China DOS Union

At present, the machines at my workplace all do not support USB boot, but I have a 20G mobile hard disk in hand, and I really want to make it bootable, so I can use the large amount of maintenance software on the mobile hard disk for daily maintenance. The current floppy boot disks can already recognize USB hard disks, so I was wondering whether it would be possible to boot the machine from drive A:, recognize the USB hard disk, and then choose to boot the computer from the USB hard disk. Since selective booting from floppy/hard disk/CD-ROM can already be done, USB booting should also be possible. I hope some expert can give me some advice!

I oppose making this kind of software. Old machines not having USB boot is only a small problem, and everyone can tolerate it. If this kind of software is made, then there will no longer be any way to force BIOS manufacturers to directly support USB sector read/write calls from the BIOS. These BIOS manufacturers may then get lazy and go back to that miserable era like CDROM had (an era like that is easy for people to exploit, causing everyone to get tricked). Everyone should call on, and force, BIOS manufacturers to provide USB sector read/write support directly for DOS users, just like reading and writing hard disks, using int13. Quite a lot of new motherboards already meet this requirement, and we should not continue giving those substandard motherboards room to survive. We are users, or software developers, and we must work to protect our own interests, so that software follows standards and a good environment for survival can be created. We should refuse to let others control us, and refuse those nonstandard, seriously defective products, whether intentionally or unintentionally so.

My position is this: for our ultimate benefit, even if some expert is capable of making this kind of software, you should not make it. If you have already made it, then you should not make it public either.

This is really unbelievable, the poster above actually thinks like that? Then for the sake of convenience, we might as well not study the DOS operating system either, since Windows is so much more convenient! Your view is too biased, isn't it? 99% of the machines at our workplace don't have USB boot capability—does that mean just to save money we should neither update the motherboard software nor want this?

Any person capable of making this software has the freedom to make it, and also the freedom to release it. However, my view is that I hope he will not make it, and even more so not release it. Someone capable of making it is not an ordinary person; he must be capable of making his own judgment, and he must be able to distinguish right from wrong and weigh the pros and cons. He follows his own judgment in deciding his actions, and that is completely free.

I also have my freedom, which is to state my viewpoint here. I do not force others to agree with my viewpoint; everyone can reserve their own opinion. If you do not agree with my viewpoint, that is fine. The world is diverse by nature, and there is no need to insist on complete uniformity.

Brother Budian has great foresight. From a strategic point of view, it really should be like this! Support Budian!

Brother blue, thanks for your support. But as for those who do not support it, I thank them too. My personal view is that when BIOSes can all support USB sector read/write int13 calls and this has already become a de facto industrial standard, then it will not be too late to write compatibility drivers for old machines. Our room for survival seems very tight, and our huge industry is far too fragile in its ability to deal with a tiny handful of ugly forces. Some people have already noticed that Microsoft has always been unwilling to let its operating systems boot from USB mobile devices, just as it originally did not want people to run from CDROM. Obviously, giving users too much convenience does not fit Microsoft's interests. Microsoft's goal is to obtain final control and hold the keys to the users' front doors. If I were Microsoft, I would also do everything possible to muddy the waters, make USB boot a total mess, keep USB boot from forming regularity and standards, and in the end have Microsoft step forward to formulate a new standard under which USB devices can only be used under windows, while at the BIOS and DOS stage everyone would be unable to use USB devices. Of course, on the other hand, Microsoft may not necessarily have that degree of control this time. It has already moved far away from DOS development, so it has no way to control people's continued attention to and research into DOS. So we do not need to worry too much. But no matter what, I personally still feel it is better to be on guard. After all, with CDROM we really suffered a huge loss. Maybe I am too sensitive myself, but that is because I have the feeling of “once bitten by a snake, ten years afraid of well ropes.” Look at the CDROM situation—it is our eternal pain. We will always have to keep sticking on band-aids, and always have to live with the awkwardness.

Actually, it is not only USB itself. In the future, if 1394 becomes popular, it could also follow the currently successful experience of USB. If USB were to fail like CDROM did before, then future 1394 and even newer devices might continue to suffer from Microsoft's sabotage.

Simplicity is beauty, but those evil forces insist on making simple things abnormally complex and distorted; their purpose is ugly.

I agree with Budian's opinion. Back when I discussed GRUB for DOS with Budian regarding support for booting CD ISO files, I already came into contact with this issue, and it can be said that I directly experienced the harm caused by M$'s self-made standards. This time I strongly agree with Budian's viewpoint. “Of course, on the other hand, Microsoft may not necessarily have that degree of control this time. It has already moved far away from DOS development, so it has no way to control people's continued attention to and research into DOS.” This is also absolutely correct. Fortunately, because people have continued paying attention to and researching DOS, support for USB devices such as USB flash drives under DOS is now basically no longer much of a problem. After loading the USB driver under DOS, my USB flash drive can be operated normally under DOS on all several of my computers.

So you guys are just treating me like a newbie and not going to help me, huh!

Why doesn't the poster above understand? What we're saying isn't aimed at you personally, it's a call for motherboard BIOS developers to improve support for new devices such as USB. The phonix homepage introduced their next-generation BIOS, with perfect support for USB2.0, IEEE1394, and many other new devices. Looking forward to motherboard manufacturers adopting it as soon as possible.

Brother, we're all the same. I myself have one machine at home and one in the office. I am not a computer room administrator, so I do not have many machines available. From that point of view, I'm far worse off than you—you at least have many machines you can use.

Whether at home or in the office, the machines I use are all old machines and do not support USB boot. I do not feel this causes any great inconvenience, and I have not set about solving this problem either. What I am considering now is that the machine at my home is too old, and it is time to buy a new one. I am wondering, if I buy a new one, will I definitely happen to get a good one that can perfectly support USB boot? I am very worried about this, because it is hard to see wording like “XXX motherboard supports USB boot” in advertisements, so I really am worried.

USB boot is a major feature. I think that if a new motherboard has this function, it must clearly state it in advertisements, to show that this motherboard is not an outdated product. Not only should it support USB boot, it should also allow DOS to directly use the int13 disk interrupt, without relying on any drivers, to directly read and write USB sectors. That is the most proper solution. This is the task of the motherboard BIOS, and any motherboard that can complete this task beautifully is without question a good motherboard worthy of trust. A motherboard with this capability must make its identity clear, show that it is different from other motherboards, and show that it was carefully crafted rather than crudely manufactured.

Brother, you can replace your own machine, or replace it with one of those so-called new motherboards in the future, but do the hundreds of machines at my workplace that don't support USB boot all have to be replaced too? Using a USB hard disk for maintaining hundreds of machines is very convenient. The only regret now is that they can't boot directly from USB, so I still have to carry CDs and floppies! What's even more troublesome is that some machines don't even have a CD-ROM drive, and the floppy drive is broken too. When the system crashes, the hard disk has to be removed. If the function I mentioned could really be achieved, it would save administrators a lot of time. Actually I think this is not much trouble for you people, but without this thing, will hardware makers really move any faster?

Let me analyze this situation.

For motherboard manufacturers, they certainly hope everyone will eliminate old machines as quickly as possible and buy new ones instead. That way, their customers will increase and their sales will keep going up. For that reason, they will not develop or release this kind of software. Developing this kind of software would be a thankless task and would bring no benefit to those motherboard makers.

Since the USB technical documentation of various companies has not been made public, and the technical details are hard to unify, anyone who wants to develop this kind of software will find it far from easy. Often a piece of software may work with one company's products but not another's, resulting in limited usability, or what may be called incompatibility. In the long run, just dealing with the BUGs reported by users would probably be impossible, and I believe the developer would eventually lose patience. Because of the compatibility problems that are hard to escape, even if such software were successfully developed, it would not become especially popular.

In short, I think no one is willing to take on this thankless job.

I think at present, with USB flash drives and CD-ROM drives, basically there's nothing that can't be done! There's no need to go that far!