From owner-freebsd-chat Tue Mar 5 19:51: 5 2002 Delivered-To: freebsd-chat@freebsd.org Received: from mailhub.yumyumyum.org (dsl092-171-091.wdc1.dsl.speakeasy.net [66.92.171.91]) by hub.freebsd.org (Postfix) with SMTP id 5652437B4D6 for ; Tue, 5 Mar 2002 19:50:43 -0800 (PST) Received: (qmail 7512 invoked from network); 6 Mar 2002 03:50:36 -0000 Received: from dsl092-171-091.wdc1.dsl.speakeasy.net (66.92.171.91) by dsl092-171-091.wdc1.dsl.speakeasy.net with SMTP; 6 Mar 2002 03:50:36 -0000 Date: Tue, 5 Mar 2002 22:50:36 -0500 (EST) From: Kenneth Culver To: Terry Lambert Cc: Nate Williams , "Steve B." , "Eugene L. Vorokov" , Subject: Re: C vs C++ In-Reply-To: <3C857080.44C5236B@mindspring.com> Message-ID: <20020305224840.D7488-100000@alpha.yumyumyum.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Tue, 5 Mar 2002, Terry Lambert wrote: > Kenneth Culver wrote: > > I need to learn to say what I mean in a better manner. I've been trying to > > say the last comment for this whole thread and just couldn't get it into > > words. Thanks. > > Whatever. > > The bottom line is that the original poster is being paid > by someone to code, and he who pays the piper calls the > tune. If he doesn't like it, there are plenty of other > companies to work for, and if his manager doesn't like it, > there's plenty of people he can hire that *will* do the > code in the required language to meet the corporate goals > which led to the choice of C++ in the first place. > > C++ is a good mapping for problems that are subject to > object decomposition for solution, and in a design-before-code > environemnt (explains why you haven't seen good -- IYO -- Open > Source C++ code), it is easier to verify that the code matches > the design, and the correctness of the design, as well as being > able to use the design document 10 years later to successfully > maintain the code. > > FWIW, the University of Kentucky did a Bell Labs/USL > sponsoered OS research project called "Choices", written > in C++. It had a stacking VFS architecture implemented > using a pure virtual base class and inheritance. In the > demo for this code, I saw extended attributes and ACLs > added to a filesystem in less than 20 minutes. I have > yet to see FreeBSD's stacking VFS architecture handle > the cache coherency problem for getpages/putpages correctly, > without explicit coherency using read/write to implement > the stacking of a top level getpages/putpages, with the > associated decoherency in the mmap case not being fixed > for the msync() or for the file-and-mmap simultaneous access > cases. Well, I've never seen code written in C++ that well, which could very well be because I've only really looked at Open Source projects. Also, where I work, we use C, not C++, but I think that's mainly because we use FreeBSD, and practically nothing in FreeBSD is written in C++. Ken To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message