Date: Fri, 6 Jan 1995 18:50:28 +0100 From: "DI. Christian Gusenbauer" <cg@uhura> To: "Jordan K. Hubbard" <jkh@time.cdrom.com>, hackers@freebsd.org Subject: Re: guest account: Yggdrasil information Message-ID: <199501061750.SAA00134@uhura> In-Reply-To: "Jordan K. Hubbard" <jkh@time.cdrom.com> "guest account: Yggdrasil information" (Jan 4, 9:53pm)
next in thread | previous in thread | raw e-mail | index | archive | help
> This will take a little while to read and believe me, I know how much > mail most of you get, but it's nonetheless worth slogging all the way > through it. Yes, most of it is unabashed sales hype. Ignore that and > concentrate on the features list. I have this CD and it's not lying > about the degree of flexibility it provides. NetBSD is not the > competition. This is. When and where we can match some of these > features (and I not only think we can, I KNOW we can!), we should > strive to do so! I certainly will be! Hi Everyone! After reading this forwarded info about the features, watching discussions here at the University, speaking with friends I think it's time to plan the future of FreeBSD. If we plan to support these features mentioned in that article, we should think about how FreeBSD will present itself to the future user. Here at our institute, many people fell in love with Micro$oft's NT (not me!), because it's so simple to install and to administer. Just clicking around and it works. Even our hardware-technician, who doesn't know what TCP/IP, NT services, etc exactly are, can configure these things. So I think, if FreeBSD should have a future, we'll have to do some work: 1. We have to make configuring as easy as possible for those users, who never want to look 'behind the scenes'. There should be a way to install and configure FreeBSD for *non-experts*! 2. It's a matter of fact, that all popular OS use GUIs (OS/2, Windows, Apple, NextStep, ...). So why don't we use X right after the installation to communicate with the user (yes, I know X needs lot of space, but if we shrink X to a minimum it's perhaps possible?). So install a minimum OS on the HD, copy a small X version to the disk, and start the X server. I also think, that - in future - only a couple of users will install an OS from disks. The installation medium will be the CD-ROM. 3. All configuration (administration) tools should look and work the same way. I myself hate to have 10 tools and 10 different ways to install these things (eg. some use a DOS program, some need MS Windows; or: look at named, inetd, nfs or even passwd, where I can configure every tool when I understand and know the correct syntax of its ascii-file: it's impossible for a novice user to work with that). 4. Plug'n Play: that sounds fine to me, but I do also know, that there are many cards which can't be installed like that. There may be problems de- tecting the hardware or something else. But I think the autoconfiguration mechanism of FreeBSD is the first step, where we should start working. It sounds good: autoconfiguration! But why do I have to fill in IRQs, DMA, Memoryadresses etc in the config-file of a kernel, when there is an autoconfiguration?? No, I'm not a frustrated user; I'm really fascinated from UNIX especially FreeBSD. But I think, if we want to compete with LINUX, OS/2 and Windows NT, there is a long way for us to go, and we should start walking at once ;-). Please don't flame me for my thoughts, but if we discuss how the installation procedure for 2.1 should look like and what the features of LINUX are, we can even discuss how FreeBSD 3.0 or 4.0 should look like and direct the current work in this direction. As long as I have to recommend OS/2 or Windows NT to my friends, FreeBSD or Linux is lost in space ;-). > > Thanks! > > Jordan Christian. cg@fimp01.fim.uni-linz.ac.at
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199501061750.SAA00134>