Date: Fri, 26 Mar 1999 00:21:59 -0500 From: Brian Adkins <brian@lojic.com> To: Brett Glass <brett@lariat.org>, advocacy@FreeBSD.ORG Subject: Re: Swan song Message-ID: <4.1.19990325235434.00a39f10@mailbox.iwaynet.net> In-Reply-To: <4.2.0.32.19990325203830.00a2b440@localhost> References: <85720.922223339@zippy.cdrom.com> <Your message of "Tue, 23 Mar 1999 10:57:21 MST." <4.2.0.32.19990323104122.009e8c70@localhost>
next in thread | previous in thread | raw e-mail | index | archive | help
At 09:43 PM 3/25/99 -0700, Brett Glass wrote: Well you certainly know how to get my attention Brett, just mention Java. I have a few comments below, but before that, I have to say that even though I'm still doubtful that the existence of a FreeBSD emulator on Linux would accomplish what I think we all want to accomplish, I must confess that when I read Jordan's statement for the very first time (prior to even installing/using FreeBSD) encouraging ISV's to write to Linux, I was very surprised. As someone evaluating the OS, it struck me as an odd thing to say. I'm not saying that it's a bad thing for ISV's to do, but I think many individuals, or companies, might interpret that statement as an endorsement of Linux over FreeBSD (that was my *first* impression). Promoting the fact that we can run Linux binaries may be perceived as a strength of the OS; whereas, encouraging ISV's to write to Linux may be perceived in a negative way. Yes, companies will probably draw that conclusion anyway given the numbers, but that's different than the FreeBSD community encouraging writing to Linux. Sometimes the subtle things are important. I'm not really commenting on the emulation issue because that's been beat into the ground, but I am saying that perception is very important and we need to think carefully about the image we want to promote. >... >Now, let's contrast this with Java. Java doesn't emulate anything else; >rather, the Java Virtual Machine makes other platforms run Java binaries >and emulate the Java API and ABI. As a result of this, it is succeeding >despite the fact that it is awkward, clunky, and slow to load. Have you actually written any significant Java code? Your last sentence would indicate not. I think many Java developers would attest to the elegance of the language compared to C/C++. >AND despite >the fact that it has been badly marketed. If by badly marketed, you mean over-hyped, then maybe, but otherwise there has been a significant and successful Java marketing campaign Sun, IBM and others. When my computer illiterate father asks me if I'm using "this new Java language" in my products because he saw something on CNN and "it's the wave of the future", it tells me that Java has been effectively marketed (maybe too well). >AND despite the fact that it has >borne the full force of Microsoft's endless PR dollars, which have been >spent liberally on a campaign to defame, fragment, and marginalize it. >(Just the hiring of J++ architect Anders Hejlsberg from Borland cost >Microsoft a 7 figure sum, according to reports from his former co-workers.) >AND despite the fact that Java has to add security to platforms that don't >have it. AND despite the fact that Java started with an installed base of >zero long after Windows was dominant. AND despite the fact that developers >had to learn an entire new language, application framework, and class >hierarchy in order to use it. Actually, the fact that Java is so incredibly similar to C++ is a very important key to it's success. I was able to make use of many years of C++ programming when I began coding in Java. If it had been an entirely new language, I would not have been as quick to adopt it. >AND despite the fact that many key parts >of the application framework were late or missing. AND despite the fact >that Java to this day isn't really "write once, run anywhere" unless >you're very good and very, VERY careful. Maybe, but I'd much rather "port" Java than C/C++. I've done *a lot* of both, and it's much easier. >FreeBSD emulation for Linux (and, for that matter, other UNIX-like OSes and >even NT) would not have anywhere near this many hurdles to overcome. It >would encourage the development of hundreds -- even thousands -- of native >applications and become a stabilizing force in the raucous UNIX world. And >developers could do what they could not do with Linux: Use the time-tested >code of the utilities as the foundation of their own applications. > >We must therefore ask ourselves : Is FreeBSD to be an OS/2 or a Java? Well, let's at least not miss out on providing the best free OS for Java. >I was interested, strongly, in trying to make it the latter. But since >at least some people on this list seem not to be able to learn from the >past nor to develop a version from the future, it will likely be >fruitless to try. > >Now, I'll readily stand up to flames from all sides when advocating >a product (and believe me, I've had asbestos underwear handy ever since >I first used Usenet more than 15 years ago). But when you encounter >total negativism from the person or persons, you are trying to do >doing advocacy FOR, it's time to leave. Brett, I'm not commenting on whether the criticism was warranted, or not; however, I *do* think it isn't anywhere near as tough as the cold hard reality of architecting, designing, programming, marketing etc. your idea. If you quit this easily with a little criticism, you surely wouldn't last long in actually creating the thing. If you believe in it, stand up for it! Someone once said, "When people of integrity are told it can't be done, they quietly push on and let time prove them right." >So, farewell. I'm sure you'll manage to run FreeBSD into the ground >without me. I realize this is probably a statement made out of frustration, but it really doesn't help accomplish anything. >--Brett Glass 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?4.1.19990325235434.00a39f10>