THE VIEW FROM CHAOS MANOR
December 14 - 20, 1998
This is a day book. It's not all that well edited. I try to keep this up daily, but sometimes I can't. I'll keep trying. See also the monthly COMPUTING AT CHAOS MANOR column, 4,000 words.
|Previous Weeks of The View:||For an index
of previous pages of view, see VIEWDEX.
See also the New Order page, which tries to make order of chaos. These will be useful.
For the rest, see What is this place? for some details on where you have got to.
If you subscribed:
If you didn't and haven't, why not?
For the BYTE story, click here.
The LINUX pages are organized as the log, my queries, and your responses and advice parts one, two, three, and four. There's four pages because I try to keep download times well under a minute. There are new updates to four.
Highlights this week:
December 14, 1998
Chores done, page updates done. Calvin Dodge has sent Red Hat 5.2, and shortly after that the Official Red Hat care package arrived, along with "Power Tools and other stuff. Now I need to figure how to upgrade what I have, and do I install Moshe Bar's suites before or after? This will take some thought, and I haven't time for thinking. The good news is that the Simon and Schuster contracts are ready so I won't starve, and USAA is clearing up all the detritus from the Bronco wreck.And I did a couple of pages on Mamelukes last night as well as paid the bills. It may be a good week. It has started well, anyway.
Interesting observation by Talin on what this place is.
I have about decided to start over on Linux: scrub the disk, repartition with few to no extra partitions (essentialy a swap, and a couple of big ones, and that's it) and install 5.2 from scratch. I am astonished that none of the other Linux suppliers have sent anything; most of them said they would at COMDEX. Of course one can download a bunch of stuff, and one day I will, but Red Hat seems to be doing a good job of keeping up with things, and they have documentation, and such like. Before I scrub I need to read what Moshe Bar has to say about the work he did on getting SAMBA to work with my system (I have not installed that) and whether it will work all right with 5.2 installed from the beginning. But the more I hear about Linux the more I like the idea of a big Linux box to play with.
On that score, Corel is about to send me their already set up Linux Box; more when I have it, but it won't be too long now. And that is probably the best way for small businesses that want to use Linux to do it: installation is not a problem and the boxes are cheap enough, and Corel is enthusiastic. And you will get Word Perfect Suite as part of the deal...
VIEWDEX, the index for VIEW, is now split in half thanks to John Rice: Viewdex1 has the first group, and Viewdex is now the current index. There is also an updated New Order. These are valuable tools for finding things. When Darnell gets the updated Front Page Extensions going we will also add a search page. I think I will then add a hidden directory which you can get to only by registering, and another you can get to with a password I'll supply by mail to subscribers. This is more in the nature of playing with the Front Page tools and anything serious, but perhaps I'll keep the current daybook page where you'll have to register to see it, and the current column in the subscriber area. Clearly all this is easily hackable, just as most locks are to keep honest people out of places they have no business in.
December 15, 1998
All of which makes for a full schedule.
I have sort of given up on getting a car until I have some more money, which ought not be all that long. Finish some work. Of the ones I looked at the GMC Jimmy and the Dodge Durango were acceptable; the Durango is almost $10,000 (out the door) more than the Jimmy, the differences being a big V8 and a lot of interior options. The Jimmy is the closest thing I have found to my old Bronco II. It's larger but not that much larger, and while the darned thing costs too much, it does seem to represent a minimum 4 wheel system. The V6 has enough power that I was able to go up Beverly Glen with a passenger and the A/C on and still have some acceleration; which is about what the Bronco II could do. The Durango on the same test had lots of power in reserve, and wasn't at as many RPM; the Durango will clearly cruise at 75 all day without any strain at all, and my excursion into an unpaved street convinces me that the off road capability is there. But it's quite costly. Time to work. When I have the money in hand I can think more clearly. (One attraction of the Jimmy is 0.9% financing, which is like they are giving you the financing for free even with our minimal inflation.)
Now to work.
Well that didn't work: Roberta had shopping to do in Beverly Hills and I ended up going with her so we could have lunch together.
Meanwhile, a report by Russell Kay on a problem with Internet Explorer 5 beta.
And purp, a useful UNIX command, for me to learn about when I get to the Linux box.
Tonight's BUFFY was wonderful. I am a big fan of Buffy the Vampire Slayer. One of the only TV shows that actually takes good and evil seriously.
December 16, 1998
Well, we are wagging the dog; thoroughly predictable, of course. One way to stop an impeachment vote. God save the troops involved, and the innocents in Baghdad.
The day was devoured, this time by a thick NASA document, draft of the history of the DC/X. I have no choice but to review it, and alas, they got some right and some wrong. Won't pay a dime to correct it, but it has to be done. Tomorrow Roberta is off to San Diego leaving me here. I WILL get some work done. I have to...
We did a lot of work on Roberta's machine last night. It will all be in the column; but for the moment the big problem is that we can't get internal modems to believe there is a dial tone. We think it is because Plug N Play assigned IRQ 12, and that has confused the modem software. What I have to do is reserve IRQ 3 for an ISA modem and set that modem to the proper settings. At the moment IRQ 3 has been assigned to the Ensonic; IRQ 5 has been assigned to the Ethernet; IRQ 12 has been given to the internal modem, and that ain't working. We need to rethink all this.
We have the first ever Chaos Manor USB device on her system: a USB Microsoft Wheel Mouse. Works fine EXCEPT that you cannot use the mouse in SAFE MODE! So you had better understand keyboard controls for Windows if you use a USB mouse! And, in fact, it turns out NOT to work fine: that is, when it was working it was smooth, but we had ODD problems and lockups with mouse intensive software, all of which went away when I changed to PS/2 mouse.
I suppose the thing to do here is delete a bunch of devices, take them physically out
of the system, reserve all the IRQ's except the one I want the next device to be, boot up
and install that device, lock it down -- Plug and Play always wants to run
reconfiguration, every darned time you put something in, and it reconfigures as it likes,
not as you had it or want it. I want IRQ 5 for sound, and 3 for the internal modem, and
the Ethernet can have 12. Incidentally, in the iWill board, if you disable the PS/2 mouse,
it really does free up the IRQ for anything you like.
December 17, 1998
Roberta is off to San Diego and the beach house and a singing gig, and I am trying to get her machine in shape. There is now an urgency: her Lexmark printer, which she loves dearly, is not working because following a paper jam, we cannot close it back up; the cartridge isn't seating properly. We think this happened once before and when we took it to be fixed, they found nothing wrong and it came back all closed up properly. If so, there is nothing in their manuals or on the web on what we are doing wrong. It must be something simple, but darned if I can figure it.
So we have an older Mannesman Talley 906 that will work with the new machine, and is hooked up to it, and we'll put that in place of her Lexmark while we send the Lexmark out to be closed properly. This seems silly, but it is a spur to get her new Big machine set up properly. Scarlet ("I'll never be hungry for disk space again.") is an iWill motherboard, AMD K6-2 3dNOW system running at 66/300 (to be conservative; Roberta has no need for the extra speed crom 100/300) and the first from the beginning Windows 98 system in the house. We spent the evening downloading upgrades to software and drivers and Windows 98 (service pack 1 is not out alas, but there are some upgrades at the site). We have her on an external Modem off COM1: and that works, although it's only a 28.8 modem, USR V.EVERYTHING, and can't be upgraded, at least not without a hardware swap. I used to get new USR modems as they came out, but 3COM seems to change PR firms every 6 weeks whether they need to or not, and I have lost contact. I still like genuine USR external modems: they work.
Anyway, her machine is working fine, and is quite stable. Now all I have to do is fight plug and play to let the IRQ's be in the right places; see yesterday. We can't get an internal modem to work on IRQ 12, and I need to lock it to IRQ 3, which, I think, requires a lot of tedious if fairly simple work.
BOY did I speak too soon on the stability of her machine. Odd lockups. Weirdnesses. But I think I know why.
We continue to bomb Iraq and for the life of me I don't see what we accomplish. His nuclear facilities are gone, and won't be back. Biological and chemical weapons are so easily made, and the facilities so easily hidden, that it's unlikely we'll get all or even many of the chemical, and as to biological, any private house, high school laboratory, old wine factory, or most anything else can conceal facilities for making such weapons. Frightening, but there it is. Which means we can kill Iraqi's all month, but we won't get Saddam Hussein, nor will we weaken his control, and if we do kill him he's replaceable; the Ba'ath Party can find another. Not perhaps as popular as he, but enough so.
For my part I thought the Gulf War unconstitutional without a Declaration of War, and this escapade even more so; and if we want to change Iraq we have no choice but to send in ground troops and do it. Exactly why we would want to do that, when Iraq is now notably weaker than, say, Turkey or Iran or Israel, is not clear to me: there are plenty of bandit regimes in the world. We have troops carrying out the garbage -- literally -- in Haiti, and we seem to have a permanent garrison in Bosnia. We're about to send someone -- God knows who will volunteer -- unarmed to Kosovo. The CIA is to play some kind of game monitoring Palestinian/Israeli peacekeeping, and God alone knows how that is going to work -- what can we do that Shin Bet doesn't do better? But this is the course of Empire, and we seem determined to embark on it. It's exciting, and will provide plenty of high presitge employment for former Senators who become proconsuls, but it has little to do with the Philadelphia Constitution of 1787. Somehow I seem to be the Constitution's only mourner. Enough on that. Computers are easier to understand, and even Plug and Play is less frustrating.
Regarding that: is there any way to nail down specific IRQ than to take all the boards out, reserve all but the one you want a board to be, install the board, nail it down there, reserve all but one again, repeat as needed? There should be a simpler way, but I don't know it.
One is suggested in Mail; I'll try it. Report when I know.
December 18, 1998
Of course you can assign an IRQ to a specific PCI slot. I have given 5 to the sound card. That works. Now to see if I can SCSI as well as IDE. We have found that if the EZ SCSI drivers get confused as to which IRQ the SCSI card is on, the system can blow sky high, rebooting after the first Windows splash screen. Removing the SCSI board and uninstalling all SCSI drivers -- not just in System but Uninstall the whole darned thing -- seems to be the only remedy.
I have put more time into RJP's system that it was worth: I could and should have bought a new one for her. But it was an interesting experiment. We now have the network, and sound. SCSI is so she can have a CD burner.
There really must be a simpler way to do all this. LINUX isn't it: installation with LINUX is even worse. But as these machines get more powerful, they get more like CP/M, when you had to write and assemble your own BIOS. I shouldn't have to spend all day getting Plug and Play to do what it ought o be able to do easily enough.
Well that was goofy: it is possible to install a PCI board such that it appears to be seated and Microsoft will recognize that there is now harware: but it is not seated well enough that the BIOS actually sees it as a SCSI board. This can waste a lot of time as you try to install drivers.
Does anyone have experience with USB, and in particular a USB mouse? I am beginning to believe my real problem here is the USB mouse on this machine. I get goofiness in varous ways. I am going to disable USB, enable PS/2, and put a PS/2 mouse in this system; and see what happens. I don't care for what is going on here.
And in fact switching to a PS/2 mouse has ended the weirdness. I do not think USB mice are ready for prime time; Windows 98 does not seem to know how to support them. This can be serious. My diagnosis is pretty positive since we had all kinds of problems with this system, and now after changing to PS/2 and disabling USB we don't. I need some other USB devices to test. But for mice I cannot recommend USB with present drivers.
I would have thought CNET would know that the overseas BYTE magazines are licensees, are not owned by CMP, and have nothing to do with CMP; but apparently not. There seems to be an announcement of a new BYTE web magazine. So be it, but if CNET got that part wrong what else did they get wrong? See http://www.news.com/News/Item/0,4,30128,00.html for the story.
December 19, 1998
Well, they did it. What will happen now isn't clear. The whole scene in Washington reminds me eerily of the last stage of the Roman Republic. I hope I am wrong. Neither side came off well in that. I'm glad I don't have any part in this. I do note that as power grows greater, having it is a greater reward: and worse, fewer men and women of honor want that kind of great power, leaving it to those who do want it, and will do anything to get it. That is the point of limiting government: it makes it less of a prize to grab. A lesson Washington and Adams knew well.
I didn't hear the final debates because I got up late: I spent much of the night infuriating myself reading The Baltimore Case by Daniel J. Kevles, and didn't stop until it was grey daylight outside. This is the story of Congressman Dingell's exploitation of a scientific dispute that ended up with David Baltimore resigning as President of Rockefeller University, and shows precisely what happens when you take government money for science: the politicians win, the scientists lose, and the effect on science is interesting. Especially in direct funding systems like NIH (as opposed to NSF which can be something of a buffer in these matters), the politicians will exploit anything they can. Few cases are as raw as Dingell's decision to "get" Nobel Laureate Baltimore because Baltimore dared stand up to Dingell and his gang of thugs posing as Congressional staff; but there are and always will be many more.
The lessons here are not clear. Without churches and great families, in this era of death taxes, it is difficult to find mechanisms for looking far to the future or funding sources for long term research and basic research and the argument that only government can do this now is compelling; but The Baltimore Case shows this can be fraught with danger. In this case the conclusion was found almost in the first weeks after Margot O'Toole, a brilliant but dangerously obsessive and flawed post-doc who expoited her role as "whistle blower" raised the questions about the publication in the journal CELL that started the whole thing.
O'Toole was unable to duplicate some of the work done by Baltimore's co-author Thereza Imanishi-Kari. In frustration she came up with an alternate interpretation of the results Imanishi-Kari and Baltimore obtained. And in the first meeting with Baltimore, Baltimore said that he didn't accept O'Toole's views but they were a possible interpretation -- and "no one is going to settle this by arguing. This indicates a need for further research." That is where it should have ended, but it didn't end there; research in an area critical to understanding lupus was set back by years, careers were ruined, tens of thousands of man hours of brilliant scientific people were wasted, and Dingell's thugs ran roughshod through the scientific community establishing dangerous precedents. And at the end the conclusion is that the questions can only be settled by more research because no one is going to settle it by arguing.
Well, Roberta's machine is stable, but I had to turn off the power management in the BIOS; it was going to sleep, but would never wake. That is, the monitor would show the yellow light that indicates no signal, moving the mouse would do nothing to wake it, and while the machine was "on" nothing would recover it. The odd part is that earlier while I was having other problems, this didn't happen. It may be that I changed some BIOS setting in getting things set up, but I don't know. I do know that this machine has consumed more time than I like, but it HAS to be stable since it will be Roberta's. I'll make another attempt to see if there's a BIOS power management setting that will recover, but for the moment what's happening is that Windows 98 is putting the monitor to sleep, and that we can recover from. The power consumption of a PC is not large when the monitor tube power is off, so I don't worry much about power management anyway.
With that fix the machine seems very stable. I ran THIS MEANS WAR and TOTAL Annihilation (games), both of which put a LOT of objects in motion on the screen, and neither locks up; both did when I had the USB mouse installed. I am now certain that Windows 98 doesn't really understand USB mice, and USB is contraindicated; stay with PS/2 or serial. Both of those work.
Alex has taken away one of my machines, leaving me a work station bench free, so I will get the other Windows 98 experimental machine going, then turn my attention to Linux. If Moshe Bar is reading this, I have sent mail several times, but it is bounced; if this wasn't temporary, please send me a good mail address, and my thanks for the SAMBA suites. I'll get at all that Monday or so.
I find some amusement in the current stories of what happened to my ARPANET account many years ago. The real story has far more to do with politics than anything else. When my account was cancelled I was invited by friends in the Pentagon (where I still did some briefings for USAF Plans and Doctrines) to make a formal protest, but that would have been to return evil for good; Professor Moses (MACSYMA) and his executive assistant Ellen Golden went to a lot of trouble to set things up for me, and I sure would not have wanted to see her harmed directly or indirectly. Marvin Minsky offered to intervene as well, but again it seemed a lot of trouble to keep an account at MIT when I had one most people didn't know about through John McCarthy at SAIL. As far as I can see, one Cstacy wanted to show that he could kill my account, and I didn't have any real reason to pay attention.
I didn't see them at the time but now I notice there are stories of threats about suing McGraw Hill if I were to "libel" the people involved in BYTE. Man, they didn't want to DO that; McGraw Hill legal were aggressive tigers! All I ever had to do was prove truth; they would take it from there. But I think that threat came about through a casual remark I made at a science fiction convention, in which I said something to the effect that they could tell Cstacy that he had managed to annoy me and I'd have a lot more to say in public. I promptly forgot about it, since I had many other means of communication if I wanted them.
The theoretical reason was that I was supposed to be giving away ARPA secrets by mentioning something or another in BYTE; this about an unclassified tax funded DOD Defense Communications Agency test system, in which the very people who were complaining about me were having "at sign" parties ("if you have an @ in your name you're invited") and generally seeing that this was the worst kept 'secret' in the world. Eric says these people were the Net equivalent of Welfare Queens who never paid for accounts, and were terrified that somehow if it got to the public that a defense agency was allowing all these students to use their assets, the game would be over.
But in any case my public discussion of public supported facilities was the ostensible reason I was dropped. One supposes that my views on defense -- DOD was paying for the net -- had a lot more. Here we had people who loathed the Defense Department working on a DCA project; they had to do things to relieve their consciences, and show they weren't "really" helping defend the United States, but just taking the nation's money. Or something like that. I never quite understood that position. Some of them were using their university connections to avoid conscription, and again seemed determined to show they didn't REALLY support defense.
My views on the Viet Nam War were published at the time, and have not changed. It was undelcared, and I did not like that; but there was the color of the Tonkin Gulf Resolution. I thought that illegal, but the courts held otherwise.
I thought the Viet Nam war horrible--Sherman was right, War is Hell-- but necessary: it was a campaign of attrition in the Cold War (the Seventy Years War), and while wars of attrition are slow, costly, frustrating, and often tragic, they can be decisive; this one was, as enormous percentages of the USSR's truck and transport industry worked to send their products to be destroyed in South East Asia and thus were not available for building USSR infrastructure. This was decisive.
But it was a horrible war, and as Allard Lowenstein said one night in a debate, it looked as if the goal was not to win and get out but lose and stay in. (One of the few times I have been floored in a debate, and this on national TV. Allard said "Jerry, you want to win and get out. I just want to get out. Your friends McGeoge Bundy and Robert McNamara want to lose it and stay in.")
But there were many ways to oppose the war. My friend Phil Lemons, my first tech editor at BYTE (later Editor in Chief at BYTE and now PC WORLD) was opposed to the war in Viet Nam but his protest took a costly (to him) form. It's his story, not mine to tell, but it does him honor. And Allard Lowenstein, the man who brought Lyndon Johnson down and one of the few people to defeat me in a national debate, and I understood and respected each other's positions. But that was at a level a bit above the petty gossip that was the early ARPANET.
The only significant thing I ever did on ARPA NET was to get GPS funded. Of course that is too strong a statement; actually it was a lot more complicated than that.
What happened was that Francis X. Kane, the silent co-author of Possony and Pournelle THE STRATEGY OF TECHNOLOGY, had invented Global Positioning System while Director of Plans at USAF Systems Command, and had managed to get USAF to request and get funding. Then in the 70's Congress zeroed GPS out of the budget. Kane called me in hopes I had some resources to get it revived. I broadcast a message about that on ARPANET, and Lowell Wood, Livermore physicist and one of Teller's inner circle, saw it, sent me email to get details, got in touch with Kane, and went to Washington where he lobbied among his Congressional contacts to get it restored. It was restored and I think it would not have been done without Wood, who would not have heard about it were it not for ARPANET. GPS turns out to be important, but then we all knew it would be; the trick was convincing Congress.
That done, I had books to write and work to do. Commercial net accounts were coming, Loudon was inventing what became CompuServe after several evolutions and transformations and acquisitions, and BYTE was inventing BIX; I still held a clearance through having been President of Pepperdine Research and I still did some policy briefings for Air Force Plans and Doctrines, and so had all the communications I needed. Stallman, one of my friends from the MIT ARPANET, was done with most of his work there (I was an early beta tester of emacs, and some of my na´ve observations found their way into the final product), and I found I spent most of my time on ARPANET playing ZORK, an activity for which I had no time; it was a gift of time to find that account closed. I resented the way it was done, but not enough to speak to anyone in the Pentagon about it. I see that the story still reverberates after twenty years (it all happened in CP/M S-100 days, mostly at 300 baud; 1200 was a godsend). Peace.
Saturday night: Eric and I went out to Fry's where I decided to solve Roberta's Modem Problem once and for all: I plunked down a hundred bucks for a Creative Labs external V.90 56K Modem Blaster. I could probably get one of those from Creative, but I want to get her system set up and tested and DONE, and this ought to do it. And if I do have any problems I can get hold of people at Creative. I gather their tech support can be slow but it is sure; and in my case I can by-pass much of the slowness. One advantage of my position. I'll need to know about this modem for the O'Reilly book anyway.
For the record I have not yet managed to get any AGP board to work with a super-7 AMD K6/2 chip, and I have tried three mother boards. I'll keep trying, but it's a little discouraging. As I said I got one to work until I installed sound card and then it blew up... I will keep trying.
And it's about time to build a new Pentium II system. May as well see just how much advantage you get from Pentium II vs super 7 vs Celeron A. Or that I get, anyway.
December 20, 1998
What my military friends call Operation Free Willy ( a few prefer Impeachment Storm) is over, with Saddam victorious. He's firmly in power, the arab coalition against him is gone, the typical arab man in the street in Syria, Jordan, and elsewhere sees Saddam as a courageous hero standing up to the Great Satan; we are a billion dollars poorer; there will never again be UNSCOM inspectors on the ground in Iraq; Syria, Iran, and Jordan will openly defy the UN sanctions, opening up a flow of money to Iraq; and impeachment was delayed by one day. Oh, and in Russia the moderates who advocated cooperation with the US were pretty thoroughly discredited with respect to the hard liners who said the US would never really consult with them before taking action; and impeachment was delayed by one day. Saddam is stronger than ever, and impeachment was delayed by one day. We have sown the wind.
Mail is getting enormous, but I won't break it up since today is the last day. It is clear I have been doing too much of this stuff, though, when the downloads get this long. Today I'll knock it off: I have an Intellectual Capital column to write, and this place is a mess, and Roberta's machine is stable enough now that USB is gone so I can get that downstairs and transfer all her files to it. Sufficient unto the day..
I need some help. Does anyone know Eudora and PGP well enough to help me move them from Roberta's old machine to her new one? I am sure there is an easy easy way but I do not know what it is. She has a credit cqard verification system and PGP on her old machine; those need Eudora. I have the original disks for Eudora Pro 2.1 which is what she was using so I can install that; how do I get it to find her old mail, and such like? And transfer the PGP stuff? Moving from one machine to another is a pain; the problem is that "Joizy", her Gateway 2000 Pentium 200, uses rather odd memory so memory upgrades are not easy; otherwise I would just put a second hard disk drive in her machine, use Partition Magic to make her original 4 gig drive one bug C:, and make the new 4 gig a D drive. I may yet do that. I have drives...
I have her systems linked by 100 Base T Ethernet, and that works fine for file transfers. That is for some file transfers. See below.
Any advice appreciated.
And of course I got some. Thanks to Moshe Bar on NetBeui, and Donald MacArthur on transferring PGP.
NetBuei sucks rocks. Roberta has made some sound files. They have VERY long file names. Attempts to transfer those files by Ethernet Microsoft Network NetBuei are painful. About half the time the machine locks up. Not always, but enough. Can drive me nuts... One file at a time is OK. Often multiples. But sometimes with multiples it jasn't recovered from the last before it blows up; that's an hypothesis. All I know is that it locks up sometimes. Then I have to reset. Then start over. Feh.
The other night I let Alex contact the Microsoft site with Scarlet, the new machine. He was looking for an upgrade to Windows 98, hoping for a service pack. There wasn't one, but there were a bunch of upgrades that Microsoft says you can download and install automatically. There does not seem to be a way to download and install yourself.
My advice is DO NOT DO IT. We did, and that was the beginning of the problems that ended with my disabling USB.
Worse, now SCANDISK is an awful program to use with a BIG disk like 8 gigabytes. It takes forever, it shows no real progress indicator, you have no idea of what it is doing, and it sits there going click/click on the disk driver for HALF AN HOUR. It was showing cross linked files in Windows/Java. I have never installed Java on this machine. No reason to. I suspect MICROSOFT installed its version of JAVA when Alex tried to upgrade Windows 98. The result was a ghastly mess. I think I will have to delete all that and be done with it. Microsoft has managed to become both arrogant and incompetent with this stuff: my advice is DO NOT go to the Microsoft site and try to upgrade Windows 98. The result can be horrible. I am not sure what it has done to my system, because there isn't really a log of what it did. But now Scandisk is running, I am hearing click/click/click on the disk drive, there is no real progress, and no tool that will let me find out what the devil is happening. And after half an hour one does wonder.
Stay away from those automatic installation updates from Microsoft. For me they did no good and have cause endless grief. Your milage may vary.
NOW I need more help: does anyone know of a decent WIndows 98 disk tool? I have Norton for Windows 95, but I haven't got it for 98. Is there a downloadable upgrade? What does exist?
File transfer problems: Apparently Windows cannot stand a combination of long file names and a long list of files to transfer. It will send those files across the Ethernet a few files at a time, but if you select a whole bunch of files some with long file names, it will lock up along the way. This is interesting, but it makes for tedious file operations. Now we know.