Date: Fri, 19 Mar 1999 13:34:35 -0800 (PST) From: Mark Diekhans <markd@Grizzly.COM> To: brett@lariat.org Cc: advocacy@FreeBSD.ORG Subject: Re: Netscape browser Message-ID: <199903192134.NAA19894@osprey.grizzly.com> In-Reply-To: <4.1.19990319134858.03fd24e0@localhost> (message from Brett Glass on Fri, 19 Mar 1999 13:56:16 -0700) References: <4.1.19990319114734.00b794b0@localhost> <4.1.19990319103804.00a8ec60@localhost> <4.1.19990319114734.00b794b0@localhost> <4.1.19990319134858.03fd24e0@localhost>
next in thread | previous in thread | raw e-mail | index | archive | help
>From: Brett Glass <brett@lariat.org> >Because the incremental cost of putting it on the disk is roughly zero, and >every user who has to call in and ask how to load it explicitly represents >a potential support call and hence an expense. The incremental cost is not zero, there is a lot more to manufacturing than cutting a CD-ROM (cheaper than dirt); its the testing of an additional platform. Having to test the emulator installation adds signficantly to this cost. If there where only one `Linux', it would probably be more costly than testing the application. Given that there are several `Linux' and `Linux' in practice is a testing nightmare, the cost of testing the emulator would be very high. > The user might even give >up and SWITCH to FreeBSD if the Linux vendor is not helpful. Hence, it is >in the Linux vendor's best interest to include the emulator. Huh? Why would a vendor be movated to make their product look bad? I doubt that they would be willing to incure the support cost of the emulator unless it was obviously outweighted by profit. >It's not a long shot at all; it's really the biggest chance FreeBSD has >to leverage the success of Linux. Linux emulation, by contrast, was the >long shot -- and in fact a very bad strategy. He who emulates, suffocates. I see little agreement with this assertion; and my experience doesn't agree either. >>A practical strategy for application development would be just that: >>a development environment that lets a single source generate native >>binaries for both Linux and *BSD. Not that I am proposing this, >>as it will still need market pressure to get it adopted and >>incure costs for support and manufacturing. > >The emulator could be part of that scheme. The object would be to >have the binary be a FreeBSD binary rather than a Linux binary. You are failing to see the difference; a development environment would allow producing the both binaries at lower cost than a runtime installation of the emulator. It would still be doubtful that this would be enough to get them to do the port without more market pressure. >Again, get it into the mainstream Linux distributions. How? Why would the Linux distributors do this? Has RedHat been asked? >>In my personal experience, Linux emulation has been one of the best faetures >>in terms of promoting FreeBSD to individual users. > >Alas, every time you do that, FreeBSD gets another strike against it >in the eyes of application developers. No, what I have another user who is using FreeBSD instead of Linux. (I did lose several to inferior laptop support). Granted, its not as good as having them use FreeBSD and FreeBSD native apps, but its better than using Linux and Linux native apps (which is better than Windozes and Windozes apps). >Remember: the popularity of a platform depends on the number and variety of >apps targeted explicitly for it. History has shown that very little else >matters. If we do not learn from history we are doomed to repeat it. I have been doing commercial software development for 19 years and my experience does not agree with this assertion. This is not saying that native applications are not important, simply that they are not as important as the availablity of applications in general and user base So where does the use of FreeBSD is 66% of Linux statistic come from? Mark To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-advocacy" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199903192134.NAA19894>