Ah yes, the never ending quest for new and better ways to bog down the Linux desktop... this time with a new load of resident interpreters. But I guess this shouldn't be a surprise... bloat is well on it's way to becoming a Linux standard.
New lanaguages and concepts are cool and help Linux to grow. New users do too. A lot of people try Linux on older hardware where they've been told it's supposed to work the best. Then it's "welcome to the modern distro" where new installs are frequently in swap before the default Gnome or KDE desktop finishes booting. Watch enthusiasm turn to grief as the new user clicks on OpenOffice or Mozilla... and waits and waits and waits on the same machine that previously ran Win95, IE and MS office just fine.
It would be nice if Linux had the clout to inspire people to upgrade their hardware like Microsoft does with every new version... but we don't. If we want carve out a beachhead for Linux on the desktop, we'll to make it run well out of the box on the older hardware that people are willing to try it on.
A group of us recently talked a local school into recycling some of their old P-III, 128 mb systems to Linux instead of scrapping them. It took a LOT of effort to unravel the bloat, provide a nice desktop and get the boxes to run Mozilla and OpenOffice without swapping. Requiring any kind of resident interpreter to support the basic desktop would have made this success story impossible.
Techie trends and toys are great. But if desktop developers don't factor in the end user experience, they'll be creating a lot of code that nobody will have the time or patience to want.
Please, guys... use efficient compiled languages for basic operating system components!
Kylix alternative
on
Kylix in Limbo
·
· Score: 2, Informative
It's a shame about Kylix. Fortunately there's an open source alternative.
The Lazarus IDE has made a lot of progress over the last year. It's built on the cross platform and self contained Free Pascal Compiler... so all a Lazarus app needs to install and run is GTK and a Linux kernel with elf support. This makes writing and packaging trans-distro apps a relatively easy process. Lazarus and FPC can currently produce full featured graphical apps on Linux and FreeBSD. A The Win32 version is also progressing nicely, for those who are interested.
The Lazarus IDE and Free Pascal Compiler are written in Object Pascal and compile themselves. The latest RPM's and source tarballs can be found at http://lazarus-ccr.sourceforge.net.
Slashdot Mirror
Ah yes, the never ending quest for new and better ways to bog down the Linux desktop... this time with a new load of resident interpreters. But I guess this shouldn't be a surprise... bloat is well on it's way to becoming a Linux standard.
New lanaguages and concepts are cool and help Linux to grow. New users do too. A lot of people try Linux on older hardware where they've been told it's supposed to work the best. Then it's "welcome to the modern distro" where new installs are frequently in swap before the default Gnome or KDE desktop finishes booting. Watch enthusiasm turn to grief as the new user clicks on OpenOffice or Mozilla... and waits and waits and waits on the same machine that previously ran Win95, IE and MS office just fine.
It would be nice if Linux had the clout to inspire people to upgrade their hardware like Microsoft does with every new version... but we don't. If we want carve out a beachhead for Linux on the desktop, we'll to make it run well out of the box on the older hardware that people are willing to try it on.
A group of us recently talked a local school into recycling some of their old P-III, 128 mb systems to Linux instead of scrapping them. It took a LOT of effort to unravel the bloat, provide a nice desktop and get the boxes to run Mozilla and OpenOffice without swapping. Requiring any kind of resident interpreter to support the basic desktop would have made this success story impossible.
Techie trends and toys are great. But if desktop developers don't factor in the end user experience, they'll be creating a lot of code that nobody will have the time or patience to want.
Please, guys... use efficient compiled languages for basic operating system components!
It's a shame about Kylix. Fortunately there's an open source alternative.
The Lazarus IDE has made a lot of progress over the last year. It's built on the cross platform and self contained Free Pascal Compiler... so all a Lazarus app needs to install and run is GTK and a Linux kernel with elf support. This makes writing and packaging trans-distro apps a relatively easy process. Lazarus and FPC can currently produce full featured graphical apps on Linux and FreeBSD. A The Win32 version is also progressing nicely, for those who are interested.
The Lazarus IDE and Free Pascal Compiler are written in Object Pascal and compile themselves. The latest RPM's and source tarballs can be found at http://lazarus-ccr.sourceforge.net.