From owner-freebsd-chat Wed Nov 28 23:34:54 2001 Delivered-To: freebsd-chat@freebsd.org Received: from guru.mired.org (okc-65-31-203-60.mmcable.com [65.31.203.60]) by hub.freebsd.org (Postfix) with SMTP id 76D3437B428 for ; Wed, 28 Nov 2001 23:34:46 -0800 (PST) Received: (qmail 16599 invoked by uid 100); 29 Nov 2001 07:34:39 -0000 From: Mike Meyer MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15365.58639.39658.89837@guru.mired.org> Date: Thu, 29 Nov 2001 01:34:39 -0600 To: "Anthony Atkielski" Cc: "Mike Meyer" , "Andrew C. Hornback" , Subject: Re: Feeding the Troll (Was: freebsd as a desktop ?) In-Reply-To: <01fe01c178a1$001d1be0$0a00000a@atkielski.com> References: <004801c17872$98e47b40$6600000a@ach.domain> <017f01c1788c$8cb71d90$0a00000a@atkielski.com> <15365.52562.394957.602907@guru.mired.org> <01fe01c178a1$001d1be0$0a00000a@atkielski.com> X-Mailer: VM 6.90 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid X-face: "5Mnwy%?j>IIV\)A=):rjWL~NB2aH[}Yq8Z=u~vJ`"(,&SiLvbbz2W`;h9L,Yg`+vb1>RG% *h+%X^n0EZd>TM8_IB;a8F?(Fb"lw'IgCoyM.[Lg#r\ Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Anthony Atkielski types: > Mike writes: > > Note that Apple examined an OS that didn't have > > all the multi-user complexity that Unix has, ran > > - like a bat out of hell - on Apple hardware, and > > on MP boxes with those processors. They decided > > against adopting it. > What operating system was that, and what was their reason for rejecting it? BeOS, and I don't know. > > Not to mention VMS. > I'm not familiar with VMS. What elements of NT are inherited from it or > influenced by it? Most of the internals. The lead designer for WNT had previously been the lead designer for VMS. Which is why WNT has the same relation to VMS that HAL has to IBM. > > No, it doesn't cost billions of dollars. > Yes, it does. Even some application systems can cost hundreds of millions of > dollars to develop. Be - including BeOS - was bought by Palm for US $11 million . Be's revenues were well under a million dollars a year . You can find financial information about them at ; none of the fiugres were over US$100 million per year. To the nearest billion dollars, BeOS was free. I know people who view writing "another little RTOS" for some custom hardware project as a minor part of the project; the single-board hardware design is the hard part. They are writing OS's for thousands of dollars. > > Sun, Apple and DEC have all done OS development > > projects that they abandoned. > Why did they abandon them? Well, for Sun and Apple because they decided that their wasn't a viable market for the product. For DEC, it was because the group that developed them was a blue-sky research group whose charter prevented them from using *any* current company product. There wasn't anyone on that coast who could give someone permission to run a competing product, no matter who they worked for. Though I did set up a Sun while at DEC, then watched the hardware designers disassemble it. > > I'm pretty sure Be didn't have billions of dollars > > to spend, and they put a very nice little OS on the > > market, though it wasn't suitable for much off the > > desktop. > Why wasn't it suitable for the desktop? You got that backwards - it was suitable for some peoples desktop, but it wasn't suitable for much else. It wasn't multi-user, and the hardware didn't scale beyond four processors. For me, it wasn't suitable for the desktop because they pretty much cloned the Mac desktop, which I can't stand. > > Of course not - the underlying OS is largely irrelevant > > if it provides all the needed functions. > So their choice of OS cannot be used to argue that it was technically superior > for the task. I'm not arguing that it's technically superior. I'm arguing that it's acceptable for the task at hand. > > Horse puckey. If you've used batch environments, you > > might be familiar with SAS, which communicated large > > volumes of data, but did it by passing a data set > > (or a small number of them) from one command to the > > next. > Sequential processing, not parallel. Yup, but it's still data analysis. > > Most - if not all - modern Unices allow multiple > > independent threads of control in a single address > > space. > Threads, or processes? Threads. Process don't share an address space. They can share memory if you need them to, but that's not as tightly coupled as threads in the process, and you wanted tightly coupled communications. > > I recommend that most of my relatives run Windows, > > not FreeBSD. > So do I. And when they ask about Linux, I discourage them in the strongest > terms. If you can't support a Unix desktop, and can't recommend someone to them who can, that's probably the best thing to do. If there's a local Linux users group, you might point them at that with a warning that if they get in trouble, they'll have to get help from the users group, not you. That's *especially* true if they are complaining about problems in the Windows GUI. > > Come on - what do you want to do on your > > desktop that you can't do on FreeBSD? > I have a hundred or so applications on my desktop that will not run under UNIX. That means Unix is an inferior processor for those applications, not that it's an inferior desktop. Which ones can you not find an acceptable alternative for on Unix? But I think I already answered that question. > Additionally, I can assign permissions via ACLs in NT, to both files and > objects. I don't have to run anything special to get a GUI, since that is the > native environment. Wait a minute, I thought all that multi-user protection stuff was *bad* for a desktop. > > I have as yet to see you list a single thing you > > wanted FreeBSD to do as a desktop that it couldn't. > How do I start Adobe Illustrator on FreeBSD? With the command "gimp". They changed the commands and UI, though. > > For a real giggle, consider that I've seen > > Windows running on the MIPS processor running > > with the opposite endianness of the MIPS boxes. > I saw a MIPS machine in a lab once; I don't recall if I actually used it. It > was running NT. MIPS didn't sell very many of them, though SGI used the processor, then bought the company to make sure they could get enough of them. For a while, the second most popular Unix workstation on the market was the DEC MIPS-based box. They swapped the endianness of it to match their other hardware. We did an in-house port of NT to that box in the early 90s, before DEC dropped the product line in favor of the Alpha. > > The same was true of OS/2. > Yes, but OS/2 was designed around the MS-DOS paradigm, which was already dying. > NT was designed around the Windows paradigm. IBM wanted the MS-DOS look and > feel, and Microsoft wanted the Windows look and feel, and that's why they parted > ways. The rest is history. Since paradigm means pretty much whatever you want it to mean, that statement is true, but empty of meaning. http://www.mired.org/home/mwm/ Q: How do you make the gods laugh? A: Tell them your plans. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message