Analyzing AT&T's Anti-Anti-Spam Patent

An anonymous reader writes "Dan Gillmor is reporting in his eJournal taken, in turn, from Gregory Aharonian: AT&T has apparently been awarded a patent for circumventing certain spam filters, thereby providing slimeball spammers with yet a bigger hammer!" The patent covers "A system and method for circumventing schemes that use duplication detection to detect and block unsolicited e-mail (spam.)", although it's unclear exactly what AT&T want it for.

Comments

Hey! Shortsighted people!

[KFury]

Has it occured to anyone that by patenting an anti-anti-spam technique, AT&T can legally forbid spammers from using that technique?'

Yay AT&T. I applaud you.

Re:Hey! Shortsighted people!

[KrispyKringle]

See, that occurred to me. But I sorta doubt they'll use it to track down spammers and sue them for patent infringement, considering that spammers are already very often violating state laws, violating their ISP AUP, and peddling illegal scams and therefore make themselves hard to find.

But on the other hand, I doubt ATT will be selling circumvention technology. Now, a fair guess would be that they won't sue the spammers for infringement, but may sue those who sell software used for spamming (who are generally a bit more findable).

Re:Hey! Shortsighted people!

[m_chan]

Or it occurred to them that they can make a mint by selling/licensing the technology to "spammers" or slightly more legitimate advertisers. It's probably just perception, but I think that a good chunk of the dinner-time phone-spam, and a large portion of the direct mail I used to get was from the Death Star.. oops.. I mean good ole Ma' Bell.

Re:Hey! Shortsighted people!

[incom]

Please someone with some money, patent all possible future DRM techniques.

Re:Hey! Shortsighted people!

[Steve B]

Has it occured to anyone that by patenting an anti-anti-spam technique, AT&T can legally forbid spammers from using that technique?

True, though it's unfortunate that the government hasn't already done so on the grounds that circumventing an anti-spam filter is a form of cracking.

Re:Hey! Shortsighted people!

[GreyWolf3000]

Hold your applause until they demonstrate that their intention is good.

Re:Hey! Shortsighted people!

[Shakrai]

spammers are already very often violating state laws, violating their ISP AUP, and peddling illegal scams and therefore make themselves hard to find

Hey, I hope this doesn't get me modded flamebait but I've had this thought for awhile and this seems like the ideal article to raise it in. Disclaimer: I am not endorsing or defending SPAM or the people behind it.

Has anyone else thought that the most effective way to combat SPAM would be with education not filters/lawsuits/etc?

It would seem logical to me to assume that at least a large number of (if not a vast majority of?) spammers are ignorant as to why it's a bad idea. They don't know much about the Internet, and some idiot with a spam-software outfit approaches them and tells them about this "Great Marketing Idea", sells them some software (that may or may not do various bad things like hiding headers/etc), and off they go!

My boss approached me once with some literature he received from one of these software companies. After my initial "WTF??? You aren't serious???" reaction I sat down with him and explained some of the history behind spamming, why it's a bad idea, would piss off our existing customers/alienate new ones, etc etc etc. Based on this experience it would seem to me that the most logical solution would be to educate the companies behind the spamming as to why it's a "Bad Idea".

Of course, this theory doesn't hold any water when you look at pornographic spam, Nigerian bank fraud spam (my personal favorite), pyramid schemes, etc etc. But it probably would be a better approach when dealing with the idiots who have been duped into thinking that unsolicited e-mail is a legitimate marketing tool. At the very least it can't hurt any.

Just a thought I've had for awhile now.

Re:Hey! Shortsighted people!

[MillionthMonkey]

This patent describes the simple use of hash-busting characters in email messages.

System and method for counteracting message filtering

Abstract

A system and method for circumventing schemes that use duplication detection to detect and block unsolicited e-mail (spam.) An address on a list is assigned to one of m sublists, where m is an integer that is greater than one. A set of m different messages are created. A different message from the set of m different messages is sent to the addresses on each sublist. In this way, spam countermeasures based upon duplicate detection schemes are foiled.

This isn't "providing slimeball spammers with yet a bigger hammer". It's a bread-and-butter spamming technique. Almost all the spam I get is salted with random letters or dictionary words in the address or message body to change the hash (and is therefore infringing on AT&T's new patent). We just saw a story a few days ago where spammers were sprinkling fraudulent scam emails with hash-busting characters to get past filters.

One of the nice things about spammers is that (unlike their opponents) they rarely patent the circumvention mechanisms they use, leaving their bag of tricks open for intellectual property land grabs like this one. Compared to laws against spam, which for the most part hardly exist, patent law rests on sound international footing and gives AT&T much greater leverage against spammers who are now patent infringers. Good for AT&T. I wish I'd thought of it first.

It's lunacy to assume that AT&T secured this patent for any other reason- like productizing this stupid patent. Are they going to sell a new software suite for spamming? Spammers aren't an ideal software market by any reasonable standard. There's only 180 of them. AT&T would sell one copy, it would get pirated 179 times, everyone with a copy would start spamming warez versions of it, and that would be the end of it. Assuming that spammers cared about using patent-encumbered software at all- which they don't. And AT&T would alienate its customers in all the other markets they're in. It would be like a Christian bookstore opening a bondage videos section. It makes no sense. I can't understand how anyone could possibly take the outrage in this article at face value.

What is really amazing about this patent is what it says about the research done by the USPTO. I bet the USPTO examiner received a dozen examples of prior art in his own inbox the very day he approved this patent, and he approved it anyway!

Re:Hey! Shortsighted people!

[Gherald]

> International patent law, however, is another matter.

The matter being that unless sizeable amounts of money are involved, nothing gets enforced.

Re:Hey! Shortsighted people!

[Trepalium]

Most of the people who could be educated aren't the ones that end up putting most of the e-mail out there. These people often only send out a single mailing and then get quickly educated as the complaints, DoS attacks, and ISP services terminations arrive. No, most of the big spammers don't care that it's inconvenient. I've been getting 50 worm.swen.a e-mails per day, and I'm pretty sure every virus and worm writer out there knows it's wrong to make these things.

Personally, I agree with the other people who've replied. Educating the people to not respond to spam (or con artists in general) would be a more worthy cause. Far too many people get taken by both internet, and the more traditional phone scams. The fact that for many people, their brains shut down when they get within 10 feet of a computer has made internet scams pretty bad.

PRECISELY!

[Chas]

Now, instead of being well-nigh untouchable due to spam's precarious placement as little more than a highly undesireable activity, AT&T can go after spammers IN COURT on grounds of PATENT INFRINGEMENT.

And going to court over something like this takes megabucks. Especially against a company the size of AT&T. Even if the spammers somehow weasel out on technicalities (like they didn't actually infringe on the patent directly), they're still going to be out so much money that their great grandkids aren't even going to be able to go to any educational institution after public high school.

Wait a minute ...

[obsidianpreacher]

1) Patents are a way of restricting rights to certain ideas/methods/etc.
2) AT&T can prevent anyone else from circumventing anti-spam filtering software with this patent
3) Ergo, AT&T are the good guys

...

wait a minute, I thought they were the bad guys

...

I'm confused now ...

Re:Wait a minute ...

[Corydon76]

The patent is not for sending multiple messages with the same text. It's for altering messages to fool hashes. I hadn't seen that technique employed by spammers until at least 2001 or 2002.

Then again, I suppose I'm lucky that I block only 200 spam messages a day, with only about 5 getting through.

Re:Maybe AT&T is just disorganized

[Phroggy]

I can't imagine that AT&T would sell spam technology, because it would be a public relations nightmare.

You don't think they'd sell it under the "AT&T" brand name, do you?

Several distinct companies operate under the AT&T brand name; I'm sure AT&T owns several companies that operate under different brand names as well.

How many normal people do you suppose make a connection between Bugs Bunny, WinAmp, Mapquest and CNN? They wouldn't make the connection between AT&T and whatever subsidiary sold the spam software either.

I'll chime in on the anti-patent side

[rsilvergun]

Since slashdotters seem to hate spam so much all reason gets abandoned when it's involved, I thought I'd point out why this is so awful. Basically, it's a math algorythm. Like Quick Sort. Now stop and think about what computing would be like if Quick Sort was patented. The same sytem that allows this to be patented would also allow Quick Sort to be. We're fortunate that most of the ground work for computing was layed before this mess started. Anyways, I just wanted to make the point that there's no such thing as a good software patent.

ATT will be selling circumvention

[djupedal]

1980...
Remember being charged for an unlisted number?

1990...
AT&T sells us caller-id, and then sells caller-id avoidance devices to marketeers, then sells us next-gen caller id to thwart their devices...etc...etc.

AT&T has been playing the middle for years...I see no reason for them to stop now. Patents just mean more money, faster.

Re:ATT will be selling circumvention

[Anonymous Coward]

So, how did AT&T manage to sell anyone caller-id in 1990 when local service was split off during the 1980's and they no longer were supplying local service? And how do comments that contain rubbish that a simple Google search can disprove get modded to "5 Insightful"?

probably just a fluke

[martin-boundary]

Actually, this whole thing is probably blown out of proportion. The patent summary looks a lot like a paper by Robert J. Hall. I expect that ATT has a policy of patenting everything any of their researchers works on, regardless of what it is. The paper itself is mainly mathematics with the spam theme thrown in to make it interesting.

The reason why...

[toupsie]

The patent covers "A system and method for circumventing schemes that use duplication detection to detect and block unsolicited e-mail (spam.)", although it's unclear exactly what AT&T want it for.

If they cannot call you to get you to change your long distance service, maybe they are doing to "telemarket" to your inbox. The Federal 'Do Not Call List' is changing the way a lot of traditional telemarketers are doing their business. Since they are now being fined for calling you, they need another way to invade your life and bombard you with offers. Having a technology that can circumvent spam blocking would be a step up on the competition.

Why, and WTF?

[lwsimon]

By patenting this "technology", they are showing the weaknesses of current spam filters. Maybe that's what they intended all along... And now that i think about it, wouldn't this come under the heading of a software patent? I mean, its not code, but its an algorithm, right?

Alternatives to Quicksort

[yerricde]

Now stop and think about what computing would be like if Quick Sort was patented.

Easy. C's qsort() would heapsort instead, as it in fact does on some C library implementations such as Metrowerks CodeWarrior's. If heapsort were patented as well, qsort() would merge-sort on large-memory machines and Shell sort on small-memory machines. If more of the efficient sort algorithms were patented, programs would be designed to manipulate data in search trees instead of arrays. There exist several sorting algorithms; unlike patented file formats such as GIF and MP3, these have minimal to no interoperability disadvantages. You're going to need a broader example than that.

Re:Maybe AT&T is just disorganized

[KjetilK]

I see at as just another research project. The people have researched how to foil spam filters. Why? For the same reason I just ran nessus against my server, to better guard against attacks. Finding a way to foil spam filters is just as legitimate as writing something like nessus, sure it can be used to break in, but it is important if you want to improve system security.

What we should do now is to read the patent, understand where the weaknesses are, and improve the filters now, before the spammers start using it (OK, from /.ers reaction, it seems like they just patented adding random rubbish, not exactly new, but did those /.ers actually perform a analysis of the patent?)

It's harder to explain why they patented it rather than just published. Probably, they have some sort of incentive program: Employees get a bonus for patents. It is quite common. You get the bonus regardless of whether it is useful or not, it is the size of the company's patent portefolio that counts on the stock market. So, they just patented it.

It is not because they are disorganized, evil spammers, or not even because they plan to go after spammers, it is simply because the research was done, and the incentives in the organization says "patent, don't publish".

What's good for the goose

[dacarr]

Remember, everyone on the 'net is affected negatively by spam - including the Death Star themselves. As such, if they patent this, they make it harder for spammers to deal with circumventing filters.

That AT&T came out and did this, frankly, rocks. Good show, guys.

The only concern I have is that there is prior art, which will come up as a double-edge sword again. Prior art will protect the good guys from frivolous patent filings (Amazon, anybody?), but as such I'm concerned that the spammers will pull the prior art card against AT&T. On the other hand, AT&T's interest - protecting their network - and the fact that they probably have infinitely larger amounts of money than your spammers just might put an end to them for now.