"Donald E. Kimberlin" <0004133373@mcimail.com> (05/08/91)
The "Hayes AT Command Set," a defacto standard used by
virtually every dial-up modem for more than a decade, has had some of
its functions patent protected, but that has not kept it from being
virtually freely used. Perhaps Hayes has kept a low profile on the
matter, preferring to let its name be spread by such wide use.
However, some news did recently occur that Hayes will, on occasion
protect its rights. The following is excerpted from an electronic
newsletter called {The Teleputing Hotline}, dated April 30, 1991:
"HAYES WINS DOUBLE DAMAGES IN PATENT SUIT
"The Hayes modem standard now has the force of law. U.S.
District Judge Samuel Conti doubled damages owed by Everex Systems,
VenTel and OmniTel to Hayes Microcomputer Products for "willful
violation" of the Hayes modem patent. The patent, for an escape
sequence with guard time, is at the heart of the `Hayes AT' command
set since it specifies how a PC will go from the online mode to the
command mode. Hayes compatibility has become a de facto standard in
PC modems of all speeds. Conti, who also awarded Hayes court costs,
said that willful infringers must not be allowed to wait five or six
years, then pay a low 1.75% royalty to the owner of a valid patent.
"Dennis Hayes said that, since a jury found in January his patent was
valid, `a number of people have come forward and talked to us about
licenses. Some have been concluded agreements and some negotiations
are underway.' The defendants in San Francisco will appeal, and Hayes
must also defend itself in Minneapolis against another infringer,
Multitech. Hayes added that the U.S. policy, increasing protection
for copyrights and patents, is now being emulated worldwide."
-----------
(Those interested in obtaining a subscription to {The
Teleputing Hotline} can contact the publisher at: 215 Winter Avenue,
Atlanta, GA 30317; FAX: 404-378-0794; Phone: 404-373-7634; MCIMail:
409-8960; GEnie: nb.atl; CompuServe: 76200,3025.)splee@gnu.ai.mit.edu (Seng-Poh Lee, Speedy) (05/11/91)
In article <telecom11.347.3@eecs.nwu.edu> is written: > The "Hayes AT Command Set," a defacto standard used by > virtually every dial-up modem for more than a decade, has had some of > its functions patent protected, but that has not kept it from being > virtually freely used. Perhaps Hayes has kept a low profile on the > matter, preferring to let its name be spread by such wide use. > However, some news did recently occur that Hayes will, on occasion > protect its rights. The following is excerpted from an electronic Interesting how Hayes goes after the smaller retail modem manufacturers. IBM and AT&T both also make and sell AT modems. I don't see Hayes challenging them in court. There are also a host of other modem manufacturers such as Codex, Universal Data Systems (both owned by Motorola), General DataComm (who I work for), Racal-Milgo, etc who also manufacture AT modems. However, these manufacturers don't concentrate on the retail markets, prefering to sell to corporate and RBOC accounts. These companies also make muxes, network management systems, and AT modems form only a small part of their business. I'd like to see Hayes go after these companies. You can bet that they will have a long court battle with Hayes before they will pay Hayes any royalty. Seng-Poh Lee splee@gnu.ai.mit.edu
Jerry Leichter <leichter@lrw.com> (05/13/91)
In a recent issue of TELECOM Digest, Seng-Poh Lee comments on the Hayes patent by finding it "interesting" that Hayes went after "smaller retail modem manufacturers" when many others, including the likes of AT&T and IBM, also make Hayes-compatible modems. He anticipates "a long court battle" with these guys before they pay up. In fact, he's probably dead wrong. Historically, the larger vendors are usually much more willing to pay reasonable royalties than to fight. There are a number of reasons for this. First, the big guys are not trying to survive by being the absolute lowest-cost producers. As a result, they can more easily pass through a small royalty than the scrappy "cut everything to the bone" price competitors, to whom every penny counts. Second, they have much more to lose than to gain: The courts have become much more willing to enforce patent rights of late, sometimes with BIG penalties. Companies that ask their lawyers whether they can safely ignore a patent claim are likely to be reminded of Polaroid's case against Kodak, which culminated in huge damages and in Kodak being forced entirely out of the instant photography market. Finally, the big guys sell "peace of mind": The last thing their customers want to hear is that the stuff they purchased is under a legal cloud. Naturally, to avoid being taken for chumps by anyone with any kind of tenuous patent claim, even the big guys will check a claim out carefully. But for a claim for which there is a reasonable case -- and certainly one that has already been accepted by a court will look VERY reasonable -- and for which the licensing demands are reasonable, the decision will usually be that it's better to pay than fight. Jerry
Henry Mensch <henry@ads.com> (05/13/91)
Seng-Poh Lee, Speedy" <splee@gnu.ai.mit.edu> wrote: > Interesting how Hayes goes after the smaller retail modem > manufacturers. IBM and AT&T both also make and sell AT modems. The chances are excellent that both IBM and AT&T license this technology from Hayes ... neither outfit is known for taking risks with this sort of stuff; they want their intellectual property respected, and behave in kind. Henry Mensch / Advanced Decision Systems / <henry@ads.com>
tnixon@uunet.uu.net> (05/13/91)
In article <telecom11.354.5@eecs.nwu.edu>, splee@gnu.ai.mit.edu (Seng-Poh Lee, Speedy) writes: > Interesting how Hayes goes after the smaller retail modem > manufacturers. IBM and AT&T both also make and sell AT modems. I > don't see Hayes challenging them in court. There are also a host of > other modem manufacturers such as Codex, Universal Data Systems (both > owned by Motorola), General DataComm (who I work for), Racal-Milgo, > etc who also manufacture AT modems. Some of the companies you mention have existing patent cross-license agreements with Hayes. Thus, no need for any lawsuits. There are dozens of other companies you didn't mention that also already license Hayes patents, either because they accepted Hayes' invitation to do so or as part of the settlement of other patent lawsuits (e.g., USRobotics and Prometheus were parties in the same lawsuit you referenced, but settled and have taken licenses). Many other companies have approached Hayes seeking licenses since the verdict was announced in the recent trial. The 170+ other US modem manufacturers have received, or soon will receive, invitations to do the same. So, don't jump to the conclusion that just because you haven't seen a media report of Hayes suing to enforce their legitimate patent rights against a particular company that (a) there's any need to, (b) they haven't, or (c) they won't. And remember that in the case of patents, there's no REQUIREMENT that the patent holder license to anyone, or license to everyone in a non-discriminatory fashion. An injunction prohibiting the further shipment of infringing products would not be a pleasant thing for your employer to endure, would it? > I'd like to see Hayes go after these companies. You can bet that > they will have a long court battle with Hayes before they will pay > Hayes any royalty. I'm sure GDC management would be pleased to know that you're publicly and actively seeking lawsuits against your company, and that you think a long and expensive court battle (in the face of a strong precedent-setting decision) would be desirable. Given GDC's financial condition, I'm sure your management would think twice before making such a decision, but they may not think twice about laying off a few people to raise money for the cause. Be careful. I won't comment on this further, except to say that I'm VERY surprised to see ANYONE, in the current highly-litigious high-tech industry, virtually inviting lawsuits, to the extent of laying down public challenges to do so! At the very least, such comments should be kept private. "So sue me" might be a cute thing to say to a neighbor who has complained of your dog crapping in his yard, but it's insane in business today. You sure won't catch ME leaving online messages asking other companies to sue MY employer! Toby Nixon, Principal Engineer | Voice +1-404-840-9200 Telex 151243420 Hayes Microcomputer Products Inc. | Fax +1-404-447-0178 CIS 70271,404 P.O. Box 105203 | UUCP uunet!hayes!tnixon AT&T !tnixon Atlanta, Georgia 30348 USA | Internet hayes!tnixon@uunet.uu.net
Ronald Greenberg <rig@eng.umd.edu> (05/13/91)
I don't know what are the exact details in this Hayes patent case, but
it sounds like an attempt to patent "look and feel", which can be
highly annoying. I don't have any recent information about activity
on this topic, but people may be interested in this old message from
Richard Stallman:
[4155] daemon@LCS.MIT.EDU bboard 05/30/89 15:02 (56 lines)
Subject: Protest Against Lotus Successful; Let's Organize Permanently
Date: Tue, 30 May 89 14:54:09 EDT
From: rms@ai.mit.edu
To: bboard@ai.mit.edu
Despite the threat of rain, we had large turnout for the protest
against user-interface copyright on Wednesday: 160 to 180, depending
on whose count. (The counts failed to include a couple of professors
who showed up just as we were leaving.) Bryan Kocher, president of
the National ACM (Association for Computing Machinery), marched with
us.
The organizers made around 30 signs, and many of the other
participants made their own. The best sign showed a strait jacket and
the caption, "Don't make me wear your suit." Once we were there, the
picketers all tried their hand at inventing euphonious chants. The
best ones were:
1-2-3 is not for me / Say no to monopoly.
Put your lawyers in their place: / No one owns the interface.
Hey, hey! Ho, ho! / Software tyranny's got to go.
Apple, Lotus, Look-and-feel: / Let's go reinvent the wheel.
And the world's first protest chant in hex:
1, 2, 3, 4 / Kick the lawsuit out the door.
5, 6, 7, 8 / Innovate, don't litigate.
9, A, B, C / Freedom, not monopoly.
D, E, F, 0 / Look-and-feel has got to go.
We were covered by reporters from the Associated Press, Reuters, Info
World, Computerworld, PC Week, MIS Week, MacWeek, Computer Reseller
News, National Public Radio, the Boston Herald, and the Boston Globe.
The stories I have seen are sympathetic and present our arguments
well. The AP article was carried in newspapers around the country.
All in all, we have done good work for the cause.
Numerous people said they could not attend but would like to help
fight "Look-and-Feel" in some other way. To make this possible, I
would like to make the League for Programming Freedom into a permanent
grass-roots organization.
Therefore, I'd like to announce the League's organizational meeting on
Thursday, June 15 at 8:30pm in the 8th floor "playroom" at 545 Tech
Square in Cambridge. This building is on Main Street, next to the
railroad tracks.
Those of you who came to the demonstration are invited as well.
I hope we will be able to choose officers at this meeting and start
enrolling members. Then we can aim for various interesting sorts of
protests starting this summer.
--[4155]--barmar@bloom-beacon.mit.edu> (05/13/91)
In article <telecom11.354.5@eecs.nwu.edu> splee@gnu.ai.mit.edu
(Seng-Poh Lee, Speedy) wrote --
[Moderator's Note: See earlier messages this issue for text. PAT]
If you had read the article carefully, you would have seen that it is
not the AT command set that is patented, it is the escape sequence
with guard time (i.e. the fact that you have to pause for a second
before and after the "+++" to go to command mode). Not all AT modems
do this.
Also, do you know for a fact that the modem manufacturers that you
mentioned above have not signed licenses with Hayes? I'll bet most of
them have. Maybe not specifically for this patent, but probably
general cross-license agreements. That's common in the manufacturing
world.
Barry Margolin, Thinking Machines Corp.
barmar@think.com {uunet,harvard}!think!barmarbud@uunet.uu.net> (05/14/91)
In article <telecom11.347.3@eecs.nwu.edu> 0004133373@mcimail.com (Donald E. Kimberlin) writes: > The "Hayes AT Command Set," a defacto standard used by > virtually every dial-up modem for more than a decade, has had some of > its functions patent protected, but that has not kept it from being > virtually freely used. Perhaps Hayes has kept a low profile on the > matter, preferring to let its name be spread by such wide use. In the interest of stemming panic out there, let's be quite clear. The enforcement was to a specific patent, not to the "Hayes AT Command Set". That patent, although quite important to modem users, does not cover the "AT" _command_ set. As a matter of fact, as PN-2120, a document from TIA committee TR-30.4, the command set has been submitted to the CCITT as part of draft recomendation "V.25 ter". One of the regular contributors to TELECOM Digest, Toby Nixon of Hayes is the chairman of TR-30.4, and can probably post an update of the status of this proposal, if asked. Communication software writers can, quite freely, write software which sends the string "AT whatever" and DCE equipment may recognize that string and act upon it. The patent upheld is on the method of notifying the DCE equipment that the next data arriving should be treated as a command to the DCE, as opposed to data to be transmitted to the far end; that is, switching to command mode. Most software defaults to a one second pause, transmitting three plus (+) signs, followed by a one second pause. Hayes patent is broader, covering any time delay, followed by any unique sequence. This patent, however, covers only async data ports used for both data and command. Synchronous ports, obviously, cannot pause in sending data, and the use of separate command and data ports is also outside the scope of the patent. For those in the external async modem business, the Hayes patent is a business expense that they will have to factor in. The rest of us can continue to type in "ATxxx" without worrying that the feds are going to come after us. Bud Couch - ADC/Kentrox standard BS applies
Randy Bush <news@psg.com> (05/14/91)
> The patent upheld is on the method of notifying the DCE equipment that > the next data arriving should be treated as a command to the DCE, as > opposed to data to be transmitted to the far end; that is, switching > to command mode. You mean kinda like one tells an X.25 PAD (i.e. Telenet et al.) to drop to command mode from data mode, <pause> "@" <cr> <pause> Seeing as the above and similar uses have been in use since the '70s, how did our friends from Norcross manage to patent it? Randy Bush / news@psg.com / ..!uunet!m2xenix!news
sbrack@cis.ohio-state.edu> (05/14/91)
kentrox!bud@uunet.uu.net (Bud Couch) writes: > As a matter of fact, as PN-2120, a document from TIA committee > TR-30.4, the command set has been submitted to the CCITT as part of > draft recomendation "V.25 ter". One of the regular contributors to > TELECOM Digest, Toby Nixon of Hayes is the chairman of TR-30.4, and > can probably post an update of the status of this proposal, if asked. > Communication software writers can, quite freely, write software which > sends the string "AT whatever" and DCE equipment may recognize that > string and act upon it. > Most software defaults to a one second pause, transmitting three plus > (+) signs, followed by a one second pause. Hayes patent is broader, > covering any time delay, followed by any unique sequence. This patent, > however, covers only async data ports used for both data and command. > Synchronous ports, obviously, cannot pause in sending data, and the > use of separate command and data ports is also outside the scope of > the patent. This is a valid patent?? I haven't studied patent law in depth, but I would think that a patent like that would fall under the realm of overbredth (sp?). Ha this patent been upheld in court? It certainly appears that Hayes has patented a PAUSE-DATA-PAUSE sequence. If that's true, then Hayes holds license on nearly every piece of data manipulation equipment ever manufactured. If a patent that broad covering a process that basic is admissible, then the first person to patent the breathing process is in for a fortune in royalties from all those people using his "technology" all this time 8) 8) 8). > For those in the external async modem business, the Hayes patent is a > business expense that they will have to factor in. The rest of us can > continue to type in "ATxxx" without worrying that the feds are going > to come after us. Steven S. Brack | sbrack%bluemoon@nstar.rn.com Jacob E. Taylor Honors Tower | sbrack@bluemoon.uucp The Ohio State University | sbrack@nyx.cs.du.edu 50 Curl Drive. | sbrack@isis.cs.du.edu Columbus, Ohio 43210-1112 USA | brack@ewf.eng.ohio-state.edu +1 (011) 614 293 7383 | Steven.S.Brack@osu.edu
max@uunet.uu.net> (05/15/91)
In article <telecom11.354.5@eecs.nwu.edu> splee@gnu.ai.mit.edu (Seng-Poh Lee, Speedy) writes: > Interesting how Hayes goes after the smaller retail modem > manufacturers. IBM and AT&T both also make and sell AT modems. I > don't see Hayes challenging them in court. There are also a host of > other modem manufacturers such as Codex, Universal Data Systems (both > owned by Motorola), General DataComm (who I work for), Racal-Milgo, > etc who also manufacture AT modems. Perhaps it's because IBM pays Hayes a licensing fee and the smaller retail modem manufacturers do not. I don't know about AT&T, though, sorry. max@gupta.com Max J. Rochlin decwrl!madmax!max
nanook@eskimo.celestial.com (Robert Dinse) (05/15/91)
In article <telecom11.355.6@eecs.nwu.edu>, kentrox!bud@uunet.uu.net (Bud Couch) writes: > In the interest of stemming panic out there, let's be quite clear. > The enforcement was to a specific patent, not to the "Hayes AT Command > Set". That patent, although quite important to modem users, does not > cover the "AT" _command_ set. > The patent upheld is on the method of notifying the DCE equipment that > the next data arriving should be treated as a command to the DCE, as > opposed to data to be transmitted to the far end; that is, switching > to command mode. > Most software defaults to a one second pause, transmitting three plus > (+) signs, followed by a one second pause. Hayes patent is broader, > covering any time delay, followed by any unique sequence. This patent, > however, covers only async data ports used for both data and command. Since the timed escape is an essential part of the command set, it is impossible for a modem manufacturer to claim Hayes compatability without it. Thus, Hayes, if successful at enforcing this patent, has a monopoly on this kind of modem. Indeed, I can't think of a way you could escape from data mode to command mode not using a timed delay and some unique sequence, that wouldn't be possibly contained in transmitted data. I agree with a previous poster that says this reeks of look and feel, it's more than that, it's functionality. It's not just like someone said, hey you can't make a spread-sheet that has that user interface, they've said, hey, you can't make a spreadsheet that works at all. The ability to escape from data mode to command mode is essential in a smart modems operation. The ability to do that in a way that guarantees that escape to command mode won't accidentally be invoked by the data stream would be difficult (I can't think of a way) without timing and a unique string being an essential feature of the escape from data mode. This type of BS really torques me. I have one Hayes modem and nine clones here, I will not buy another Hayes product. Not only must modem manufacturers figure this as a cost they have to figure in, but so must consumers. And since Hayes compatables comprise nearly all consumer type modems, we are essentially all being held hostage by Hayes, they, if successful in enforcing this patent, have a complete monopoly on the field. My feeling is that the Justice Department, gutted by Reagan and Bush, really should be filing anti-trust suits against corporations that participate in monpoly by litigation.
Mark Fulk <fulk@cs.rochester.edu> (05/16/91)
Others have made some good points about overbreadth of the patent, and prior art. I am bothered by another aspect: triviality. My three year old son has more complicated and better ideas every day. If the Hayes patent holds water, I'm going to patent the phrase "excuse me" in its use to interrupt a conversation. Seriously, a patent should only be granted if the invention solves a problem that many have found difficult, or if the new solution is not trivial to find and offers significant advantages over older methods. I AM in the process of patenting a few of my own ideas, all of them much more complex than the Hayes patent, and have in fact rejected the notion of patenting other ideas as, well, patently ludicrous. (pause) +++ (pause) is, in my view, not even an idea. I have decided not to patent much better inventions (such as a nice way to keep track of transient allocations in LISP, for example). Mark
"Michael H. Riddle" <riddle@hoss.unl.edu> (05/17/91)
In <telecom11.364.4@eecs.nwu.edu> fulk@cs.rochester.edu (Mark Fulk) writes: > Others have made some good points about overbreadth of the patent, and > prior art. > I am bothered by another aspect: triviality. My three year old son > has more complicated and better ideas every day. If the Hayes patent > holds water, I'm going to patent the phrase "excuse me" in its use to > interrupt a conversation. > Seriously, a patent should only be granted if the invention solves a > problem that many have found difficult, or if the new solution is not > trivial to find and offers significant advantages over older methods. [My comments are not particularly directed to Mark; rather, his was the post to which I replied for convenience.] I'm just a little surprised at the tone and direction most of the comments have taken, just as I'm a little surprised by Toby Nixon's silence. Perhaps the company has not allowed him to say anything, although I'd think an approved press release might be available and, if so, would certainly help clear the air. (If I missed one you posted Toby, then I apologize. I know you try hard.) Some points need to be made: a. The Patent and Trademark Office issued the patent. b. While the PTO isn't infallible, their acts get a presumption of validity. They have the job, not you or I. c. The PTO obviously felt the Hayes application met the requirements. d. A number of modem vendors agreed, or decided licensing was easier and cheaper than a patent challenge. My understanding is that quite a number of them went the licensing route, and that even more are holding discussions with Hayes. e. A relative few companies decided to challenge the PTO decision in court, and they lost. The courts, after complete briefing and arguments, agreed with Hayes and the PTO. Personal reaction: some of the simplest and most useful inventions seem obvious /after/ invention: paper clips and rubber bands would be good examples. Yet, /before/ the invention, they were unique and nonobvious. Sometimes what may be involved is the practical method of manufacture, or some other intermediate step that is not obvious or easy to implement. The point is that Hayes has an obviously defensible patent, one which patent lawyers for a number of companies have been unable (so far) to overturn and which even more patent lawyers have felt valid enough to have their clients license the technology. We need to keep this in mind. <<<< insert standard disclaimer here >>>> riddle@hoss.unl.edu | University of Nebraska ivgate!inns!postmaster@uunet.uu.net | College of Law mike.riddle@f27.n285.z1.fidonet.org | Lincoln, Nebraska, USA [Moderator's Note: Actually, Toby Nixon did send a good response to te Digest on this. Perhaps you are behind in your reading? PAT]
Ron Heiby <heiby@mcdchg.chg.mcd.mot.com> (05/18/91)
think!barmar@bloom-beacon.mit.edu (Barry Margolin) writes: > If you had read the article carefully, you would have seen that it is > not the AT command set that is patented, it is the escape sequence In fact, if memory serves me right, Hayes itself had to license the patent held by another (all but forgotten, probably) modem manufacturer who had patented the controlling of a modem by commands sent to its RS-232 port or to the same port as the data stream. Perhaps someone who knows/remembers better the details would chime in with them, or correct me if I'm wrong? Ron Heiby, heiby@chg.mcd.mot.com Moderator: comp.newprod
johnl@iecc.cambridge.ma.us (John R. Levine) (05/20/91)
In article <telecom11.371.3@eecs.nwu.edu> is written: > The ability to do that in a way that guarantees that escape to > command mode won't accidentally be invoked by the data stream would > be difficult (I can't think of a way) without timing and a unique > string being an essential feature of the escape from data mode. The other approach is to reserve some character sequence to mean switch to command state, and to have some way of protecting that sequence if it appears in data, most typically by doubling the first character of the sequence. This works perfectly well, and is what one does with synchronous modems, but means that the communications software on each end has to do some of the filtering, while the timed technique has the advantage of the escape sequence being so unlikely in the normal data stream that no protection is necessary. > My feeling is that the Justice Department, gutted by Reagan and > Bush, really should be filing anti-trust suits against corporations > that participate in monpoly by litigation. The whole point of a patent is to allow monopoly by litigation (or more typically by the threat thereof) for a limited period. On the other hand, someone pointed out that the way you escaped to command mode on an arpanet PAD was delay-@<cr>-delay, which is basically the same as Hayes approach and was almost certainly invented and more important published earlier. I haven't heard whether that example was presented to the court. Regards, John Levine, johnl@iecc.cambridge.ma.us, {spdcc|ima|world}!iecc!johnl
Eric A Rasmussen <ear@wpi.wpi.edu> (05/20/91)
In article <telecom11.371.3@eecs.nwu.edu> nanook@eskimo.celestial.com (Robert Dinse) writes: > In article <telecom11.355.6@eecs.nwu.edu>, kentrox!bud@uunet.uu.net > (Bud Couch) writes: >> Most software defaults to a one second pause, transmitting three plus >> (+) signs, followed by a one second pause. Hayes patent is broader, >> covering any time delay, followed by any unique sequence. This patent, >> however, covers only async data ports used for both data and command. > Since the timed escape is an essential part of the command set, > it is impossible for a modem manufacturer to claim Hayes compatability > without it. Thus, Hayes, if successful at enforcing this patent, has a > monopoly on this kind of modem. Indeed, I can't think of a way you > could escape from data mode to command mode not using a timed delay > and some unique sequence, that wouldn't be possibly contained in > transmitted data. Perhaps I am missing something obvious, but for computer equipment such as PC's (as opposed to dumb terminals) which have control of basically all the serial port lines, what would be so hard about using the terminal ready line to toggle between command mode and data mode? I always disable the 'feature' in my modem where it disconnects when the terminal ready line is not high. After all, what's the point if I can use the <pause> +++ <pause> ATH method to disconnect when I want. If I ever have to reboot while on-line I don't want to lose my connection. (Yes, this does happen. Say you download a program, run it, and it crashes.) Thus, I really have a free line which could readily be used for this purpose if I didn't have a Hayes compatible modem. Eric A. Rasmussen ear@wpi.wpi.edu ear%wpi@wpi.edu
"Marc T. Kaufman" <kaufman@neon.stanford.edu> (05/20/91)
In article <telecom11.379.1@eecs.nwu.edu> johnl@iecc.cambridge.ma.us (John R. Levine) writes: > In article <telecom11.371.3@eecs.nwu.edu> is written: >> The ability to do that in a way that guarantees that escape to >> command mode won't accidentally be invoked by the data stream >> would be difficult (I can't think of a way) without timing and >> a unique string being an essential feature of the escape from data >> mode. > The other approach is to reserve some character sequence to mean > switch to command state, and to have some way of protecting that > sequence if it appears in data, most typically by doubling the first > character of the sequence. This works perfectly well, and is what one > does with synchronous modems, but means that the communications > software on each end has to do some of the filtering, while the timed > technique has the advantage of the escape sequence being so unlikely > in the normal data stream that no protection is necessary. Funny you should say that. In fact, the usefulness of "+++" is fast coming to a close, because many modems are at the far end of a network with buffering, and there is no good way to insert "time" into a buffer. My understanding is that the next generation of modems will use the old BiSync technique of DLE (Data Link Escape) + character to send commands to the modem. DLE + DLE will be sent as a single DLE to the other end. There is no more problem with this (from the point of view of computers) than there is with XON/XOFF. (Yes, I know that BiSync used DLE as a framing escape rather than a modem escape, but the principle is the same) Marc Kaufman (kaufman@Neon.stanford.edu)
Alan Bland <mab@druwy.att.com> (05/20/91)
In article <telecom11.371.3@eecs.nwu.edu>, nanook@eskimo.celestial.com (Robert Dinse) writes: > monopoly on this kind of modem. Indeed, I can't think of a way you > could escape from data mode to command mode not using a timed delay > and some unique sequence, that wouldn't be possibly contained in > transmitted data. One technique I've seen is to use a timed BREAK to switch to command mode. A short BREAK is passed on the communications line, but a longer one (e.g. two seconds) goes to command mode. So there is indeed another way to do it without infringing on the Hayes patent, though you'd probably be violating someone else's patent :-) Alan Bland att!druwy!mab == mab@druwy.ATT.COM AT&T Bell Laboratories, Denver CO (303)538-3510
"Elliott S. Frank" <esf00@uts.amdahl.com> (05/21/91)
My, how short the memory of almost everyone commenting on the validity
of the Hayes patent.
Not too long ago, the way that you got a modem to "dial" was that you
used an ACU (Automatic Calling Unit) in conjunction with the modem.
You gave the ACU the number to call, and it dialed the line on behalf
of the modem. When your modem call was completed, you could drop DTR,
signalling the modem to drop the line. You then redrove the ACU for
your next outbound call. [There are probably hundreds of WeCO 801
ACU's still in service ...]
The first modems that allowed dialling over the data line used various
schemes to get the attention of the dialler. The Hayes scheme was only
one of several available in the late Seventies/early Eighties. The
ubiquity of the Hayes scheme is due to brilliant marketing, not to
being the only viable mechanism for escaping the dialler.
Elliott Frank ...!{uunet,sun}!amdahl!esf00 (408) 746-6384
or ....!esf00@amdahl.com
[the above opinions are strictly mine, if anyone's.]Charlie Mingo <Charlie.Mingo@f421.n109.z1.fidonet.org> (05/21/91)
nanook@eskimo.celestial.com (Robert Dinse) writes: > [S]ince Hayes compatables > comprise nearly all consumer type modems, we are essentially all being > held hostage by Hayes, they, if successful in enforcing this patent, > have a complete monopoly on the field. > My feeling is that the Justice Department, gutted by Reagan and > Bush, really should be filing anti-trust suits against corporations > that participate in monopoly by litigation. The purpose of patent law is to grant a limited monopoly to the inventor of a new product or process. Every patent, therefore, creates a monopoly which would be illegal under the antitrust laws, except that patents are specifically excluded from antitrust prohibitions. Now if Hayes were attempting to enforce an invalid patent, or attempting to expand the scope of the patent beyond that granted by the government, it might be a different story. In this case, the patent is valid and, as a consequence, so is any monopoly created thereby.
tnixon@uunet.uu.net> (05/21/91)
In article <telecom11.371.1@eecs.nwu.edu>, riddle@hoss.unl.edu (Michael H. Riddle) writes: > I'm just a little surprised at the tone and direction most of the > comments have taken, just as I'm a little surprised by Toby Nixon's > silence. Perhaps the company has not allowed him to say anything, > although I'd think an approved press release might be available and, > if so, would certainly help clear the air. (If I missed one you > posted Toby, then I apologize. I know you try hard.) Since the matter is still in litigation (both the appeal of the Everex/Omnitel/VenTel suit, and the Multitech trial and others), I cannot comment except on historical facts. As much as I might like to, I cannot comment on anything pertaining to matters that might be the subject of a trial. One thing I will say: another writer (not Mike Riddle) stated his belief that the US Justice Department should "go after" companies that try to monopolize by lawsuit. As a matter of historical fact, I should point out that the defendants in the recent trial (Everex, Omnitel, VenTel), and several others which settled before the trial, ORIGINATED legal action AGAINST HAYES. They filed suit claiming that Hayes was attempting to monopolize the modem market by enforcing the Heatherington patent. Hayes counter-sued for patent infringement (i.e., we didn't start it, they did). The claim of monopolization was thrown out of court by summary judgement -- the court found no merit at all in their claim. The judge did, however, find merit to Hayes' claim of patent infringement, which eventually went to trial, the result of which everyone is now aware. > [Moderator's Note: Actually, Toby Nixon did send a good response to te > Digest on this. Perhaps you are behind in your reading? PAT] My previous response was to an employee of another modem company that appeared to be inviting a lawsuit. I considered that to be a separate matter from the issues related to the validity of the Heatherington patent and the related cases. I was subsequently contacted by the management of that company, and it was clearly stated that the opinions expressed by that employee were not those of the company or its management (as I suspected). Toby Nixon, Principal Engineer | Voice +1-404-840-9200 Telex 151243420 Hayes Microcomputer Products Inc. | Fax +1-404-447-0178 CIS 70271,404 P.O. Box 105203 | UUCP uunet!hayes!tnixon AT&T !tnixon Atlanta, Georgia 30348 USA | Internet hayes!tnixon@uunet.uu.net
andys@ulysses.att.com (05/21/91)
In article <telecom11.371.3@eecs.nwu.edu> is written: |> Since the timed escape is an essential part of the command set, |> it is impossible for a modem manufacturer to claim Hayes compatability |> without it. Thus, Hayes, if successful at enforcing this patent, has a |> monopoly on this kind of modem. Indeed, I can't think of a way you |> could escape from data mode to command mode not using a timed delay |> and some unique sequence, that wouldn't be possibly contained in |> transmitted data. |> The ability to escape from data mode to command mode is essential |> in a smart modems operation. The ability to do that in a way that |> guarantees that escape to command mode won't accidentally be invoked |> by the data stream would be difficult (I can't think of a way) without |> timing and a unique string being an essential feature of the escape |> from data mode. |> This type of BS really torques me. I have one Hayes modem and |> nine clones here, I will not buy another Hayes product. |> Not only must modem manufacturers figure this as a cost they have |> to figure in, but so must consumers. And since Hayes compatables |> comprise nearly all consumer type modems, we are essentially all being |> held hostage by Hayes, they, if successful in enforcing this patent, |> have a complete monopoly on the field. |> My feeling is that the Justice Department, gutted by Reagan and |> Bush, really should be filing anti-trust suits against corporations |> that participate in monpoly by litigation. You really ought to talk to somebody who knows a little anti-trust law and intellectual property law (which you *CLEARLY* do not) before you make such wild and irresponsible statements. Seeking patent protection for an invention has never been considered anti-competitive behavior. Why have a patent office if you prosecute everybody who uses it? Hayes invested the R&D dollars into developing an essential technology for the implementation of smart asynchronous modems and now will get a return on their investment. This is wrong? Tell me, does celestial.com (whatever that is) give away its assets? If so, I will watch for your Chapter 11 filing, since you won't be in business very long. TI gets royalties from nearly everybody manufacturing integrated circuits (although a lot of the basic US patents are probably due to expire) including the Japanese. They didn't do every single chip design, but they did invent the enabling technology, and are getting paid for it. The Hayes patent is no different. Enforcing patents, and deciding who may and may not license your patents is not a matter of restraint of trade under current anti-trust case law. This is not a lack of vigilance by our current fascist government, it's the way anti-trust law and patent law have intersected for a long time. Patent protection usually enables publication of technology, since the inventor is given something in return for publishing the patent. Go talk to somebody who knows about such things. You don't. Andy Sherman/AT&T Bell Laboratories/Murray Hill, NJ AUDIBLE: (908) 582-5928 READABLE: andys@ulysses.att.com or att!ulysses!andys What? Me speak for AT&T? You must be joking!
bud@uunet.uu.net> (05/22/91)
In article <telecom11.371.3@eecs.nwu.edu> nanook@eskimo.celestial.com (Robert Dinse) writes: > Since the timed escape is an essential part of the command set, > it is impossible for a modem manufacturer to claim Hayes compatability > without it. Thus, Hayes, if successful at enforcing this patent, has a > monopoly on this kind of modem. Indeed, I can't think of a way you > could escape from data mode to command mode not using a timed delay > and some unique sequence, that wouldn't be possibly contained in > transmitted data. I was writing from my own rather parochial point of view. My company (that is, the one I work for, not own) makes Switched 56 DSU's. These are 56 kB _synchronous_ units. We use a separate _async_ command port (separate from the data port which handles the signal going to the far end) to take care of configuration, maintenence, and call control. The language used by that command port is the AT set. To call a number, the command ATDxxxxxxxx is entered to the command port. When the far end answers, the data is sent through the data port. To disconnect - ATH0. No "(pause)+++", no patent infringment. Bud Couch - ADC/Kentrox If my employer only knew. standard BS applies
nanook@eskimo.celestial.com (Robert Dinse) (05/23/91)
Several points:
Regarding XON/XOFF - When is the last time you tried to use this
with UUCP? If you have, then you would know why that is not a workable
scheme. Indeed, when you don't know what data is going to be
transmitted over a link, there is no way to guarantee any particular
stream of data will be unique.
In regards to the old schemes that used a seperate dialer, with
two serial ports, or if you rely on control lines to signal an escape
to command mode, yes then there are alternatives. But there are
computers that have neither of those options available to them (mostly
low end machines).
Of the schemes that did allow escape from data mode to command
mode with a single port, with no control line intervention, and
without break, only using ASCII data, because that is all some systems
can generate, what other options are available?
The other problem I have with Hayes is philosophical. It's one
thing to patent a paper clip, it's quite another to patent >ANY<
method of binding loose papers together.
If Hayes had patented say using a pause of some defined value,
followed specifically by '+++' followed by another pause that would be
one thing, quite different from patenting ANY time delay followed by
ANY unique character string followed by another time delay. Also, if
Hayes had enforced this from the beginning >BEFORE< it had become an
accepted standard I'd have different feelings about it. But I feel
waiting until now, until everybody is using it is slimey at best.