Linked by Hadrien Grasland on Mon 28th Feb 2011 11:23 UTC, submitted by Joao Luis
Thread beginning with comment 464435
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[11]: So long, and thanks for all the fish
by _txf_ on Tue 1st Mar 2011 16:19
in reply to "RE[10]: So long, and thanks for all the fish"
Please, do provide specific examples.
Init Systems and Package Management Come to mind..
Well, depends on what you consider to be "soon" and Gtk support can clearly be marked as legacy and/or deprecated, much in the same way Apple had supported Carbon on OSX, even though Cocoa was the right way to do it. Clearly it can be done.
The problem is that application developers would have to start out with a gtk api then have Qt apis. So some apps would be using gtk libs and others Qt libs. And then you have both libs loaded on a memory and power constrained device
Also I have a particular problem with the "we want customers, not developers" attitude that many companies exhibit. They assume they are in it alone and that they ship a closed and boxed product that you throw away after having used it more or less without modification. This was true for dumb and feature phones, but not at all for smart phones. With smart phones, people don't just want the "out of the box" experience, they download software, modify settings and often do crazy things with their hardware. That's the difference between platform and ecosystem. The aforementioned attitude is exactly what got Nokia into this mess.
Problem is that to get the best developer output you have to provide them with a sane base to work with. This means no deprecated apis at the start of a products lifetime (that is absolutely insane but it is what you're proposing), no conflicting apis or two apis to do the exact same thing, sensible apis, a sane development environment, decent documentation. Qt has all of those things, keeping gtk in the forefront would have just muddied the waters.
RE[12]: So long, and thanks for all the fish
by spiderman on Tue 1st Mar 2011 16:27
in reply to "RE[11]: So long, and thanks for all the fish"





Member since:
2007-04-18
Is there even system wide support for rotation, a modern kernel, recent X?
I agree that it is unmaintained, but "old" is a very relative word. Debian can easily bridge package version gaps far larger than ~1 year.
Please, do provide specific examples. My experience is that I've ported many years old and unmaintained software to newer systems with a mere recompile (if the binary didn't just work out of the box anyway), even between fundamentally different operating systems (Linux, Solaris, Windows, etc.). If you plan ahead and don't tie yourself to too many platform specific features, porting becomes relatively trivial.
Well, depends on what you consider to be "soon" and Gtk support can clearly be marked as legacy and/or deprecated, much in the same way Apple had supported Carbon on OSX, even though Cocoa was the right way to do it. Clearly it can be done.
Also I have a particular problem with the "we want customers, not developers" attitude that many companies exhibit. They assume they are in it alone and that they ship a closed and boxed product that you throw away after having used it more or less without modification. This was true for dumb and feature phones, but not at all for smart phones. With smart phones, people don't just want the "out of the box" experience, they download software, modify settings and often do crazy things with their hardware. That's the difference between platform and ecosystem. The aforementioned attitude is exactly what got Nokia into this mess.