I use PFE, the Programmer's File Editor. It doesn't even have syntax highlighting, but it's light, fast, very configurable and the macro function rocks (Shift + F7, Do some stuff with Ctrl+Cursor or whatever you want, Ctrl + F7, repeat with F7).
Regrettably it's out of development, though there still is a bug that sometimes occures: when editing, lines disappear and you should not save the file if that happens. But it does not happen very often.
The Mode feature is not very intuitive, but once you figured out how it works, it allows you to switch things like line-indenting, wrapping, etc. based on the file type you are editing.
http://www.lancs.ac.uk/people/cpaap/pfe/
I don't use other editors because of the time to get used to them... and because of basic stuff like Ctrl+Cursor, where PFE stopps at far more characters than whitespace only. I need this a lot and most other editors don't do this.
Which is filled by varios.anything files and directories. This is as annoying as Windows Apps creating their own files and folders in "My Documents" - the way they do it is just not how I like it (and differs from app to app). That's why I store "My Documents" somewhere else which is easier under Windows than it is under Linux because of the drive letters. No, that's not a good invention either. Man, that's weird...
Hiding away the chaos by front ends just does not feel right IMHO. Sometimes you have to make a cut und rebuild everything based on the experience you have. That's evolution. Linux is a dinosaur;) He can try as hard as he wants, but cleaning the house won't be possible for him. Sorry. He's not designed for it. But he has other valuable qualities.
The problem with hiding the chaos is that you have to hide anything then. I don't think that's a fun job, and in a recent/. article about OS developers we saw that this is a motivation killer. The result is holes. Sometimes there is no other way to do stuff, important stuff, than to dive into the chaos and edit a strange configuration file or something. At this point you need to tell the user the truth. "Yes. Linux is a mess. Yes, you have to use the command line to do that, I'm sorry. Use the pipe sign. Yes, use Alt Gr + the key with the arrow brackets on them. No, not that one. Use chmod 755. Yes, C, H, M, O, D..." Not to *bash* the command line. Haha;)
Hiding the chaos is easier than starting from scratch, this is right. The typical Linux user is fine with that chaos because he got used to it, so it does not stand in his way. But for newbies the learning curve is too big and I don't think that KDE can cover all holes. So chances look bad for Linux on the Mainstream Desktop, I think.
Yet I think there must be a way to build an operating system in a way that both allows complete control for the gurus and easy and consistent configurability and usability for the average Joe. But, again, if this was Linux, it would be Linux the kernel plus a bunch of stuff developed after one standard (or what is the correct expression? Develop according to a standard? I'm not sure). Not Linux as we know it today.
To me Linux feels untidy right out of the box. There is a system after which the things are done, but the system itself is... not what I would do. It starts with the directory structure. It is not intuitive where you will find a certain binary -/bin?/usr/bin?/usr/local/bin?/sbin?/usr/sbin?/usr/local/sbin? And so on. After a while you begin to understand that sbin is for the super user. But why do I find the user data in/home and not in/usr? It seems weird right from the start. But changing only this aspect will cause a really big amount of problems.
It already sucks that you have to set a shebang line with an absolute path in most cases. If that script is used on your PC (maybe it runs under Windows, so you might have to use #!c:/Progra...) and on several websites on several servers, you have to edit this line for several servers. At least if they don't use/usr/bin/perl.
In this regard, even the windows file extension approach is good (it knows that.pl has to be opened with perl.exe in dir X).
Nevertheless I understand that the shebang line has other advantages.
But Linux is a bit to much chaos for me. This is why every Linux guru really impresses me. I'm not willing to explore this mess to finally find the advantages over Windows (and Linux surely has them) though I consider myself a geek. But probably this is why I read/. though all my friends lough at me;)
And it's not KDE 3. It's like wearing second hand, it just does not feel as good. And Windows users will turn away from KDE if you show them a KDE without all the eye candy and speed that KDE 3 seems to have (have not tried it - yet)
Your complaint seem to be more that there are multiple possibilities or too much variants of everything to have standarized products. The point is valid! But that does not mean Linux is crap.
I never said so. Linux stands for lots of cool techniques, choices over choices and so on. It's the coolest playground for programmers on earth. With playground being a universe that allows the programmer to be god because he could change really anything.
Unfortunately this also means that Linux is chaos. No sandbox looks tidy, right? Linux is a huge sandbox where you can play as much as you want.
Linux rocks for developers, but end users will always feel a bit helpless, if they want a clean system (which I guess they want). It's like a child's room of the size of 1000 m and you are the one who has to tidy it. You wouldn't even want to make a list of the stuff in there unless you go into the room because of the stuff in there. The typical Windows user that wants to get work done in the first place does not want to explore the complete room to know how he can do his stuff the best way. He's fine with dictated solutions, as long as they more or less work for him.
Making an Open Source Operating System for the masses could use Linux (the kernel) as the bases and probably some KDE stuff, too (like KParts), but the applications of it have to be developed after new common standards so that the programs integrate much better. You can reuse the core of many existing programs, I think, but all the front end stuff and the interfaces (at API level) will have to be changed to make this possible. This is a lot of work. And it's not what's going on in Linux today, because that's not what Linux is about. That's okay, but then don't expect Linux to become mainstream ever.
Linux is Linux and it should stay that way (because this is freedom). An OS-OS should be a seperate project IMHO (because there you want control, which is managed freedom if it's done right).
The standards KDE apps follow? The standards Gnome apps follow? The common standards that should make them use the same clipboard? The common standards that should allow to define a set of icons you want to have on your desktop and when you switch from KDE to Gnome, they are still there? The common standards that make them use the same start menu? The common standards that make it so simple to write a configuration program that can be used with every distro - after all they are all called Linux, so what? The common standards that allow having to configure one aspect of the system at a single place instead of two or three (what was it, fonts?). The common standards that allow setting a "default browser" in one single central place? Did I miss anything?
Linux distributions are no operating systems like Windows is. Linux is the kernel and a bunch of applications that different people developed for different reasons in different ways with different aims. Windows is explorer, registry, dial up network, system control panel, etc. plus apps that use them somehow. Common standards!!!
I don't want to say that Microsoft did those things the best way - far from it. But Linux does not even have something like this. There is no central unit. KDE brings a bit of this feeling back, but there are too many important Linux apps that were not developed for KDE. Now read the paragraph again with s/KDE/Gnome/g;
Complete decentralization is cool for P2P but it's the worst thing you can do with operating systems IMHO.
I work for a web company and besides having to program in PHP I also have to design here and there. To test these things IE is a must have, because most people will use IE to view the site.
In most other cases the reason for not "being able" to switch are tools I got used to, like Windows Commander, Programmer's File Editor, Corel Photo Paint and some other things.
I installed Linux several times (several versions and distributions of course) just to see how far it has come. Once I tried to get the extra buttons on my Logitech keyboard to work. No chance. There seemed to be some hacks out there according to Google, but nothing simple. And that's a big problem, because those keys really increase my productivity. I miss them every time when I use different computer.
And you find that in other areas of Linux as well. Of course there are editors, file managers, browsers and so on. But most of them (at least the easy-to-use-ones) are not available for Windows, so if I try to switch, I have to learn ALL of them from scratch because my favourite Windoze apps are not available for Linux. Last time I checked there was a Windows Commander clone. But like many things in Linux it seemed to be a project that somebody started 3 weeks ago - not a full blown product with all the little details that make it rock.
I hate the feeling of not knowing how to do the simplest things - like setting an environment variable so that it remains set when I reboot - though almost nobody I know comes close to my knowledge of and experience with computers - I'm talking about people in RL, of course.
So when I use Windows, I am the master, I control the system. When I use Linux, I feel like Alice in Wonderland. Lots of cool things here and there but I don't relly know how to use them the best way.
Currently I lack the time of getting used to Linux to use it efficiently.
But I think porting KDE to Windows would help a bit. Then I can get the Linux desktop experience on Windows. When I get to a point were I need to do something NOW and not after I read a really long HOWTO or something, I can easily use the techniques I know under Windows and learn the Linux way later. Or not. Who cares;)
Then please explain why (before Slashdot was commercialized) that the _public weblogs_ said that roughly 90% of the users was viewing slashdot through WIn(X ver) on IE ?
I am one of those (IE 6 on XP), and though I don't use Linux, I enjoy reading about improvements of KDE (including Alpha Releases).
I'd like to use Open Source everywhere, yet there are still too many things that I can use under Windows only. Plus, I don't think Linux is ready for the desktop. It will never be. Otherwise it wouldn't be Linux anymore. To make a good Open Source operating system for the masses, everything you will include into this operating system (distribution) would have to developed after common standards - it has to look and feel like ONE product. And that's just impossible with Linux as we know it today. But I don't think any Linux zealot will admit that. Right?
Yet, after all, I think Linux (the kernel) and KDE are both technically great products. And I love to see what people can create if they only want to.
Is it just me? I thought Perl 6 will "simply" translate from Perl to Parrot Assembly language. A usable Parrot has nothing to do with a usable Perl 6. You can use Perl 6 when both Parrot AND Perl 6 itself are done. Right?
Not if these functions can be used standalone. Then you can simply #include them in another project instead of #including all the other (probably useless) functions in the file as well and instead of having to copy it. If you copy it, you might want to synch changes afterwards which would be painful if you had to edit the file instead of just overwriting it.
If the functions belong to a library, you can still have a library file that includes all functions, so that the projects that use this library only have to include one single file.
Each function in an own file could also ease version management if you are not the only one working on the project. At least if you do not use cvs for whatever reason.
Does it still sound ridiculous? Is there any real argument AGAINST using separate files for each function, class, etc.? Then please tell.
For instance, at what point do you split that massive source file into multiple files?
Right from the start, I'd say. Each function/class/whatever should have their own file.
At what point do two functions approaching similar functionality need to be merged, despite the cost of digging through the source and making changes to call the new function?
As early as possible, when as few as possible other functions or parts in the code already use it.
I'd rather see a list of/. readers, the city they live in and a link to mail them, maybe over a site-internal system to avoid spamming.
Meetings could be organized privately. I'd like to know whether there are any Slashdotters here in this area... that's the only "problem" I have, not a meeting place.
Slashdot Mirror
I use PFE, the Programmer's File Editor. It doesn't even have syntax highlighting, but it's light, fast, very configurable and the macro function rocks (Shift + F7, Do some stuff with Ctrl+Cursor or whatever you want, Ctrl + F7, repeat with F7).
Regrettably it's out of development, though there still is a bug that sometimes occures: when editing, lines disappear and you should not save the file if that happens. But it does not happen very often.
The Mode feature is not very intuitive, but once you figured out how it works, it allows you to switch things like line-indenting, wrapping, etc. based on the file type you are editing.
http://www.lancs.ac.uk/people/cpaap/pfe/
I don't use other editors because of the time to get used to them... and because of basic stuff like Ctrl+Cursor, where PFE stopps at far more characters than whitespace only. I need this a lot and most other editors don't do this.
Which is filled by varios .anything files and directories. This is as annoying as Windows Apps creating their own files and folders in "My Documents" - the way they do it is just not how I like it (and differs from app to app). That's why I store "My Documents" somewhere else which is easier under Windows than it is under Linux because of the drive letters. No, that's not a good invention either. Man, that's weird...
;) He can try as hard as he wants, but cleaning the house won't be possible for him. Sorry. He's not designed for it. But he has other valuable qualities.
/. article about OS developers we saw that this is a motivation killer. The result is holes. Sometimes there is no other way to do stuff, important stuff, than to dive into the chaos and edit a strange configuration file or something. At this point you need to tell the user the truth. "Yes. Linux is a mess. Yes, you have to use the command line to do that, I'm sorry. Use the pipe sign. Yes, use Alt Gr + the key with the arrow brackets on them. No, not that one. Use chmod 755. Yes, C, H, M, O, D..." ;)
Hiding away the chaos by front ends just does not feel right IMHO. Sometimes you have to make a cut und rebuild everything based on the experience you have. That's evolution. Linux is a dinosaur
The problem with hiding the chaos is that you have to hide anything then. I don't think that's a fun job, and in a recent
Not to *bash* the command line. Haha
Hiding the chaos is easier than starting from scratch, this is right. The typical Linux user is fine with that chaos because he got used to it, so it does not stand in his way. But for newbies the learning curve is too big and I don't think that KDE can cover all holes. So chances look bad for Linux on the Mainstream Desktop, I think.
Yet I think there must be a way to build an operating system in a way that both allows complete control for the gurus and easy and consistent configurability and usability for the average Joe. But, again, if this was Linux, it would be Linux the kernel plus a bunch of stuff developed after one standard (or what is the correct expression? Develop according to a standard? I'm not sure). Not Linux as we know it today.
To me Linux feels untidy right out of the box. There is a system after which the things are done, but the system itself is... not what I would do. It starts with the directory structure. It is not intuitive where you will find a certain binary - /bin? /usr/bin? /usr/local/bin? /sbin? /usr/sbin? /usr/local/sbin? And so on. After a while you begin to understand that sbin is for the super user. But why do I find the user data in /home and not in /usr? It seems weird right from the start. But changing only this aspect will cause a really big amount of problems.
/usr/bin/perl.
.pl has to be opened with perl.exe in dir X).
/. though all my friends lough at me ;)
It already sucks that you have to set a shebang line with an absolute path in most cases. If that script is used on your PC (maybe it runs under Windows, so you might have to use #!c:/Progra...) and on several websites on several servers, you have to edit this line for several servers. At least if they don't use
In this regard, even the windows file extension approach is good (it knows that
Nevertheless I understand that the shebang line has other advantages.
But Linux is a bit to much chaos for me. This is why every Linux guru really impresses me. I'm not willing to explore this mess to finally find the advantages over Windows (and Linux surely has them) though I consider myself a geek. But probably this is why I read
And it's not KDE 3. It's like wearing second hand, it just does not feel as good. And Windows users will turn away from KDE if you show them a KDE without all the eye candy and speed that KDE 3 seems to have (have not tried it - yet)
Unfortunately this also means that Linux is chaos. No sandbox looks tidy, right? Linux is a huge sandbox where you can play as much as you want.
Linux rocks for developers, but end users will always feel a bit helpless, if they want a clean system (which I guess they want). It's like a child's room of the size of 1000 m and you are the one who has to tidy it. You wouldn't even want to make a list of the stuff in there unless you go into the room because of the stuff in there. The typical Windows user that wants to get work done in the first place does not want to explore the complete room to know how he can do his stuff the best way. He's fine with dictated solutions, as long as they more or less work for him.
Making an Open Source Operating System for the masses could use Linux (the kernel) as the bases and probably some KDE stuff, too (like KParts), but the applications of it have to be developed after new common standards so that the programs integrate much better. You can reuse the core of many existing programs, I think, but all the front end stuff and the interfaces (at API level) will have to be changed to make this possible. This is a lot of work. And it's not what's going on in Linux today, because that's not what Linux is about. That's okay, but then don't expect Linux to become mainstream ever.
Linux is Linux and it should stay that way (because this is freedom). An OS-OS should be a seperate project IMHO (because there you want control, which is managed freedom if it's done right).
Which common standards are you talking about?
The standards KDE apps follow?
The standards Gnome apps follow?
The common standards that should make them use the same clipboard?
The common standards that should allow to define a set of icons you want to have on your desktop and when you switch from KDE to Gnome, they are still there?
The common standards that make them use the same start menu?
The common standards that make it so simple to write a configuration program that can be used with every distro - after all they are all called Linux, so what?
The common standards that allow having to configure one aspect of the system at a single place instead of two or three (what was it, fonts?).
The common standards that allow setting a "default browser" in one single central place? Did I miss anything?
Linux distributions are no operating systems like Windows is. Linux is the kernel and a bunch of applications that different people developed for different reasons in different ways with different aims. Windows is explorer, registry, dial up network, system control panel, etc. plus apps that use them somehow. Common standards!!!
I don't want to say that Microsoft did those things the best way - far from it. But Linux does not even have something like this. There is no central unit. KDE brings a bit of this feeling back, but there are too many important Linux apps that were not developed for KDE. Now read the paragraph again with s/KDE/Gnome/g;
Complete decentralization is cool for P2P but it's the worst thing you can do with operating systems IMHO.
I work for a web company and besides having to program in PHP I also have to design here and there. To test these things IE is a must have, because most people will use IE to view the site.
;)
In most other cases the reason for not "being able" to switch are tools I got used to, like Windows Commander, Programmer's File Editor, Corel Photo Paint and some other things.
I installed Linux several times (several versions and distributions of course) just to see how far it has come. Once I tried to get the extra buttons on my Logitech keyboard to work. No chance. There seemed to be some hacks out there according to Google, but nothing simple. And that's a big problem, because those keys really increase my productivity. I miss them every time when I use different computer.
And you find that in other areas of Linux as well. Of course there are editors, file managers, browsers and so on. But most of them (at least the easy-to-use-ones) are not available for Windows, so if I try to switch, I have to learn ALL of them from scratch because my favourite Windoze apps are not available for Linux. Last time I checked there was a Windows Commander clone. But like many things in Linux it seemed to be a project that somebody started 3 weeks ago - not a full blown product with all the little details that make it rock.
I hate the feeling of not knowing how to do the simplest things - like setting an environment variable so that it remains set when I reboot - though almost nobody I know comes close to my knowledge of and experience with computers - I'm talking about people in RL, of course.
So when I use Windows, I am the master, I control the system. When I use Linux, I feel like Alice in Wonderland. Lots of cool things here and there but I don't relly know how to use them the best way.
Currently I lack the time of getting used to Linux to use it efficiently.
But I think porting KDE to Windows would help a bit. Then I can get the Linux desktop experience on Windows. When I get to a point were I need to do something NOW and not after I read a really long HOWTO or something, I can easily use the techniques I know under Windows and learn the Linux way later. Or not. Who cares
I'd like to use Open Source everywhere, yet there are still too many things that I can use under Windows only. Plus, I don't think Linux is ready for the desktop. It will never be. Otherwise it wouldn't be Linux anymore. To make a good Open Source operating system for the masses, everything you will include into this operating system (distribution) would have to developed after common standards - it has to look and feel like ONE product. And that's just impossible with Linux as we know it today. But I don't think any Linux zealot will admit that. Right?
Yet, after all, I think Linux (the kernel) and KDE are both technically great products. And I love to see what people can create if they only want to.
Is it just me? I thought Perl 6 will "simply" translate from Perl to Parrot Assembly language. A usable Parrot has nothing to do with a usable Perl 6. You can use Perl 6 when both Parrot AND Perl 6 itself are done. Right?
Not if these functions can be used standalone. Then you can simply #include them in another project instead of #including all the other (probably useless) functions in the file as well and instead of having to copy it. If you copy it, you might want to synch changes afterwards which would be painful if you had to edit the file instead of just overwriting it.
If the functions belong to a library, you can still have a library file that includes all functions, so that the projects that use this library only have to include one single file.
Each function in an own file could also ease version management if you are not the only one working on the project. At least if you do not use cvs for whatever reason.
Does it still sound ridiculous? Is there any real argument AGAINST using separate files for each function, class, etc.? Then please tell.
I'd rather see a list of /. readers, the city they live in and a link to mail them, maybe over a site-internal system to avoid spamming.
Meetings could be organized privately. I'd like to know whether there are any Slashdotters here in this area... that's the only "problem" I have, not a meeting place.