sid.
/etc/X11/XF86Config-4 file there are NO fontpath entries and there have not been for over a year. Since its introduction and kde supporting it fontconfig has worked pretty much flawlessly for me. There where a few buggy packages in debian sid where the font cache did not get updated correctly and you had to run fc-cache but it has been months since that has been a problem.
In my
I compile almost nothing on my systems except for the code that I am working on. I use a debian system because it has just worked for me however I have also used knoppix and recent versions of Mandrake and I have NEVER seen the issues you are talking about. Actually I have not seen library issues of any kind on linux sine redhat 5.2 or so many years ago and I have used many versions.
I have not seen libc issues in a long time on redhat, mandrake, suse, debian, knoppix, etc.
Overall it seems to be cool to knock linux and while it does have issues it would be better if you stuck to real issues that existed. If you are compiling all this software yourself then you should probably stop until you learn how to compile software correctly.
I have used Mac OS/X, Windows XP, OS/2, etc and I find KDE to be far more usable then any of them and I think it is more usable then they ever can be for some very simple reasons.
In KDE I configure spellcheck ONCE. The settings are system wide, the dictionary is shared and it works the same in EVERY KDE app. I don't have to deal with different applications having different ideas and having to set that up for each one.
In KDE I configure the proxy settings for my network ONCE. Every KDE app uses those settings from viewing webpages to using webdav to grab data remotely. I just set it once and don't worry about it again.
In KDE I configure my editor ONCE. I set what I want my default editor to be customize it to how I want to view text, work with it etc. I currently use the kate embedded component as my default editor and it works the same way showing embedded data in konqueror, in the kate application, in kwrite, in kdevelop 3 etc. I could change it to kvim in one place if I wanted and use it everywhere. I can set my editing component to ANY registered text editing component. I have seen NO equivalent functionality in any other environment.
KDE has io slaves. EVERY file save/load, the url bar in konqueror etc are url transparent. I just work with files and don't have to worry about what app I am using. I can just open a file in my editor, word processor, spreadsheet etc and not need a special sftp, ftp, webdav,imap etc client to grab the data. I just worry about what data I need to work with and where I need to save it to and not about what app I am using.
KDE saves me a large ammount of time getting work done and while some areas are still buggy and need work it is far more usable now then I see OS/X, XP, etc as EVER being. I don't think proprietary software can agree to the point that I can choose what my system wide editor is, what my spellcheck system is etc.
Mac OS/X looks prettier but I also find that for what I do it makes the job far slower and harder then it needs to be. It has lots of pretty graphics but things like io slaves just save too much time. I am sure it is great how easy spellcheck is to setup in various apps but I setup these things ONCE for the system and now I just get about doing my work. I have 3 monitors setup and I have 6 virtual desktops and on startup I tend to have about 50 gui apps running along with zope that I use to do my work.
You can have mac osx and all of these other environments. KDE is becoming more usable all the time and it is becoming more integrated and easier for more settings to be system wide.
Note: I know that GNOME has some of these same features and will likely have all of them in the future and that GNOME and KDE are working together on freedesktop standards etc. I just commented on the KDE side because that is what I use the most. I want both to continue to exist and grow stronger through competition and cooperation.
In the end I don't think multiple proprietary vendors can ever really work together and so I don't see it as possible to have an environment as integrated as KDE is already and will become more so in the future.
Python does compile to bytecode. It just does not require a separate step to compile to bytecode like java does. If the bytecode is out of date it will be recompiled and used automatically.
Konqueror supports proxy config like that (1 for all proxy types) and it also prompts for password on any connection type that returns some kind of denied thing I have tested it with ftp, sftp, webdav etc.
Konqueror in kde 3.2.1 is also a nice improvement it renders faster and more correctly then it did before. Overall in my experience it is a tossup as to which browser konqueror/mozilla is more standards compliant some stuff one does better and some stuff the other does better.
KDE 3.2 supports SVG trasnparently so konqueror will render svg just fine.
KDE has transparent webdav to every kde application through the io slaves. You could use konqueror like filemanager and do webdav://server or webdavs://server since it also has ssl webdav support but more useful is that you can use those urls from any kde application so in your word processor, code editor, sound recorder etc you can save to webdav just like you would your home dir. To make it easier to get back to it you can bookmark it.
I normally don't use the webdav functionality however stuff like sftp works the same way.
Konqueror does all of that because it is not a file manager, web browser etc. It is a kpart embedder. It embeds khtml to do html rendering and other stuff for other things. I can embed kate, kword, kpdf etc for other things. It can speak all of those protocols because every kde app can. You can use sftp, smb, imap, pop3 etc from ANY kde application. Although why you would want to open a pop3 connection from kword I am not sure but it does work. Any file dialog box in kde can use any of the io slaves.
That also means that you can paste any url that kio understand into a file upload box in a webpage and it will deal with it and have the webpage the content of that url. That saves such a large ammount of time doing web apps and lots of other things. Give an sftp url to some data and just have it load it up.
Use KDE then.
I have tried it at lots of dpi settings and every program in it scales correctly since there is a layout manager.
Shortcuts by default are the same in all kde applications for the same command. You can also change that system wide. If you change cut to be F12 it is F12 in every kde application.
However KDE also shows a large degree of caring with the kio-slaves. Can you tell me that windows or macosx have EVERY app url transparent to probably 50 or more different protocols. Do you have any idea how much time can be saved in a web app when the client can work with any of the urls transparently? Overall from what I have seen of users using kde they have found it far easier to use then windows and macosx and more productive. Also once you show them how the io slaves work they don't want to go back to anything else because it is so easy.
As a small example of what has saved customers and other developers a fair bit of time is that kio-slaves work in an html file form control. I have often seen it where you wanted a file on some remote webpage uploaded to a different one. All you do is take the url for the one you want and paste it into the file upload box and when you send the form kde will grab the remote file and send it as part of the form submit. No messing with downloading files locally and then uploading them. Everything just works transparently for you.
That kind of thinking is for the end user and it helps a lot and gnome is the only system other then kde that has these kinds of features although from what I understand it does not work as well yet but it is progressing rapidly and likely the kde and gnome ones will becoming unified at some point.