Date: Fri, 07 Jul 2000 21:09:54 -0600 From: Brett Glass <brett@lariat.org> To: Jeroen Ruigrok/Asmodai <asmodai@wxs.nl> Cc: "Jordan K. Hubbard" <jkh@zippy.osd.bsdi.com>, Narvi <narvi@haldjas.folklore.ee>, Dann Lunsford <dann@greycat.com>, chat@FreeBSD.ORG, advocacy@FreeBSD.ORG Subject: Re: No port of Opera? (Was: ((FreeBSD : Linux) :: (OS/2 : Windows))) Message-ID: <4.3.2.7.2.20000707203753.04804100@localhost> In-Reply-To: <20000707211621.C35215@daemon.ninth-circle.org> References: <4.3.2.7.2.20000707003817.04760d40@localhost> <4.3.2.7.2.20000706225433.0475b4d0@localhost> <Your <4.3.2.7.2.20000706193313.04a8ca40@localhost> <54201.962945434@localhost> <4.3.2.7.2.20000706225433.0475b4d0@localhost> <20000707082704.Z35215@daemon.ninth-circle.org> <4.3.2.7.2.20000707003817.04760d40@localhost>
next in thread | previous in thread | raw e-mail | index | archive | help
At 01:16 PM 7/7/2000, Jeroen Ruigrok/Asmodai wrote: >We are not emulating Linux. We are merely mapping the Linux syscalls to >appropriate BSD ones. Please use the correct terms for that, since this >is essential to the debate you wish to carry. To me, performing that mapping (with thunking in some places!) is what I would consider to be emulation. Granted, it is a "light" form of emulation, since the APIs are close. In any case, the discussion here doesn't hinge on what one calls this facility. I think you'll agree that WHATEVER it's called (call it a Frobnicator if you'd like), the effect on the availability of natively compiled applications is the same. (And whatever you call it, if you tout it as being especially efficient at running Linux binaries, this will make developers MORE likely to tell FreeBSD users to run their Linux versions rather than porting their software.) >With the help of the linuxulator we got it easier to ride the hype which >Linux created. Any "riding of the hype" which has occurred has been, IMHO, utterly independent of the existence of the Frobnicator. The people who I see switching are using FreeBSD as an Internet server and are not, as a rule, running desktop or commercial applications. My clients who run such things as office suites (e.g. Corel Office) are sticking to the OSes for which they are natively compiled so that they can get support. >Now that the hype is reaching its momentum it becomes obvious that >FreeBSD/BSDi is being a major player in the whole ballgame as well. >Have you looked around recently? There's more and more mention and >support for the FreeBSD cause (and indirectly for OpenBSD and NetBSD). I am making an effort to generate those mentions and create that support, and so are others. But FreeBSD is not even really at the status of "second fiddle" yet; it's more like "third viola." >We are here, we are demanding our portion of the marketshare, there's no >denying us. You cannot "demand" market share. Rather, you must offer things of value. And one of the most important things of value -- the thing that has historically proven to make or break a platform -- is native application support. >Now tell me again, how did the linuxulator hurt us again? By encouraging developers to support only Linux. It's Linux on the box; Linux in the ad; Linux on the support line. "Oh, you're running FreeBSD? Sorry, Sir, but the box says 'Linux.' If you're trying to run it on something else, we can't provide you with tech support." >Companies >will always work according to the demand/supply principle, so the >userbase needs to, politely, ask Opera to release a _native_ port. "Asking politely" does not constitute demand. PAYING CUSTOMERS constitute demand. Now, without emulation, those customers have clout equal to their numbers times the price of the product. But with emulation, this clout is diminished by the number of "defectors" who are willing to run the non-native binary. Worse still, the vendor can't tell who bought the product to run it under emulation. So the marketing numbers show sales onto the emulated platform as larger than they should be. And the number of users who want the native port therefore seems smaller. It's a vicious effect: ALL of the incentives run against the native port. >We pulled this off with Oracle (IIRC) and we also amassed tons of votes >for a native JDK and for more products, and we got them and/or are >getting them. Maybe. But will they deliver? And will they support it for more than one version? I am hoping that we can keep Applixware and Xi Graphics. (As you may recall, a representative of Xi Graphics told me emphatically that they were "dropping all support for BSD" about half a year ago. It turned out that she was mistaken and they were dropping BSD/OS only, but it this is still not a good thing -- especially as BSD/OS and FreeBSD move closer together.) Bottom line: the Linuxulator/Frobnicator/Prognosticator/Whatever is hurting the platform. I've proposed a strategy to counter this. Shall we take action? --Brett 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?4.3.2.7.2.20000707203753.04804100>