I mean, we all know that the conclusion is broadly true — the libraries and services available across Linux distributions make for a scruffy, frustrating and often unreliable target. That just doesn’t seem to me like the primary problem here.

I mean:

– it’s hard to port anything to or from Windows

– a modern web browser does a lot of stuff and is hard to port anywhere

– you have a few more decisions to make on Linux than on Windows (though really, not that many more — there are plenty of competing toolkits on Windows too)

– Chrome seems to do sandboxing at some level that I can’t pretend to understand anything about but that is apparently fiddly to do and needs kernel support

– Chrome developers have put a lot of work into making sure it feels shiny and nice and snappy etc to use on Windows, and much of that work doesn’t port at all

– you can’t please everyone

I’m not really seeing the part that explains whether, or why, it’s been any harder porting Chrome to Linux than porting it to OS/X, for example.

(Of course the one area where Linux really _is_ harder is in packaging and deployment, and they aren’t even talking about that yet! What joys we have yet to come.)

One thing I did find amusing about that Slashdot thread was seeing it descend once again into the argument about whether we were or weren’t sufficiently warned about the instability of KDE 4.0 when it first came out. I wonder when, or whether, that argument will ever stop?

Chris