Date: Mon, 22 Mar 1999 18:14:59 -0700 From: Brett Glass <brett@lariat.org> To: "Jordan K. Hubbard" <jkh@zippy.cdrom.com> Cc: advocacy@FreeBSD.ORG Subject: Re: FreeBSD emulation for linux Message-ID: <4.2.0.32.19990322180649.040bf100@localhost> In-Reply-To: <44316.922150448@zippy.cdrom.com> References: <Your message of "Mon, 22 Mar 1999 16:15:28 MST." <4.2.0.32.19990322160933.00aaf6c0@localhost>
next in thread | previous in thread | raw e-mail | index | archive | help
At 04:54 PM 3/22/99 -0800, Jordan K. Hubbard wrote: >> The reason why it ultimately stopped working so well was that Microsoft >> moved the API out from under them. > >At which point they should have fixed it, regardless of the degree of >technical difficulty involved. I wouldn't just say that off-handedly >about anyone, but I find it hard to believe that IBM, with all of its >incredible resources and arsenal of software patents (for convenient >cross-licensing purposes), couldn't have kept up with the football as >Microsoft attempted to kick it around the field. It involved the use of entire libraries of code that IBM didn't have access to. It wasn't just a few APIs! What's more, to implement it would have violated OS/2's protection model, allowing Windows apps to crash the whole machine. The Win32s and Win32 architectures were incredibly dirty and nasty and could not be virtualized. >> Well, quite frankly, Jordan, your lack of support for it could well >> poison the effort. > >I have pointed out that it's technically difficult to do something >like this given the differences between the FreeBSD and Linux APIs. >This is no more than correct. Could it be any harder than going the other way? >I have pointed out that it would need buy-in by the major Linux >distros before this could be the kind of out-of-box solution ISVs >would want it to be before targeting the Linux platforms with FreeBSD >native binaries, and such buy-in isn't trivially contemplated. It's easy to lay the strategic groundwork for this. Again, you're being negative, and as "spiritual leader" of FreeBSD you're dampening enthusiasm for something that's NECESSARY. >I have pointed out, just now, that I simply don't see the kind of >talent it would require being available. Motivation is great and >technical abilities are great, but they need to be enbodied in the >same person before you get movement here. What's happened to the people who were working on Linux emulation? > You've also suggested that >the Linux emulation team would be the ideal candidates when I happen >to know that everyone on the Linux emulation team is busy up to their >eyeballs and very definitely *not* even motivated in this direction >(and if Soren subscribed to advocacy, you'd have been flamed to toast >by now for even proposing him in absentia). This is no more than >silly. If they're not motivated in this direction, Jordan, then you should help to get them motivated. This is needed. >If pointing out glaring flaws in a proposal truly counts as >"poisoning" it then I would submit that it also constitutes a mercy >killing. None of the things you've pointed out are glaring flaws, IMHO. However, again, as the "spiritual" leader of FreeBSD, you can kill even good ideas. Sad but true. >Many of your suggestions would be far motivational, in fact, if you >just took the time to actually do some of the "devil's advocacy" >yourself before committing them to paper (so to speak) and then having >everyone else shoot noisily at the more obvious holes in them. Even >advocacy can sometimes stand a little bit of the old vigorous >scientific inquiry process (before publication) when it starts getting >into specifics, as you've been wont to do lately, and just shooting >ideas from the hip and seeing what they hit afterwards doesn't quite >cut it when you're getting into the area of suggestions which are >actually technical in nature. What sort of inquiry would you recommend? There's nothing about this project which is fundamentally different from emulating Linux on FreeBSD, so technical feasibility is not in doubt. --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?4.2.0.32.19990322180649.040bf100>