From: Digestifier To: Linux-Misc@senator-bedfellow.mit.edu Reply-To: Linux-Misc@senator-bedfellow.mit.edu Date: Sat, 21 Aug 93 03:33:30 EDT Subject: Linux-Misc Digest #34 Linux-Misc Digest #34, Volume #1 Sat, 21 Aug 93 03:33:30 EDT Contents: Re: NT versus Linux (Howlin' Bob) Anyone get FlexFax working with a Class 1 modem under Linux? (Jeffrey Perry) Re: NT versus Linux --- mem needed for X11 (Jim Graham) Re: NT versus Linux (Howlin' Bob) Re: NT versus Linux (Bryson Lee) NT vs Linux ? why? (Paul Cardwell) Re: Modem XTERM emulation? (Gary McTaggart) Re: NT versus Linux (Steve Sheldon) ---------------------------------------------------------------------------- From: gt8134b@prism.gatech.EDU (Howlin' Bob) Crossposted-To: comp.os.ms-windows.advocacy Subject: Re: NT versus Linux Date: 21 Aug 93 05:40:29 GMT drew@romeo.cs.colorado.edu (Drew Eckhardt) writes: >For most people, a microkernel is going to do more harm than good. With I'm rather tired of seeing the microkernel listed as a feature. I'm glad Linux isn't a microkernel or a message-passing OS. There is no *inherent* value of having a smaller "kernel" and having the rest in "processes." Look at Mach: they're technically amazing, but on single-CPU machines, they're always playing performance catchup to the monolithic systems. >>>>kernel size 50k 350k >> Doubt it very much >The 50K figure doesn't include device drivers, filesystems, etc, while >the Linux figure does. It's interesting that the 50k figure is harped upon. Linux runs in 4 MB like a dream (I know, most of my time using Linux has been on a 4 MB machine). From what I hear, the NT betas thrashed the disk all to hell with even 12 MB. >Unless Microsoft used a "bignums" package for all the internal >structures (unlikely because of performance concerns), they can't >be unlimited, although they may be for "all practical purposes" >unlimited as Linux is. Well, they probably use floating point offsets :-) integer bignums would be interesting, though. >>>>memory model flat internally segmented You *always* use segmentation on the 386. You can't avoid it. The TSS is a segment. The IDT is a segment. You must have valid segments for cs,ds,es,fs,gs,and ss. You simply cannot run without segments. The process memory model is not segmented: it sees a flat 3 GB address space. The kernel sees a flat 1 GB address space, but it can use segment overrides to access the process's address space. >>>>protable yes no (x86 only) >> debatable >Linux has been ported to the Amiga. Is being ported... it's still rudimentary. >>>>runs DOS apps yes no >Linux runs *some* DOS applications under the DOS emulator. Hey, I'm working as fast as my little fingers can go. >As far as realistic configurations, Linux works well, with X (provided that >you've tweaked things properly) and 8M of memory I can do builds without >paging. Technically, no. Every executable you use has to be paged in. I think the incorrect term "swapping" is actually more correct here. Linux can simply "release" executable pages without having to copy them to swap; there is still some memory reclamation at work. -- Robert Sanders Georgia Institute of Technology, Atlanta Georgia, 30332 uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt8134b Internet: gt8134b@prism.gatech.edu ------------------------------ From: j.perry@lynx.dac.northeastern.edu (Jeffrey Perry) Subject: Anyone get FlexFax working with a Class 1 modem under Linux? Date: Sat, 21 Aug 1993 05:08:44 GMT Has anyone gotten FlexFax to compile under linux when configured for a Class 1 Fax modem? If so could you please email me any changes that were necessary. Thanks in advance! I have not tried to compile yet but wanted to check if anyone knew it was possible or was not possible...before spending time on it. Please respond via email and I can summarize. Thanks again! jeffrey perry ------------------------------ From: jim@n5ial.mythical.com (Jim Graham) Crossposted-To: comp.os.ms-windows.advocacy Subject: Re: NT versus Linux --- mem needed for X11 Date: Fri, 20 Aug 1993 13:34:23 GMT I won't touch the majority of the errors in this ``feature'' list, which was obviously written by someone who has never used Linux---most of them have been thrashed about to the point where I'm sure people know about the errors.... But there are a couple of things that really shocked me.... In article <1993Aug19.224831.30686@kf8nh.wariat.org> bsa@kf8nh.wariat.org (Brandon S. Allbery) writes: >In article <250hid$bai@bach.seattleu.edu> aehall@calvin.seattleu.edu >(OUTTA HERE!) writes: >>In <930818233207.23008@lambada> ed.duomo@lambada.oit.unc.edu (Cousin Ed) >>writes: >>>Feature NT Linux >>>------- -- ----- >>>min required mem 12 meg 16 meg (w/ X) 16 Meg??????? Who dreamed this up? I have 4 Meg in my machine, and I can tell you now that it runs rather nicely, thank you. >> I use 8 with X just >> fine. > >Use fvwm and rxvt and 4 Meg is enough! Take *that*, Microshaft! I run X (XFree86 1.2) in my little 4 Meg sometimes, myself...and I do *NOT* use fvwm---I use twm. I do use rxvt for a couple of xterms that I don't use much (e.g., I keep a window open that I just use to monitor UUCP stuff), but I use the real xterm for anything where I plan to actually do much. It's a bit slow sometimes (not always, though), but it does work. Of course, I've gotten so used to just using the virtual consoles and selection (for snarf-n-barf) that I don't really use X unless I need the graphics aspects of X. >>>min required disk space 60 meg 120 meg (all series) Wait a minute....hold on here! Are you saying that without 120 Meg, I can't run Linux? My entire hard disk isn't even that big, and I still have a small dog partition (I have a shareware program that I still need to compile every now and then, and I'm still using emTeX for its 24-pin DVI driver, plus, well, games ). Oh, and of course, you can have a limited system on just one 1.44 Meg floppy (basically just good for demos, admin., maint., etc.).... You'll notice that I haven't said anything about NT---there is a reason for that....I don't want to be guilty of the same thing that [whoever made up the list that started all of this] is apparently guilty of. In other words, I don't want to start making claims about an OS that I've never even seen, and have virtually no facts on. 16 Meg of RAM.... 120 Meg of disk.... Geez.... Next thing you know, someone will be saying that msdog no longer has any of these prehistoric ``features'' like 640k limits, 64k barriers to some data structures, different memory models, etc.---yeah, right. Umm, while we're at it, I have some great land for sale. Really nice location---it's about 25 miles south of Ft. Walton. {\tinyprint Submarine and/or scuba costs extra.} :-) --jim -- #include 73 DE N5IAL (/4) ==========================< Running Linux 0.99 PL9 >========================== INTERNET: jim@n5ial.mythical.com | j.graham@ieee.org ICBM: 30.23N 86.32W AMATEUR RADIO: (packet station temporarily offline) AMTOR SELCAL: NIAL ============================================================================== E-mail me for information about KAMterm (host mode for Kantronics TNCs). ------------------------------ From: gt8134b@prism.gatech.EDU (Howlin' Bob) Crossposted-To: comp.os.ms-windows.advocacy Subject: Re: NT versus Linux Date: 21 Aug 93 06:00:13 GMT brileary@world.std.com (Brian Leary) writes: >0.1%: A handful of people that do the actual linux development and >application writing. Those are descent humen beings, and other than >being fanatics (ALL LINUX USERS ARE!), I can say no ill about them. >Not very competent, but otherwaise hard working. They rarely post or >participate in the linux groups, and are mostly silent. This is an amusing paragraph. If I were a violent man, I'd tie you down and tattoo it on your belly with a hot poker. It's that amusing. Now, see, *that* is fanatical. I'm a lot more laid back. I was just trying it out. >I think that's pretty much an accurate and objective breakdown. Sir, I do not think you understand the meaning of the word "objective." >> As someone else posted (and I checked), QIC-80 >For your information, it has been in alpha for over a year. All that >was done is that the MACH code was placed in a linux site, that's all. No, the new QIC-80 driver actually runs under Linux; people are actually using it to backup their systems. And, amazingly enough, the QIC-80 driver loads via the Linux loadable device driver mechanism (which is, admittedly, not included in the Linus-distributed kernel). >>>Of course, there is also the FANATIC MENTALITY, but that one is >>>obvious. >>This seems to be true of you as well. >> >No so. I will say upfront that linux has the best chance among all >unices of surviving (mostly as a toy os). All the commercial ones will I do real work under Linux. I simply do not understand how that can be considered a "toy" OS. Could you define this term? If it means "not in shrink-wrap," which is mainly how I see it used, then I agree. But if it means not really useful, I disagree. >be extinct in a mmater of a few years, due of course to the new unix: >NT. DEC and Silicon Graphics started the trend, and others will follow "The new unix" sounds more fanatical than anything I've heard from a Linux user. Sounds like a mantra to me. >>>Observation #3: >>>For a long time they kidded us about how NT is vaporware. The general >>>consensus among them is that NT will never see the light of day, and >>>our replies that it is coming out "soon" were laughed at. >> >>What do you expect? Windoze NT is *already* behind schedule. It was >>supposed to be released several months ago. We can't be responsible for >>the bogus claims of others. >> >What a joke! Linux 1.0 was supposed to be released last summer. In >December, SLS 1.0 was announced, under the impression that linux 1.0 >would be released by then. Well, we're still waiting. NT was late? >Hahaha! Hmm. I suppose your followup to this will claim how advanced NT is because it started at version 3.1? Methinks you are too susceptible to numbers. Try this trick: remove the leading "0." from the Linux version numbers. Relax. You are getting sleepy. Now repeat after me: Linux is at version 99. It is the singlemost stable and mature OS in existence. Now, let's be intelligent about this. Linux is not late. "Late" is not being released on time; Linux is in a perpetual state of re-release. Linux has more features now that were promised for 1.0. >Somebody said that and all the sheep kept echoing it! Tell me, can you >load the serial driver at boot time? the SCSI driver? the sound >driver? the HD driver? the Mitsumi driver? Er, no. Why would you want to? I rather like booting from my hard drive. And if you're going to be using the drivers, why don't you just compile them in? >some sheep would say you can enable that by recompiling the kernel. >That's loadable alright! Heh. Well, you do understand that you must compile the kernel at some point. While NT seems to spring whole from the Microsoft factories, it indeed has been compiled by little gnomes deep beneath the forest. Trust me. >The current loadable modules project doesn't have the blessings of the >linux developers, it'll never be integrated into the kernel. Larix >tried and failed. Uh, it was created -- by definition -- by a Linux developer. It's being used by the QIC-80 developers. And when there is a great enough need for it, and only then, it will be integrated into the standard kernel. That need has not been demonstrated yet. If you're always going to be loading the same set of drivers upon bootup, loadable device drivers doesn't help the user at all. >Not a single applicaion running under linux uses drag and drop, unless >of course you count the demo included with Motif! To be honest, I'm not very fond of "drag and drop." This is another magic word like "microkernel." I'm sure you can quote me some numbers showing how "drag and drop" has increased productivity? >Sun *binary* interface. It runs (supposedly) ordinary commercial >Windows apps that were compiled under DOS/Widnows. Linux has no hope >in a million years to have something like this. >All that is under linux is an attempt to make applications written for >Windows use an alternate API library by compiling them under linux and Incorrect. There are in fact 2 projects underway. The one included with SLS is as you describe, minus the incoherent ranting. The other is called Wine, and it is a Windows binary loader and emulation package. It runs a couple of simple Windows programs. They're hard at work now writing it so I won't offer it as an example of working code. But it has promise. >Well, it's getting too long. More later if find the time. Oh, please. Just how many times can you leave yourself logged in and walk away? You must have a very malicious roommate. -- Robert Sanders Georgia Institute of Technology, Atlanta Georgia, 30332 uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt8134b Internet: gt8134b@prism.gatech.edu ------------------------------ Crossposted-To: comp.os.ms-windows.advocacy From: blee@tim (Bryson Lee) Subject: Re: NT versus Linux Reply-To: blee@tim.dfrf.nasa.gov Date: Sat, 21 Aug 1993 06:10:48 GMT Brian Leary (brileary@world.std.com) wrote: : No so. I will say upfront that linux has the best chance among all : unices of surviving (mostly as a toy os). All the commercial ones will : be extinct in a mmater of a few years, due of course to the new unix: : NT. DEC and Silicon Graphics started the trend, and others will follow ^^^^^^^^^^^^^^^^ Just an interesting note--happened to hear from our SGI sysadmin that the boys from Newport Beach are _not_ intending to support NT on their boxes. Thought I'd add one more grenade to this firefight. -- ============================================================================== Bryson Lee, 1Lt, USAF Mail: blee@tim.dfrf.nasa.gov Flight Controls Engineer Voice: (805) 258-3123 ANY OPINIONS EXPRESSED HEREIN ARE MY OWN, AND DO NOT REFLECT THE POLICIES OR POSITIONS OF THE UNITED STATES GOVERNMENT, THE DEPARTMENT OF DEFENSE, OR THE UNITED STATES AIR FORCE. ============================================================================== ------------------------------ From: arcardw@indsvax1.indstate.edu (Paul Cardwell) Subject: NT vs Linux ? why? Reply-To: arcardw@indsvax1.indstate.edu Date: Sat, 21 Aug 1993 05:55:31 GMT Why would anyone want NT? Someone that loves multiple processors for parallel computing -- probably. I use OS/2 2.1 (in it right now :) ), Linux, AmigaDOS 3.0, and NeXT Step. 1. Yes, there is a FileManager for Linux and X-Windows in general. I am also creating a FileManager (similar to DiskMaster on Amiga). 2. Drag and drop is cool! I use it a lot on the Amiga and OS/2. 3. NT has a place in computing -- but it's not meant for the home user! Unless Microsoft makes an NT-Lite? 4. For mission critical Applications I rely on OS/2 2.1...nuff said. 5. Clients needing a cost-effective unix environment I suggest Linux or other 'free' unixes. 6. For Graphics,Sound,Animation (...multimedia) I use Amiga 4000/040. 7. NT received a lot of hype, but needs to prove itself. 8. I'd like to see NeXT Step come down in price -- it's a good OS -- well for Application development! Paul .___________________________________________________________________________. | Arc Wave, real life: Paul Cardwell | INet Address: arc@judy.indstate.edu | |----------\\-----//--------------------------------------------------------| | ---- \\\^ ^/// "Amiga, the wild side of power computing." ---- | | ----- \\|.|// Amiga,OS/2,Unix,C Programmer ----- | `------------\\-//----------------------------------------------------------' ------------------------------ From: gmt@beach.cis.ufl.edu (Gary McTaggart) Subject: Re: Modem XTERM emulation? Date: 21 Aug 93 02:32:09 > >> Anyway, my quest is to find a modem terminal emulation that would do xterm > >> when dialing into a unix box. > Since Seyon uses XTerm basically as a vt100 terminal emulator, couldn't > xterm be modified in this case to send vt100 escape codes to move the > cursor and stuff like that? It seems like it would be fairly easy to at Isn't it just simpler to run term/txconn and have a genuine xterm from the other side on your screen? No, because even with a 14.4 modem, the thoughput that term gives you is fine, but the interativity really lacks. (Yes, I tried lowering the packet size, etc., but I wasn't able to get an acceptable setting.) --Gary ------------------------------ Crossposted-To: comp.os.ms-windows.advocacy From: sheldon@iastate.edu (Steve Sheldon) Subject: Re: NT versus Linux Date: Sat, 21 Aug 1993 06:12:03 GMT In <930821002107.23308@world> brileary@world.std.com (Brian Leary) writes: I still want to know why I can't email you. >Ok, so you concur that a strong SHEEP MENTALITY runs among linux >users. Good. That's obvious anyway, nobody can deny it. The way I see >it, linux users can be classified as follows: You are the most biased little twit I've seen in a long time, but we'll listen to your unbiased comments some more... >0.1%: A handful of people that do the actual linux development and Make that maybe 1%. Depends on who all you include in your list. >10%: Regular users. Those also do some occasional reading of the Hmm, this is probably closer to 60%, but ok... >1%: Blood suckers. Those use the hard work of the first group to make Actually this is the .1%. >88.9%: THE SHEEP. Those rarely use linux beyonfd booting it. They do >all their work in DOS/Windows, but love to brag about how inadequate >it is, and how linux works for them. They know mostly nothing about >linux and whether it wroks or not beyond echoing what the master sheep >say, but are HEFs (Highly Effecient Flamers), and they fill the linux >group with posts that are mostly noise. Of course HEF + SHEEP >MENTALITY = LYNCH MOB! Well I'm just glad I don't fit in this group. But I think you're very mistaken. I've only seen maybe 10 people actually reply to your messages. I'm just doing it cause it's fun, and I miss the amiga days. :) >So here is the breakdown: > 0.1% developers |-- fanatics > 10% users | > 1% blood suckers |-- LYNCH MOB | > 88.9% SHEEP | | >I think that's pretty much an accurate and objective breakdown. Well it's not, but we'll let you live in your fantasy world. >>Objective comparison, eh? When you're using linux-0.12 as the basis of >>your comparison??? >> >I explained the points mentioned in the checklist in that post, and >most linux users in the know agreed with them. I don't think you quite read those posts did you? >>Interesting. This disproves your "sheep mentality" claim. After all, why >>would any Linuxer bother reading comp.os.ms-windows.advocacy if they >>weren't a free thinker? >The article was reposted by a sheep ... err linuxer to the linux group >after it appeared here. Definately didn't read those posts. I thought I disproved the sheep theory by counter-example. Would anyone like to clarify? >>>Of course, there is also the FANATIC MENTALITY, but that one is >>>obvious. >> >>This seems to be true of you as well. >> >No so. I will say upfront that linux has the best chance among all >unices of surviving (mostly as a toy os). All the commercial ones will >be extinct in a mmater of a few years, due of course to the new unix: >NT. DEC and Silicon Graphics started the trend, and others will follow >soon. In a few years, unix will be mentioned only in history books, >and X11 will be seen only working as a subsystem under NT. I seriously doubt that this will be the case. Windows NT doesn't presently stand much of a chance in this market. But go on... >The other free unices? Well, HURD has already stagnated, too late for >it now, and the BSD ones SELF-COMBUSTED. Yes, now this seems true. It's interesting how BSD fought with themselves to like they did. >Now you see how reasonable a guy I am? Well, you've got a couple good points. You just don't quite know how to put them all together in a coherent argument. >>What do you expect? Windoze NT is *already* behind schedule. It was >>supposed to be released several months ago. We can't be responsible for >>the bogus claims of others. >> >What a joke! Linux 1.0 was supposed to be released last summer. In >December, SLS 1.0 was announced, under the impression that linux 1.0 >would be released by then. Well, we're still waiting. NT was late? >Hahaha! I don't think that's a fair comparison. Linux is waiting for 1.0 to work out all the bugs. Microsoft is just releasing it hoping the users will report some bugs. >Somebody said that and all the sheep kept echoing it! Tell me, can you >load the serial driver at boot time? the SCSI driver? the sound >driver? the HD driver? the Mitsumi driver? *ANY* driver? Of course >some sheep would say you can enable that by recompiling the kernel. >That's loadable alright! Heh. I'm not currently aware of where loadable drivers are being used. It's a new development, but it does appear to exist. I don't quite think you grasp the situation yet. I personally don't think that loadable drivers is that great of a thing. None of the other Unices I work with use them. We just relink the kernel, which isn't that hard of a task. >The current loadable modules project doesn't have the blessings of the >linux developers, it'll never be integrated into the kernel. Larix >tried and failed. I don't think this really matters, and I don't understand why this was brought up as a benchmark here. >Not a single applicaion running under linux uses drag and drop, unless >of course you count the demo included with Motif! Now this is patently false. I have an application called Ftptool which was written using the Xview libraries by Mike Sullivan at Sun, and it uses drag and drop. Again I don't see this as being terribly useful, and don't understand why it's being used in comparison. >WABI is the BIGGEST LIE PERPETRATED ON LINUX USERS!, mostly as a >gimmick to push the declining sales of SLS, which were hurt by clones. What does WABI have to do with SLS? I'm kind of curious about that. And I don't quite understand what declining SLS sales has to do with anything. I don't believe that anybody is developing anything specifically for SLS. SLS is just on distribution. >Of course someone said WABI! and all the sheep started repeating WABI! >WABI!! WAAABI!!, and soon they believed it. First of all, WABI is a >Sun *binary* interface. It runs (supposedly) ordinary commercial >Windows apps that were compiled under DOS/Widnows. Linux has no hope >in a million years to have something like this. Why if Sun can do it on a Sparcstation, cannot Linux do this on a 386? Just curious, you seem to be blowing some steam out of your ears. >Well, it's getting too long. More later if find the time. Can't wait. You really do seem to have a conceptualization problem, guy. -- sheldon@iastate.edu Steve Sheldon #insert "standard disclaimer" Iowa State University ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Misc-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.misc) via: Internet: Linux-Misc@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Misc Digest ******************************