[ros-dev] list macros roundup

Alex Ionescu ionucu at videotron.ca
Thu Sep 29 02:53:15 CEST 2005


Gunnar Dalsnes wrote:

> The list macros are widely used in both Wine (4 different macros) and 
> in the linux kernel 
> http://www.gelato.unsw.edu.au/~dsw/public-files/kernel-docs/kernel-api/r802.html 
> (i counted 16 different list walking macros).

I hope the Linux Kernel won't ever become an example of how an NT 
micro-kernel shoudl be written..

>
> A googled for a while looking for similar discussions about using such 
> macros or not, but found none...

Really? Let me point you to some of the famous threads by our Microsoft 
friends, including some of their most distinguished engineers:

http://blogs.msdn.com/larryosterman/archive/2004/12/01/273160.aspx

"

     Larry: You said Macros work to hide the complexity and say so like
    it is a bad thing.. ? Excuse me but I thought that was the POINT of
    using a Macro..

Actually, in the world in which I live (writing systems programs that 
exist in the working set of hundreds of millions of users), hiding 
complexity is a very, very bad thing.

You need to be VERY careful whenever you do something that hides 
complexity, because it's likely to come back and bite you on the behind."

Also: "Rules of software engineering" touches on this:
http://blogs.msdn.com/larryosterman/archive/2004/04/06/108686.aspx


>
> Seems like people who hate macros are the people who are drawn to reactos.

Seems like people who hate macros are the people who are drawn to NT, 
too. I'm glad 99.9% of the developers here are on that line of though.

> I dont know what this means but it surely is weird. I hope it doesnt 
> mean that reactos coders are too stupid to learn macros;-P

Yeah, go tell Larry Osterman he's too stupid to learn a macro.

>
> Well, at least i tried. Better luck next time (/me pat self on shoulder)

Read Larry's posts. Read them again. Read the comments by other MS 
engineers. If you still think hiding complexity is a good thing, God 
help you.

>
> G.
>
Best regards,
Alex Ionescu



More information about the Ros-dev mailing list