On Wed, 28 Jan 2015 23:36:29 +0000, deadalnix wrote: > On Wednesday, 28 January 2015 at 23:22:34 UTC, ketmar wrote: >> On Wed, 28 Jan 2015 18:54:27 +0000, Zach the Mystic wrote: >> >>> I think a keyword is a keyword is a keyword. If it's a keyword to the >>> right it should be one everywhere. How is somethign that's a built-in >>> attribute one place and an identifier in another not context >>> sensitive. >> >> yep. that is "slave to the machine" approach. i don't really care how >> hard machine should work to understand what i want. ;-) >> >> i.e. i can't see why i have to deal with problems of >> compiler/tool/editor author. it's easy -- at least in this case -- for >> human to see where keywords are keywords. machine can see that too with >> some analysis. yet somehow situation is horribly reversed: instead of >> machine helping me to do what i want, that's me who must obey the >> machine orders and do the work that machine can do for me. > > That is a retarded mindset. It is not about how hard it is for the > machine, but for tool writer. > > I know many people look down on java, but quite frankly, the tooling is > just way better than pretty much anything else. One could argue this is > because of corporate support, but other languages like C++ also have > important corporate support. And still, C++ tooling mostly suck. > Honestly, pretty anything except C# tooling suck in comparison. > > The problem of the tool writer is your problem, because any tool that > isn't written is work you need to yourself. And then who's the slave to > the machine ?
and there are at least 4 available codebases for writing tools. for C++ there is DMD frontend. for D there is dscanner, magicport and SDC. yet instead of thinking how all that code can help me and made my life better, i should think how making my life harder will help to write great tools somewhere in the future. that is what i call "retarded mindset".
signature.asc
Description: PGP signature
