Author: David Rasmussen
Date: 00:01:22 12/21/00
Go up one level in this thread
On December 20, 2000 at 22:26:46, Robert Hyatt wrote: >>> >> >>What about the other things I mentioned? They're very important IMO. >>And 3D IS a problem. Ask in any linux forum. > >Then I suppose you have to define "problem". I have two 3D machines running >here with no problems, using xfree. I don't pay a lot of attention to the >latest graphics gizzywhigs, but I do go for advanced things like the latest >SCSI or IDE devices, or latest MP motherboards, etc... They work with flying >colors. > You're mentioning things that work. I know that a lot of things work. They work under windows too. That's not very interresting. I could mention a billion times as many things that works. I'm talking about what is not working. There are things that aren't working, that should be working. That's a fact, and no one can deny it. > >> >>> >>> >>>>I'm not saying that all these are great products (winprinters certainly isn't), >>>>but still, you have a greater chance of getting things to work without problems >>>>on windows as opposed to linux. That's just a fact. >>> >>>The problem is you don't specify "greater". I would say that with linux >>>you have a 98.5% probability of getting any hardware to work. With windows >>>that goes all the way up to 99%. So greater? yes. But not by much. The >>>"without problems" might be debatable. >> >>You might get 98.5% of all hardware to work under linux, but it usually takes >>more effort, configuring etc. All this is usually very easy under windows, >>without losing any flexibility. Of course, this (and most of the other things >>I've mentioned) is not a problem of linux itself. Linux could easily have a very >>easy and flexible and coherent hardware frontend, but it doesnt. > > >I don't see the "much effort". I run linux on old machines, on brand new >machines, on esoteric (dual/quad processor machines). I haven't had a >hardware problem in several years. Since I went to redhat in fact. Every >SCSI card, ethernet controller, graphics card, tape drive, cd writer, etc, >have been recognized by the install program and set up perfectly. That is >one main advantage of the redhat release. > > Yeah, I've run many weird linux machines with esoteric hardware too. And much of it works. And a considerable amount doesnt, and some only work after fiddling with weird non-standardized configuration files, and patching some module etc. etc. Don't get me wrong. I like the Unix idea. I like many things about it. But the "everything has to be done by hand, and an obscure configuration file is always better than a nice consistent frontend"-thinking is obsolete IMO. It is only with the advent of tools like kudzu and linuxconf that I begin to see the light at the end of the tunnel. And they still don't cover all situations. Far from it. Hardware and software should preferably work when plugged in to the system at once. I get that experience much more from windows than from linux. > > >> >>>I just rebooted my quad xeon to update >>>the linux kernel. It had been up for 97 days when I did this. Our firewall >>>machine has been up for 244 days and counting. Our news server/ftp server has >>>been up for 74 days since a reboot to replace a tape drive. My beowulf cluster >>>of quad xeons has been up for 290 days and counting. I would say that says that >>>Linux will run without a "lot of problems". Our NT machines don't stay up that >>>long without losing touch with reality and having to be rebooted. >>> >>> >> >> >>You're talking about server capabilities again. Linux is superior here, I think, >>although I have only good experiences with Windows 2000 in a server environment. > >It doesn't matter whether linux is used in a server facility or as a >workstation, the reliability is very good. > > I was never talking about reliability. > >> >>> >>>Drag and Drop? That is part of xwindows already. Clipboard? Been there >>>forever (called cut and paste). >>> >> >>Well, it isnt used at all, compared to windows. > > >That's a rash statement. Maybe not used by you or people you know. Used all >the time by me and the people I know. > > I'm not talking about people. I am talking about a unified way of doing things so that a minimum of functionality is at hand. Don't tell me that twm has the same capabilities as the windows GUI. Because it's a lie. Now, I know that KDE and others are beginning to address this problem, and I'm sure it will be solved eventually, but it isnt now. >> >>>> I know that this can be >>>>done in the window manager, but that only leads to standard rot. There is >>>>nothing wrong with X windows supporting these modern and not so modern notions >>>>that make the GUI of even Windows much better in practice. >>> >>> >>>I'm not sure why you say they aren't there... >>> >>> >> >>Because they're not used. > > > >They are used by everyone I know here at UAB. We have hundreds of linux >machines and users. And even the ones that are not on linux are using unix >(SGI, IBM, etc) and are using x-windows which is 100% compatible.. >> Look at my above remark. I'm not talking about what people use. I'm talking about what's in the system and how it is standardized. >>>> >>>>As I've said, I know that linux is good for development and server applications, >>>>but if I was stuck somewhere with a computer and a net connection, and had to >>>>solve 1000 different tasks (not only development and server stuff), I'd sure >>>>hope it was a windows machine. I have numerous times been in a situation where I >>>>said "Damn, I have to do this or that, and I only have a linux machine at my >>>>disposal. Now I have to download and install and configure all sorts of really >>>>basic stuff, to get this problem solved. But first, I have to find it. And make >>>>sure that the individual software parts work together". Of course, I have had >>>>the same problem with windows, but then it's much less of a problem to find and >>>>download the stuff I need. >>> >>> >>> >>>That is pretty funny. I have been solving all kinds of problems for 30+ >>>years. For 25+ of those years, I have been solving _all_ my problems using >>>Unix. The facilities are so much better. From I/O facilities, to >>>process control, to you-name-it. I feel just the opposite. I would be >>>severely handicapped if I could _not_ use unix (linux) for program development. >>> >>> >> >>You're talking about developer stuff. I'm not. And also, I doubt that you would >>have any problems doing anything under windows, if you had the experience. Name >>me one thing that couldn't be done in windows, that is done in unix, that you >>would miss. >> > >Several come to mind. The "symbolic link" that two guys at microsoft claim >to have spent three years developing from scratch. Been in unix forever. >select() to be able to watch for I/O possible on several descriptors at one >time. The ability to access NFS file systems transparently. The ability to >trivially set up a network to use static IP, or DHCP, or in the case of >my notebook, whichever I want, and instantly. The ability to use UDP for >networking, in a way that is portable to all flavors of unix. > I live without it everyday. So does that majority of "barin-dead end users". >Development is what I _do_. I'm not a brain-dead end user that wants to keep >the computer at arm's length. I want it to do what I need, as fast as possible, >with as few problems as possible. > I don't care if you do development. So do I. I'm not talking about development. Linux and Unix is fine for development. I am talking about the difference in quality that you get as an end user. That might not be important to you. If so, you shouldn't have responded to this thread, because that is what it is about. Designing a system that can work with many processes, many users, networks etc. has been done. It was developed in the Unix community in the last 25 years. And windows borrows from that and sometimes makes a lousy implementation of what it borrows. But there is another thing that is important in a system. Usability. For everybody. That's a task that is equally important for productivity as is the basic system. And it is equally difficult to design. Don't think that what you do (development) is what smart guys do, and then the stupid guys can do the user interface, the consistency in hardware and software management etc. I am glad that you are not one of the smart guys developing usability in systems, be it end user or developer systems.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.