MS Patents IM Feature Used Since At Least 1996

splorp! writes "Once again, a company is patenting a feature that another company implemented years before. C|Net's News.com reports that patent no. 6,631,412 grants Microsoft the rights to 'an instant messaging feature that notifies users when the person they are communicating with is typing a message.' Excuse me? Does anyone remember Powwow (now defunct)? I remember using that one back in '96 and it alerted the other people to whom you were chatting that you were typing. Or, alternately, it allowed you to SEE the other people typing in real time. Yeah, Powwow is gone, now, but that doesn't mean those features never existed."

Re:Don't forget

...AOL and Yahoo were not available for comment.

but they were typing a responce...

Re:Don't forget (actually)

I remember it distinctly because my girlfriend's Yahoo wasn't working

Don't worry, it happens to everyone.

Even older prior art

What about the UNIX "talk" command? That command allows you to see what the other person is typing in real time and it's been around forever. I wouldn't be surprised if there were cave paintings showing our ancestors using "talk" to tell their buddies how the wooly mammoth hunt was going.

--------
The fake Gzip Christ isn't not user number ~0xA6CA7

History of "talk"

There is a very interesting post (dated Dec. 2002) by David P. Reed [oreillynet.com] on the origin of 'talk' at: postel.org [postel.org]

In short, this goes back to at least 1967. I'm sure there is no way our esteemed patent office could possibly have found prior art back that far, let alone what happened last week. Someone should alert them to the existence of google.

Re:History of "talk"

As I've posted time and again on every "patent on prior art" Slashdot post since 2000 at least: the PTO has gone on record (including in an interview here at slashdot a couple of years ago) to say that the only source they have or use for Prior Art investigations is their own database. If a patent application has been filed on it, there's prior art. If it hasn't, then there isn't any prior art and it never existed before.

The PTO just automatically assumes that anything one person feels worthy of patenting is something that everybody else should have felt it worthy.

That's it. No google, no interviews with field experts, nothing. If a patent's been filed, there's prior art. If not, then it passes the "new" test.

Re:History of "talk"

In my limited experience, that's not entirely true. I had to fight like crazy with a patent examiner over a patent I obtained.

He did nontrivial outside research in the field, much of it directed by the reference materials I included in the patent. At one point he stated that a particular claim was "obvious" after you've read five different sources in different domains which he only knew about because we referenced all five in the application. None of them were patented.

From the Slashdot "IP is bad" standpoint you'd have to give him credit for the effort. He worked very hard to ensure that my patent was in fact non-obvious and not prior art. You really want a patent examiner that hard.

Except I don't. If patents are being given out like candy, why should I have to fight for mine?

Re:Even older prior art

RTFP. From this week's "great innovation for customers":

Unlike telephonic communication, when participants know that a person is speaking, participants in an instant messaging session do not know that somebody is preparing a message for transmission. Without a cue that the other person is transmitting information, it is difficult to have a smooth conversational flow. One mechanism that addresses this problem is employed by a UNIX "talk" program, which performs a character-by-character transmission of an instant message. That is, each time individual types of a single character on the computer keyboard, that character is transmitted to all other participants in the instant messaging session. Because other participants are essentially watching the person type, there are clear cues that a user is "talking."

However, this approach has several limitations. First, character-by-character transmission greatly increases the flow of network traffic because each character requires one or more data packets to be sent to each participant in the instant messaging session. In addition, many users do not like to be "watched" as they type, as their typing errors and incomplete thoughts are transmitted before they can be corrected. Finally, message recipients are often distracted by watching the flickering screen in which characters appear one time as a complete message is formed. Therefore, it can be appreciated that there is a significant need for a system and method that will provide the desired notification of user activity in a computer network. The present invention provides this, and other advantages, as will be apparent from the following detailed description and accompanying figures.

As far as I can see from a quick reading, the idea is not that you see what people are typing, but that you have an indicator which lets you know that they are typing.

Re:Even older prior art

The problem here seems to be that Microsoft is planning on using patents, simple ones, to attack OSS and the USPTO is helping. Just like SCO is banging its drum about IP when there's no proof of validity and is devised only as a stock price enhancer, Microsoft will use its patents to thwart OSS projects. Who can afford the court costs to fight these obvious patents? The USPTO is a major threat to OSS IMHO.

Anyways, to get around THIS patent, any IM application can device a single button which stays green when it receives the characters from a users input and turns red when a CR or LF character is recieved. Sure each character is still sent over the network but if it's buffered on every users machine, it can just be moved to the TALK window when the termination character is sent and thereby eliminating the whole message being sent out again when the sender has finished with the message.

This uses existing techniques and provides the notification mechanisms without seeing the senders thoughts as they are generated.

The OSS community needs a forum for debunking these patent applications. One where the USPTO trusts it for prior art inspections. Otherwise, Microsoft will litigate the OSS community into stagnation by killing OSS projects and improvements in the courts with bogus patents such as this one.

LoB

Re:Even older prior art

The OSS community needs a forum for debunking these patent applications. One where the USPTO trusts it for prior art inspections.

Now that's a good idea. Have you thought of patenting it? ;-)

Re:Unix talk

Sheesh, read the friggin patent:

One mechanism that addresses this problem is employed by a UNIX "talk" program, which performs a character-by-character transmission of an instant message. That is, each time individual types of a single character on the computer keyboard, that character is transmitted to all other participants in the instant messaging session. Because other participants are essentially watching the person type, there are clear cues that a user is "talking."

However, this approach has several limitations. First, character-by-character transmission greatly increases the flow of network traffic because each character requires one or more data packets to be sent to each participant in the instant messaging session. In addition, many users do not like to be "watched" as they type, as their typing errors and incomplete thoughts are transmitted before they can be corrected.

ICQ

ICQ had/has this as well, in the direct chat (not im) mode.

Re:ICQ

Heck, even Unreal Tournament made your character look like he was chatting into the radio while you were typing a message.

Also..

it allowed you to SEE the other people typing in real time

ICQ has done that for a while too

Are you sure?

Are you sure this patent grants them the rights to any implementation, or only their implementation?

Btw, would you need to sue MS in order to get this patent overturned, or could you do something like sue the patent office?

Prior Art may be the key

Check out this site [tms.org] for complete details but to lift a few important parts:

a person is not entitled to a patent if the invention was "known or used by others in this country, or was patented or described in a printed publication in this or a foreign country" before the date of invention by the applicant for the patent

But later there is a brief comment:

Naturally, if an inventor abandons the invention, he or she cannot obtain a patent.

And finally in support of M$'s patent, and likely the way they got it:

In a fast-changing world, finding a single piece of prior art which discloses the same invention as that claimed in a patent is not the most likely scenario. What is far more likely to occur is that the prior art will be something similar but not identical to the patented invention. The patent statutes also provide for this situation--in a negative manner. Specifically, section 103 of the code provides that a patent may not be obtained "though the invention is not identically disclosed or described [in the prior art] if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art." The test which is posed by this section is whether a worker of ordinary skill, knowing the prior art, would have found the patented invention obvious.

More text than code

What I find amusing is that it probably took a lot more time filling for this patent than implement the feature.

One must be very creative to describe such a simple feature in so many pages of text!

It's the application date that matters

Just because Microsoft was granted the patent now doesn't really mean anything. It takes years for a patent application to wind its way through the Patent Office. Because of that they are retroactive to the time of application.

The question shouldn't be, "How can they do this if had it in '96?" It should be, "When did Microsoft apply for this patent?"

Re:It's the application date that matters

if you look at the patent, it looks like december 2002.

Re:It's the application date that matters

what amazes me is that they filed this in Dec 2002 and in less than 10 months they were awarded the patent. AND there appears to be alot of prior art.

Mabye the USPTO needs to start getting emails, from us, pointing out the prior art.....

LoB

incorrect it is July 21,1999

this is a continuation of a previous application with a 1999 filing date. the effective date of the application, the begining of its patent protection, and the date of which prior art must be earlier than is July 21,1999.

No doubt the MS press release will say.....

This is another example of Microsoft's long history of "innovation".

Errrr, couldn't agree more personally.

Link to patent

Here is a link to the patent itself: 6631412 [uspto.gov]

It should be noted that UNIX talk is specifically talked about in the patent and the advantages of this system over it are mentioned. This does not get around the apparant prior art of POWWOW. Remember that it is the claims of a patent that are important, not the abstract. It appears from quickly looking at the claims, that the broadest requirements are for client A to send a message to client B that client A is typing. Then client B must indicate that client A is typing. Finally, that message is turned off when client A sends another message that it is done typing. The initial typing message must be based upon typing within a predefined period of time.

Any prior art asserted against this patent would need to have been in use on or before July 21, 1998.

Available since at least 1986...

That's the first I remember seeing it on a chat BBS, anyway. Lambda Switchboard software. At least two of the original systems are still online - I'm sure a few slashdotters know what I'm talking about. LOIS, TREX I, TREX II, and.. LOLA and LANE, I think?

The DOS-based Lambda software was replaced years ago with the Unix-based Mu clone, but it's still got the idle indicator in the 'F'ull who listing.

Patent dates back to at least 1999...

As shown in the PTO hyperlink in the article, "This is a continuation of U.S. patent application Ser. No. 09/359,337, entitled "System and Method for Activity Monitoring and Reporting in a Computer Network," filed on Jul. 21, 1999 now, U.S. Pat. No. 6,519,639."

On a personal note, there is CLEARLY prior art --as others have said talk/ytalk had this. Heck, a direct modem connection with a friend and seeing each other type exhibits this behavior even though that's hard to lump under the context of "An IM session".

This really feels like a defensive patent, not something they could turn around and sue AOL or Yahoo (or even Trillian or Jabber) over.

Xentax

Ah-HAH

Wonders never cease, I decided to continue Reading The Frickin' Article, and found some useful tidbits.

"Unlike telephonic communication, when participants know that a person is speaking, participants in an instant messaging session do not know that somebody is preparing a message for transmission. Without a cue that the other person is transmitting information, it is difficult to have a smooth conversational flow. One mechanism that addresses this problem is employed by a UNIX "talk" program, which performs a character-by-character transmission of an instant message. That is, each time individual types of a single character on the computer keyboard, that character is transmitted to all other participants in the instant messaging session. Because other participants are essentially watching the person type, there are clear cues that a user is "talking."

However, this approach has several limitations. First, character-by-character transmission greatly increases the flow of network traffic because each character requires one or more data packets to be sent to each participant in the instant messaging session. In addition, many users do not like to be "watched" as they type, as their typing errors and incomplete thoughts are transmitted before they can be corrected. Finally, message recipients are often distracted by watching the flickering screen in which characters appear one time as a complete message is formed. Therefore, it can be appreciated that there is a significant need for a system and method that will provide the desired notification of user activity in a computer network. The present invention provides this, and other advantages, as will be apparent from the following detailed description and accompanying figures."

So the claimed innovation here is simplifying real-time, continuous updates by just sending activity updates. Hmm. I'm not sure that really passes the tests for either "obvious" or actually "innovative", but at least they address talk.

Xentax

Thousands of Minitel services did it

About 10 years ago, thousands of free and non-free Minitel (french bbs-like) servers did it.

Including real-time chat that let you see every key stroke of other users.

Criminal at this point

It is obvious that the patent office is ignoring the prior art clause. Why not just file a criminal charge against the staff of the patent office and use the law to stop this kind of behavior. If the office is failing to perform it's job why not confront it in a court of law? Hell you could push as far as treason if need be (only takes two witnesses last I checked) as a conscious act to undermine the Constitution, federal law, and confront it as an act of economic sabotage. There are plenty of ways to confront the problem. I find it odd that the EFF and ACLU have not touched the Patent Office in earnest. What is protecting the Patent Office such that even basic avenues are not used? At the point that the Patent Office has ignored it's purpose I would most likely pursue legal action first based on ignoring the law and establishment of the "Prior Art" rules and if there is still no change after that, move to the treason avenue. Sad to say that treason is becoming more and more reasonable when looking at the larger picture of the Patent Office problems.

Another obvious patent

I'm not so much bothered by the prior art issue -- I have a much bigger issue with this patent. I'm willing to bet that if you were to take an average programmer and ask them "how can I modify this IM program so that the person you are talking to knows that you are currently typing without actually sending each character as you type it?", they'd come up with the exact same solution as described by this patent.

Unlike many on slashdot, I actually believe there are some scenarios where software/algorithm patents are applicable. However, the standard questions still need to be asked: does this do something useful, and is the implementation non-obvious? Why (aside from purely financial reasons) are patents like this being granted?

RTFP

If you read the patent, you will see that TALK and other prior chat systems are mentioned in the references and body of the patent. The specific "innovation" here is that the system polls for activity on a timer, and turns on and off the "user typing" message based on activity during the timer period.

While I think that it is absurd that this was granted, it is not any of the things being thrown around on /. as prior art. Even Yahoo's "user is typing" simply toggles on and never turns off if you abandon typing. Is polling periodically obvious? Surely. Remember, the USPO is a profit center, and granting obvious patents brings profit to both them and patent attorneys, so there is no motivation not to allow such simple changes to be patented.

Excuse me

There's a problem here. I have a patent on hypocrisy. It's obvious that no. 6,631,412 is a derivative work and therefore infringing upon my IP.

Quake 3 is prior art...

With those blue 'bubbles' that appear over the player's head when he starts typing, and disappear when complete.

A request for reexamination

an excerpt from here [usip.com]

A request for reexamination is commenced by filing a reexamination request along with a modest filing fee. In the request, the requestor cites the patents and other printed publications which purport to establish that the patented inven- tion is not new or unobvious as of the date of its invention. The Patent Office will then decide if the requestor has made out a prima facie case of invalidity. If so, the patent will be subjected to reexamination. Reexamination is between the patentee and the Patent Office. The requestor has no involvement after filing the request for reexamination.

if you have any interest at all on the workings of the us patent system, go here [uspto.gov], read up.

The fee for "requesting an reexamination was 2520.00 in 1999.

Perhaps we should start a fund to have this patent reexamined?

It's the weight game.

Very simple, Microsoft gets patents on existing functionality or technology. If you use them, they go after you. Even if there is prior art, Microsoft can tie it up in court. Since Microsoft has a huge advantage in the money department, they can last longer, causing the victim to surrender or agree to some crazy license from Microsoft. It's basically being a bully.

It's smart business, however if we depended on "smart business" we'd probable be still living in caves after someone patented "shelters."

This exact feature was in wide use at MIT in '89

This exact feature was in wide use at MIT in '89 if not earlier - the zephyr instant messaging system used by nearly all students at MIT when I was there ('89 to '94) had this feature, along with essentially every other feature currently use in IM clients. This is BS. I'm not sure if zephyr is still in use at MIT, but this is certainly NOT something new.

What is "prior art"?

IANAL, but I thought "prior art" meant publicly explained methodology to do something prior to the filing of the patent at hand. It doesn't matter if ICQ or whatever else program had this capability, the fact that they didn't make their methodology public (which is usually done through an RFC-like document or by filing a patent) makes them unfit to be prior art.

But then again, I may be pulling this out of my ass.

-bm

One Way to Have a Positive Attitude About This

Is to look at it as a necessary evolutionary step. The patent system is broken, in fact the whole IP system is broken. It isn't going to be fixed smoothly and painlessly. It's going to be ripped out by the roots and replaced. But to make that process happen the system has to reach a breaking point in the public's tolerance.

The public is historically slow to act, and is never good at acting on obscure issues, as is the IP world for the most part. Some good things the file-sharing debacle has done are to educate a lot of ordinary people about intellectual property, to demonstrate their willingness to ignore IP laws they don't agree with, and to give people some actual experience breaking those laws and getting away with it. This is surprising and encouraging behavior for an American public that has successfully been dumbed down and convenience-addicted to the point of virtual sheephood.

But it's going to take a lot more pain to get people's butts off their comfy couches in the IP arena, to the point where politicians find their constituents threatening enough to start representing them again. That point is years away, and I want to live through it and into the next Golden Age. So for me, anything that pushes this process along is a good thing, in its own way.