CompuServe Thread

#Hard errors?

32 messages in this thread
#109851From: Don Curtis/SYSOPFeb 19, 1988 12:15 AM
Jim, Why do you blame Commodore for an error when you pull a disk during a write? What would you have them do…lock the eject button? The user has to take responsibility for some errors. CBM has warnings printed in prominent places several times in the manuals about pulling disks too soon. As Steve has pointed out, keeping the system occupied while doing disk access is against the idea of multi-tasking. The disk access is simply another process that runs with all the other process at any one time. Pulling the disk from almost ANY computer while the disk is being written to is fatal to that disk…that's a fact of physics, if you don't complete the write…you don't have complete data on the disk. Don
#109939From: Jim VentolaFeb 19, 1988 7:56 PM
Don– *I* did not pull the disk out–a beginner did. So much for user friendly, eh? Why do you think the Mac is a success? I don't want to keep the system occupied while disk access goes on, but since there is no reason ever to eject a disk being written to, then why should disks be ejectable under those conditions. Hell, you sound like the guys in Detroit speaking to Nader. Why do we need seatbelts and safe cars. Drivers have to take some responsibility, etc. etc. Of course they do, but failsafe systems are preferable, especially in a business environment where more than one person uses a machine or even the same disk. (Ask a dealer how many new users have made this mistake.) // Jim
#109966From: Don Curtis/SYSOPFeb 19, 1988 10:55 PM
Jim, Failsafe is expensive and in the case of a computer, is not a human safety feature…are you willing to pay for the expense of a disk lock mechanism? That still doesn't prevent you from turning the machine off, or do you want an interlock on the power switch? That doesn't prevent you from pulling the power cord…or do you want a mini-UPS built in to provide enough power under any circumstances to finish a disk write? You see what I'm getting at? How safe is failsafe and how much are you willing to pay to get it? The Mac has some major problems due to the fact that you have to tell it to eject the disk. I don't use a Mac, but have heard several complaints from folks who do about have one. Using a Mac type system, where the computer ejects the disks…you can't by a physical button…you'd have disks poping out all the time, remember again…this is a multi-tasking system and if the system wants a disk that's not in the drive…it has to pop out the disk that is in the drive at the same time it popped up the "insert volumn…" requester. The disk light is the warning…when the light is on…don't pull the disk. The light is on the drive, that should be warning enough to PROPERLY trained personnel. If a business is willing to turn users loose on a computer with 0 training…then the business should expect some problems and the fault lies with the management…not the computer. BTW… 'the business computer' is the IBM PC (whether we like it or not) and it has no such failsafe system and businesses don't bitch about it, they train their users. Don
#110003From: Cumberland Sci. MuseumFeb 20, 1988 2:52 AM
I would like to expand on what Don has said. As the computer educator here at the museum, it is also my responsibility for training new employees on the IBM systems. I have a small four hour program which goes over the basics about taking care of disks, don't touch the disk drive while the red light is on, etc. A little time spent in training goes a long way in keeping me from having to go to the offices to 'fix' things. Since we implemented the training sessions, our users make fewer 'dumb' mistakes like taking the disk out when the light on the drive is on, and being sure to backup the hard disk every day. It only took one of our users accidently deleting a file and not backing up the hard disk each day to drive that point home. Richard Shores Cumberland Science Museum
#110004From: Don Curtis/SYSOPFeb 20, 1988 2:56 AM
Richard, You sent the message to me rather than to Jim (which is where I think you intended it to go), but I think he'll see it anyway. In any event, as you know from my message…I agree, if you are going to turn a user loose with a computer…you should teach them how to use it before doing any serious work…or giving them the capability of harming anything. Don
#110036From: Jim VentolaFeb 20, 1988 12:42 PM
Don– My college has spent about a million and a half on computers recently. We were offering a ridiculous $4.50 an hour for "lab monitors" and were naturally having trouble keeping any. We have ONE technical user support person. This is, as you say, a very BAD situation. We try to tell the Deans and the President how badly we neeed support personnell–trainers, esp. Guess what happens in budget setting? The real world is not as it should be. Finding a balance between a closed, almost patronizing, system like the MAC and a hacker's heaven like the II series and the Amiga is not easy. Still, I think not letting a disk get ejected while it is being written to makes some sense if it would cost less than $20. to implement. I wonder how much it would cost? // Jim
#110106From: Claude J. GoschFeb 20, 1988 9:44 PM
Jim, Try this on an IBM and you'll see what a really non-failsafe system will do for you. Load up Lotus 123 and load in a spreadsheet. Do a print to disk. The WAIT indicator in the upper right hand corner will flash a while, then stop, returning you to the menu. The disk light will be OUT. Now insert another disk with a different spreadsheet, leave the print menu, and attempt to load in a spreadsheet. You'll see what happens when the directory of the first disk is written to the second disk! SERIOUS PROBLEMS! I know someone at work who is fairly computer literate but got trapped by this. With the WAIT indicator gone and the disk light out, he assumed the disk write was complete. It wasn't – in fact it isn't complete until you EXIT THE PRINT MENU!!! How's that for "user hostile"? Claude
#110140From: Jim VentolaFeb 21, 1988 12:46 AM
Is that what they call in the trade a "gotcha"? One has to wonder if the programmers who made it did it on purpose or just never had the imagination for testing their system. // Jim
#110196From: John DraperFeb 21, 1988 4:09 PM
Jim, I'll tell you one of the costs that you may not be thinking of… the cost of supreme irritation of anyone beyond the "What's a disk" stage. I always thought the Mac's insistence on knowing best when to eject the disk was not only typical Apple arrogance, but just plain stupidity. Do you know any Mac owners that don't have a handy paper clip to eject the disk in the cases where the machine doesn't? The user must take _some_ responsibility. I have little patience for those that overestimate their own ability to the degree that they feel they don't need to read the docs. I have less patience with those that make the same mistake over and over again. Regards, Larry.
#110021From: Richard Rae/SYSOPFeb 20, 1988 8:41 AM
I agree with you completely. I also agree with Jim in one respect: the programmers who wrote the icon-driven programs for workbench should have made the assumption they were working with newbies and written their programs accordingly. Throwing up a message which says "Diskcopy completed" when it isn't is DANGEROUS with new users about, head load lights notwithstanding. It would have been simple enough to hold the prompt on that one program until the physical disk activity had been completed; at the very least they could have been crude and put in a fixed five second delay or whatever. I remember back when the Ami first hit the streets… MOST of us were complaining about what Jim is now speaking of. We rationalized it by saying "Watch the light, not the screen", but at that time most of us felt that the early prompt was a bad move. Maybe we've gotten complacent now that we know our way around. I know I have. But I also still recognize a bad move when I see on. Rick
#110121From: Marlene Zenker/SYSOPFeb 20, 1988 11:43 PM
Rick, as I recall the thing we complained about the most on the first Amiga's was not about pulling out a disk when the drive light was on but the fact that sometimes after the light went out it would come back on again! Now that's something to bitch about. Marlene
#110144From: Don Curtis/SYSOPFeb 21, 1988 12:51 AM
Marlene, That WAS something to bitch about…but since CBM fixed that in 1.2, (maybe even back in 1.1)…it's no longer a problem. The light stays on between both writes. Don
#110146From: Marlene Zenker/SYSOPFeb 21, 1988 2:23 AM
Don, I know – just didn't put it in the proper tense! Marlene
#110161From: Richard Rae/SYSOPFeb 21, 1988 10:49 AM
Granted; I did not say that we complained the MOST about the prompt, I said MOST of us complained about the prompt: a subtle, but significant difference. I won't try to say what one item was complained of the most, but I'll wager your light going off then coming back on was one of the top two or three; it WAS a significant problem and you are right: we complained a LOT. Many of us still maintain(ed) that if the prompt did not appear until the disk was safe to remove from the drive, most newbies would not have had the problems they did with munged disks. Even the head load light going off then coming back on a few seconds later wouldn't have burned the folks who are used to watching the screen instead of the drive. Rick
#110199From: John DraperFeb 21, 1988 4:44 PM
Rick, While it is a given that the WB tends to be used by the 'newbies', it should not be crippled in order to attempt to make it foolproof. That it is crippled in other major ways is bad enough without leaving that *&^%&^%$ sleeping cloud on longer than necessary. I say let the newbies learn how to operate the system properly, rather than make everyone act at the 'newbie' level like the Mac seems to do. Regards, Larry.
#110035From: Jim VentolaFeb 20, 1988 12:37 PM
Don– Your position is rational, it makes sense, and so I can agree with it. But it also remains true that training is precisely the most expensice part of owning computers and that the miserable Mac made unexpected inroads precisely because it helped reduce costs in that area. For myself, I would certainly NOT want to pay to protect from: shutting the machine or pulling the plug or maybe even power loss during a disk access. I WOULD like to be able to purchase that protection from power interrupts if deemed necessary. Minimally, I would like a delay loop inserted into DISKCOPY so the SCREEN did not contradict the light. People are not perfect-escpecially not work-study students who are in a hurry. Still, it is true that people should be trained, and I suppose that after trashing the first disk one is VERY likely to watch that little light. Indeed, I usually wait a few seconds after that. //JIm
#110123From: Don Curtis/SYSOPFeb 20, 1988 11:54 PM
Jim, I would tend to agree that for new users, there perhaps should be a delay so that the 'disk copy finished' doesn't pop up until everything is really done. Don
#110017From: Doug WingerFeb 20, 1988 6:46 AM
Jim, Failsafe? Buy a calculator. I don't trust anything that tries to make my decisions for me. The problem is not the fault of the drive, but people that don't read warnings, ignore flashing lights, and have no real idea of what is going on. Leave the user some responsibilty, or do you want the Government to assign a monitor to every user to prevent any possibility of him making a mistake? I have yet to cause a hard error on my disks by pulling it out while busy because I read the manual before using the machine. You don't start messing with something new before having some idea of how it operates. If you want an automotive metaphor, would you allow someone to drive your car without supervision or instruction if this was the first time for them? Doug
#110037From: Jim VentolaFeb 20, 1988 12:51 PM
Geez, Doug, have you been out driving lately? I come from Boston, and I can tell you there are MANY drivers around here that should NOT be allowed near a car. I have even seen cruisers ignore flagrant violations. (In England, if ONE INCH of your car is on the pedestrian crossing at a red light you will be bawled out, embarrassed, fined etc. Here, NO ONE fails to get at least half the car into the intersection. Pedestrians MUST walk around.) Morality is fine. In books. If the government were to "monitor" driving better (i.e. more enforcement of the laws) it would be very good in my book. That may be analogous to "training" in your sense. And I would NOT mind cars that had built in mechanical means of preventing death and maiming in case of accidents. And I WOULD be willing to pay more for that. Another thing: reasearch has shown that there are various learning styles. Some folks read a manual before they do anything. Others like to jump right in and resort to the manual only as needed. Neither is more responsible or moral than the other, in my view. These are just styles of learning. // Jim
#110070From: Doug WingerFeb 20, 1988 5:48 PM
Jim, While different styles of learning are all fine and dandy, the ones who at least read the labels first have a better chance of surviving the learning experience. Caveat Utilitor (my latin stinks but you get the idea) I also am of the`jump in and use' school of learning, but I have at least a good idea of the hazards and pitfalls before doing so. As to having built in safeguards; sure, but make 'em optional. I resent any machine that does things `for my own good'. I do am at least somewhat inteligent and don't like to be told by an inanimate object that I am too stupid to operate it unless I do it by the book. Having seatbelts in the vehicle, and using them, is great but requiring crash-helmets, roll cages, automatic fire extinguishing, flame resistant clothing, air-bags, and a radar controlled colision advoidance mechanism is a bit of overkill for a passenger vehicle. Now, as to the disk problem the beginner had; I bet he won't do _that_ again. Consider it the price of learning that the Amiga is a multi-tasking machine right down to the DOS. He paid the price, learned a lesson, and ruined a disk. Wouldn't it have better if he had just _skimmed_ the manual before rushing in? I knwo I at least read the warnings before using highly expensive equipment. Especially if the price of it comes out of my wallet. While you point out there is no `right' way to learn, some ways inflict a higher expense than others. This was one such case. Just some thoughts, Doug
#110198From: John DraperFeb 21, 1988 4:38 PM
Jim, Morality aside, the user who makes assumptions about the operation of a mcahine must also bear the responsibility of what happens as a consequence. Those who do not read the docs have no right to criticize the manufacturer for those consequences. I should think it would be obvious to anyone. Regards, Larry.
#110105From: Claude J. GoschFeb 20, 1988 9:34 PM
The most frustrating thing I discovered the first time I worked with a Mac was trying to eject a disk. It had no eject button, so I finally had to ask someone else how to get it out. Even powering down & up didn't eject it. That wasn't my idea of "user-friendly"! Claude
#110138From: Jim VentolaFeb 21, 1988 12:43 AM
Ever trash a valuable MAC disk by ejecting it, Claude? Seriously, I _like_ the eject button. Honest. I just think it would be nice if it were disabled whenever the green light was on if doing that would cost under $20. // Jim
#110192From: KEITH YOUNGFeb 21, 1988 3:26 PM
Jim, uh… I think that the light on the drive is RED! … I'm sure that's what you meant, just thought I'd point that out. <grin> Keith
#110014From: Ben BlishFeb 20, 1988 5:08 AM
(Not taking a stand) Doesn't the Mac lock the disk in till it's ok to remove?
#110040From: Steve AhlstromFeb 20, 1988 1:40 PM
Yes it does. It's real fun getting it out when the computer crashes with the disk locked in. Hope you have a paperclip handy!
#110054From: Douglas MartinFeb 20, 1988 3:40 PM
I personally still prefer the Mac idea in concept, although I realize that there's major problems in practice with implementing it on a multitasking system. Yes, people *should* be trained, but no, in many cases they aren't (I see lot's of idiot user's at work). By the way, if you know a not very well documented trick (Commodore does not have a monopoly on cruddy documentation), paperclipping your Mac is a very rare requirement; if you still use Macs and run into such a crash, try holding down the mouse button when you turn on or reset the machine. The *biggest* problem as I see it with the situation of ejecting too early on the Amiga is this business of remembering to WAIT after the light first goes out until it blinks again (I assume that's trackdisk writing its buffers out).
#110056From: Don Curtis/SYSOPFeb 20, 1988 3:49 PM
Doug, As far as a floppy is concerned, CBM fixed the wait for 2nd time the light goes out deal. They now hold the light on thru both writes so all you have to do is simply wait for the light to go out the first time. HardDisks handle it their own way so some of the require you to wait for the 2nd write to the HD even with 1.2 Don
#110060From: Steve AhlstromFeb 20, 1988 4:12 PM
Douglas, I guess we just disagree in concept. I have this strange idea that anyone who buys a computer should read the manual that comes with it. Too many people have the idea of "if all else fails, read the manual". With any computer, electronic gizmo, machine, whatever, if things go wrong, and there are warnings about it in the manual, instructions, etc., it is unfair to blame anyone but the user. I certainly wouldn't consider using a new band saw, VCR, computer, toy robot, etc. without reading the docs. ANY computer will mess up a disk if you pull the disk before the write is finished. You don't open up your camera before the film is rewound (and expect the film to be undamaged) — don't pull the disk before the light is out. Since switching to 1.2, when the light goes out — it stays out (at least it does on my 1000 and 2000). Thanks for the info on ejecting Mac disks — I haven't played with one in years. On a multitasking machine, having the computer lock the disks in place is not too bright — hopefully CBM will never consider that option.
#110163From: Glenn NielsenFeb 21, 1988 11:09 AM
In the case of DiskCopy, it tells the DOS that the disk's are bad while it is doing the diskcopy to prevent normal file system access. When it is finished with the diskcopy, it tells DOS that it has the disk back, and DOS then does a disk validation on it's own, just as if you had put a different disk in. Perhaps the disk validation routine should pop up a requester saying "Validating Volume in drive DFx:" or some such to warn new users not to remove the disk. <Glenn
#110094From: Ben BlishFeb 20, 1988 9:04 PM
I don't need a paperclip. I have a _real_ computer. <grin> –Ben–
#110055From: Don Curtis/SYSOPFeb 20, 1988 3:46 PM
Ben, Yes…but it requires you to hit the 'eject' button on the screen to pop the disk…it doesn't have an eject button on the drive itself. Don