How EPA works
EPAmail User's Manual
E-Mail News & Comments
E-MAIL MANAGEMENT NEWS & COMMENTARY ... March 26, 2010
... tell me what YOU think!
"Insanity: Doing the same thing over and over again and expecting
different results." --- Albert Einstein
"I'm no Einstein, but shouldn't it be spelled 'spamn' so it rhymes with 'damn'? " --- Editor
WHITE-LIST vs BLACK-LIST? --- IT's 100% RIGHT vs 99.9999999999999% WRONG
An Intellectual Property attorney reports that a USPTO Examiner with whom he has contact sees
no differences between white-list e-mail filtering and black-list e-mail filtering. For the
general population (which has minimal or no mathematics education or Software Engineering
experience), that's understandable and OK. ___ But for someone whose business is to understand and evaluate
software inventions, failure to understand is a primary indication of not having done the analysis.
And that's DEFINITELY NOT OK. ___ As an aid to that Examiner and those of you who would like to understand,
I've created a detailed analysis of the two filtering approaches. ___ I sincerely hope all
who read it find the conclusions useful in their Software Engineering efforts. ___ And, if I"m
wrong, please tell me what you think.
HAPPY NEW YEAR!! - Let's resolve to make the Patent Office effective!
The United Stetes Patent and Trade Office (USPTO) controls the flow of new intellectual property production. A major portion
of our national GNP is generated by the sales and licensing such properties. Overwhelmingly, new properties are developed in the
areas of computer hardware and software applications. USPTO' S IN DEEP TROUBLE AND NEEDS MODERNIZATION ___ 1) Each year for the last
decade, the USPTO has fallen further and further behind in it's ability to perform its' assigned responsibilities. It once took two years or less to fully assess and decide
the patent application. IT NOW TAKES TWO YEARS JUST TO ASSIGN AN EXAMINER FOR THE APPLICATION. Every year of delay has an
enormous economic opportunity cost. ___ 2) The USPTO does not understand concept of an invention created using computer sofware.
The USPTO is accustomed to employing experts in well-defined fields of knowledge (like Aeronautical Engineering and Pharmaceutical
Manufacturing) and training them to be patent attorney/examiners. What the USPTO hasn't grasped is that 'COMPUTER SOFTWARE IS A
LANGUAGE'. That language is used to describe inventions in applications which are themselves highly specialized and wildly diverse.
___ 3) Because the USPTO doesn't understand 'software", they recruit and hire the wrong people to examine inventions based in
computer software. Their job requirements for such positions emphasize educational credentials in Computer Science. But very few
patent applications involve computer science inventions. Instead, they address manufacturing technologically-advanced products for
a tightly-focussed user group with their own application-specific vocabulary (for example, 'doctors using CAT scans'). So, THERE'S
AN EXTREMELY HIGH PROBABILITY THAT A USPTO SOFTWARE EXAMINER HAS NEITHER THE APPLICATION KNOWLEDGE NOR THE IDIOMATIC VOCABULARY
TO UNDERSTAND AN ASSIGNED PATENT APPLICATION. ___ 4) Because of the size of the backlog of applications, the USPTO is desperate to recruit
personnel. As usual when demand exceeds supply, what's obtained is lower in quality than otherwise. The USPTO has chosen to
recruit from overseas populations for whom English is - at best - a second langauge. Thus, A PATENT APPLICATION'S EXPOSITION LANGUAGE
AND INVENTION-KNOWLEDGE-AREA-SPECIFIC LANGUAGE ARE FOREIGN TO OR INCOMPREHENSIBLE BY THE EXAMINER. ___ 5) The USPTO's
facilities are so inadequate that a goodly number of its' examiners work in dispersed, isolated locales where their ability to
tap a colleague's knowledge and experience is drastically inhibited or non-existent. Also, dispersed locales make proper
supervision difficult to impossible. ___ 6) Finally, the USPTO's examiners knowledge, experience and work product cannot be
challenged and appealed by an applicant ('s attorney) until the entire lengthy process of examination is completed. In general,
the examiner's capability and capacity to examine an application will be revealed by a) Office Actions written by the examiner
to an applicant's attorney; and b) discussions between the examiner and the applicant's attorney. DISCONNECTS IN LOGIC,
KNOWLEDGE AND LANGUAGE BECOME QUICKLY APPARENT. But the applicant has no recourse except to wait until an appeal can be initiated.
Thus, an already lengthy process becomes further unproductive, ineffective and costly - for the applicants and the GNP. ___
Without action at the highest level, there's nowhere this situation can go but downhill. LET'S APPLY SOME LEADERSHIP AND
FIX THIS MESS!!
ICANN, Natural Language CodeSets and You?
ICANN's recent decisions permit 107,000+ unique lanuage-specific character representations ("glyphs") to be employed in Internet traffic.
Glyphs are organized into "codesets" - named for the language they represent - which permit publication of writings in
the language. ___ The smallest codeset is US English which is the 128-glyph ASCII-standard. This codeset is the foundation of and an embedded
reserved subset of every other set because its' glyphs are used to describe the Internet and manipulate its' resources. ___ For instance,
the comnponents of an e-mail are identified using the ASCII set regardless of any other natural language present in the e-mail. The e-mail's
Header/'envelope' contains a 'Content-Type' field that identifies the codeset used in preparing the e-mail. ___ Regarldless of the codeset employed,
there is no restriction placed on intermingling ASCII glyphs with the natural language glyphs of the set in any e-mail field. Even 'TO:'
fields can contain intermingled glyphs as long as the embedded Domain Names can be resolved into valid destintaion IP addresses. ___ Domains whose
Name is composed solely or partly of the natural language glyphs can be accessed only if the relevant codeset is available for use. This means that
e-mait client software must make the appropriate codeset available for use for creating and reading e-mail.___ Spam detection methodologies which
depend on black-lists and linguistic analysis will use more system resources & processing time (becoming more expensive); and, be less effective due to intermingled glyph
usages. ___ EPA's white-list approach is insensitive to intermingled glyphs.
ICANN Approves Domain Names We Can't Type
Lo and behold! ___ Literally, 'just as I finished' writing the commentary below, the referenced article appeared
at 22:30 EDT, October 30, 2009. ___ THE FUTURE IS NOW! ___ Your preparation time has gone from a comfortable time in the future to ZERO. ___ And all of the 'black-list'
criteria and methodologies you've been using have become even more less useful to you than they were! ___ With "black-lists, every 'foreign font' e-mail will be allowed to be delivered
to the addressed mailbox. ___ Your only practical solution is to stop trying to define and exclude e-mail you DON'T want and start defining what you DO want ___ It's time you got EPA to pull your fat out of the fire. ___ Call Ralph Seifert at 561-762-7685 during US EDT business hours or leave a message after hours.
Massive, Basic Changes in E-mail Content Are Fast-Approaching - Is Your Domain Ready For Them?
*** 2009 E-MAIL ___
247+ Billion e-mails per day (= 2.9 Million per second; 74% individual; 24% organizational, 85% 'spam') ___
1.4 Billion Mailboxes ___ 184 million Domains ___ 21 Top Level Domains ('.com', '.edu', '.gov', ...)
Standard URL font: U.S. English glyphs(glyphs); Prevalent text font: U.S. English glyphs
*** 2015 E-MAIL
Total e-mails per day, the number of mailboxes and the number of domains will grow simply because the Internet user population grows.
The number of top leve domains will grow dramatically as ICANN permits (sizeable) special interest groups and nations to define them.
Language-specific font glyph sets - such as Chinese, Cyrillic, Hebrew, etc - in both URL's and text fields will proliferate to accomodate large net communities.
*** What to do?
Ensure that your e-mail mangement tools function properly when new fonts appear in URL's and text fields.
Tools which use 'black-list' criteria based on U.S. English semantics, idiom, context and display format will need to be modified to remain as (in)effective
as they are now. Better yet, switch to e-mail management tools which use 'white-list', 'I-want-exactly-this-and-this-only' criteria(like EPA) ___ Then, the user of the glyph set
can state exact criteria in the natural language represented by the glyph set.
Reduce the amount of time and resources spent processing 'spam' by identifying the mail as soon as the mail's Envelope is received, Then 'burn' (ignore/discard) the rest of
the mail.___ EPA accurately classifies incoming mail based on the mailbox users definitions of criteria that are compared to mail Envelope fields.
GET AND INSTALL EPA - NOW! Why delay achieving the highest-quality results from a superior, long-lasting tool?
Cyberwar? What's That?
You've probably noticed an increased mumber of articles, books and commentary focused on "cyberwar"
In this war, "cyber espionage" is (mistakenly) defined to be the mildest method of attack. But this "mild method" can be easily employed
by a wide variety of agents to achieve their goals at your expense. ___ YOU make it easy for them to succeed by failing to control your
domain's e-mail operations. Uncontrolled e-mail provides "the enemy" with easy access to your information and personnel. Uncontrolled e-mail
allows unauthorized access to internal information and correspondence. Uncontrolled e-mail allows internal information to be transmitted to
any Internet destination. ___ And who is "the enemy"? It's current, retired and fired employees. It's vendors and customers. It's attorneys and
altruists. It's anybody with motivation and rudiemntary e-mail skills. ___ The enemy will exploit your vulnerabilities as long and as often as you
permit them by failing to control ALL of your e-mail operations. ___ Smokey the bear says "Only YOU can prevent forest fires!" I say "Only YOU can prevent getting
burned by e-mail espionage".
U.S. Is Losing Global Cyberwar, Commission Says
And no wonder! E-mail transmission - BOTH domain-external AND domain-internal - is totally uncontrolled in almost every organizational domain. ___ So, for governmental,
business and other organizational domains, internal information can both leak out and be routed internally to those without 'need to know'. ___ Want to send
OUR budget, OUR product development plans and/or OUR legal case strategy to some outsider for profit or revenge? Just send an e-mail! ___ Big volume of documents?
Just attach all the files you want to an e-mail and send it! ___ Got a crush on SBJ over in Accounting? Send a couple of e-mails loaded with sexual innuendo! Who will know?
Certainly the lawyers when SBJ files a suit against the organization for permitting sexual harassment in the workplace! ___ Know someone who wants to buy the plans that
your group developed to build the latest in suitcase atomic bombs? Arrange for payment and send an e-mail! ___ I could go on and on but I'm sure you get the point.
Without the kind of filtering and control EPA provides, your e-mail domain is a disaster waiting to happen. ___ It's avoidable and preventable. Eliminate the potential
for this class of security problems once and for all! ___ Letting it happen on your watch is definitely not a career enhancer - more like a career ender - in today's realities.
___ Get EPA installed!!!
glyph Combinations Foil Blacklisting
When you use blacklists to identify e-mail senders or text you DON'T WANT, you're fighting a losing battle. All the spammer has to do is avoid using any
of the glyph strings you've listed and the e-mail will go directly to your mailbox. ___ And how easy is it to avoid your listed strings? Well, if you
permit 'M' different glyphs to appear in a string that's 'N' chazracters long, there are 'M' raised to the power 'N' possible combinations. ___ As an easy proof,
a string 2 glyphs long composed of 10 different glyphs - say 'O', '1', '2', '3', '4', '5', '6', '7', '8', and '9' - has 100 possible combinations
(100 = 10 squared). The same glyphs in a string 3 glyphs long has 1000 possible combinations (1000 = 10 cubed). ___ Then a font of 256 different
glyphs used to create a blacklist entry just 10 glyphs in length has '256 raised to the power 10' possible combinations (= a little over '1.2 followed
by 24 zeroes'). (To check this or other possibilities, go to Google and enter '256 ** 10') ___ And, any sender's address or text composed of more than 10 glyphs
can't possibly match the blacklist entry! ___ Is it any surprise that blacklists are a lot of effort for very ineffective results? Is it any surprise that\
whitelists which identify mail you DO WANT are totally effective?
Other Fonts Mean More Spam in Your Mailbox
By now you might have reeceived e-mails whose Subject: is expressed using the Russian/Cyrillic font. Of course you got that mail! Using a 'blacklist' of words asnd phrases
to identify unwanted mail and stop delivery will be totally ineffective when the blacklist's font is not identical to Subject:'s font. Since two different
fonts won't match, all such mail is 'wanted/acceptable' and is delivered to your mailbox. ___ EPA employs only 'whitelist' filters to identify 'wanted' mail
and permit delivery. Consequently, the presence of different fonts has no effect on the accuracy of the filtering process when criteria and e-mail both employ the same
font. ___ Moreover, when criteria and e-mail fonts are different, the e-mail can never by classified as 'wanted/acceptable'. ___Look at it another way. When 'blacklists'
are employed, every active spammer is generating entries which you must add to your filter if it is to accurately identify all 'unwanted' mail. With a 'whitelist'
filter, only you can add entries. ___ Why would you allow a huge group of strangers to create additional daily work for you?
Losing the Battle Against Spam
From the article: "Anti-spam software uses content filters to block messages that contain telling words ...". Again, no one seems to question this methodology of
'stating what one DOESN'T want'. They are satisfied that the method is correct - DESPITE the fact that in every other 'real life' situation where one is offered
choices, the selection is generally starts with 'I want ...'. ___ It's clear that 'spam' is UNwanted e-mail. And a modest amount of reflection will let one come
quickly to the conclusion that it's flat-out impossible to define everything one doesn't want. ___ So why not start by defining what is WANTED? It's a lot more
feasible. ___ Current filtering software delivers mail to your mailbox WHENEVER the DON'T WANT filtering criteria DOESN'T exactly match e-mail content. EPA
delivers mail to your mailbox ONLY when the DO WANT filtering criteria DOES exactly match e-mail content. Again, a modest amount of reflection will let one quickly
realize that DON'T/DOESN'T and DO/DOES are not inverse operations of each other. (Here's an example of how they aren't) ___ Even if someone 'harvests'
your e-mail address, there's insignificant-to-zero probability that their e-mail content will exactly match your EPA filtering criteria. There's an enormously higher
probability that e-mail content won't exactly match your 'telling words' criteria. ___ So don't give up and say nothing can be done about spam. EPA is THE solution. The
only problem is that it hasn't yet come to the attention of the right persons. You can help solve the problem by telling the right person about EPA.
REPORTED BUT AS YET UNVERIFIED NEWS:
Court Rules Company Has No Right to Inspect Content of Employee's Office Computer
And practically speaking, no organizaton wants to spend the time and money to police what employees keep on their office
computger. The problems occur when the employee takes the computer from the office or transmits their computer's content
to somewhere out of the organization. ___ Every organization can keep employees from removing tangible organizational assets from the
workplace. What the company can't currently control is - as the item immediately below addresses - transmittal of company
information via e-mail to some other venue. ___ Again - it's not "CAN'T", it's "WON'T".
Firms Fret as Office E-Mail Jumps Security Walls
This 1/11/2007 New York Times headline might just as well read 'Corporate Management Can't Control Their Employees' E-mail Transmissions'.
And the sad fact is, it's not "Can't", it's "WON'T"!!! ___ For years, organizations have made Internet browsing("surfing") an employee
'perk' - in effect, absorbing the resulting lost productivity as just one more overhead cost of doing business. ___ But, there's an ENORMOUS difference
between surfing and e-mailing. Surfing results in transient images on a screen that are viewed by a surfer; then, quickly disappear without organizational impact.
E-mailing exposes the organization to 1.) All forms of destructive 'malware' that can be piggy-backed onto or embedded in e-mail being received; AND, 2.) As addressed by the article, uncontrolled transmission of
organization-proprietary information. So WHY should organizational e-mail access be the same kind of 'perk'??? ___ Given the inability/incompetence
of all e-mail software products to precisely control e-mail receipt, it's no great surprise that e-mail transmission is not controlled. But - again no surprise -
EPA precisely controls both! HOW?___ What does it take to get organizational management to act and
demand the best from their suppliers???? Leakage of... TOP SECRET information from a US Federal department? ...IBM's 1st quarter financial results? ...Google's
5-year product plans? A class-action shareholder suit for mismanagement spearheaded by Bob Montgomery (of class action against 'big-tobacco' fame) or Denny Crane
(of 'Boston Legal' fame)? ___ The required solution is here and available. If you're liable for it, it's your responsibility to demand it! ___ Whining is NEITHER an option NOR a defense.
The Fight Against V1@gra (and Other Spam)
Here's an e-mail filtering company that performs pre-server filtering of its' clients' e-mail traffic stream using a combination
of human researchers and automated filtering to detect and delete spam. Since "spam" is semantically equivalent to "unwanted e-mail",
they are defining spam as "what WE don't want". In short, they are another entry in the continual
Red Queen's race to precisely define ALL spam. ___ They state "there are some indications that the growth rate of spam has plateaued
or even slowed" without describing the indications. ___ Too, "If individual users also have personal spam filters installed on their
computers, their in-box spam count can be reduced to a trickle". But their service does not offer EPA's mailbox-personalized control
(which reduces the spam count to ZERO). ___ Last quote, "We're just not going to kid ourselves and say we believe that spam is ever going
to go away. It's always going to be a prob-lem." And they're right as long as they continue to use the wrong - non-EPA - approach to
management and control. ___ It would seem that if the company is truly effective, then they are another Blue Security waiting to be
attacked. ___ Finally, as a company shareholder, I'd regard it as a serious failing if my company contracted
control of the company e-mail traffic to a third party. I can't begin to imagine the legal and business ramifications of mishandled
e-mail. Such a contract is an admission by management that they cannot control company operations.
Under Attack, Spam Fighter Folds
Blue Security, an Israel-based company, was effective only until a "distributed denial-of-service attack" launched by spammers
shut down the company's servers.___ The lesson to be learned here? ANY WEB ADDRESS SUCCESSFULLY ATTACKED THE SAME WAY! ___
How to avoid the problem? Perform domain filtering WITHIN the individual domain. ___ In the Blue Security case, these was a
compelling ecomonic reason for the spammers to mount an attack. Where a domain is performing it's own filtering internally,
there are no external indicators to attract spammers's attention and invite an attack.
THERE'S GOOD NEWS AND BAD NEWS:: Noted today at AOL!
First, the good news! On its' main 'Reference' screen, AOL equates 'spam' and 'unwanted mail'. I find it rewarding to see EPA's
definition of 'spam' being used by at least one major ISP. ___ Now the bad news! AOL's ability to correctly separate the
'spam/unwanted' from 'not-spam/wanted' shows no improvement. ___ Does anyone have a suggestion for AOL? ___ Do you think AOL
is handicapped by the NIH(Not Invented Here) mentality so prevalent at companies with large, in-house software development
staffs?? ___ Can anyone bring EPA to AOL's attention???
PROPOSED 2006 RESOLUTION FOR ALL E-MAIL PROFESSIONALS: Provide e-mail USERS with total, accurate, effective e-mail control!
In 2005, as in previous years:  Internet and intranet e-mail volume continued to grow  Unwanted mail continued to flow into
users' mailboxes  Organizations continued to allow unauthorized e-mail to flow into, within and out of their domains  No
major ISP or domain owner had even announced an effective solution to control unwanted and unauthorized e-mail (forget installed
and in use) ___ Many reports of 'victory' and 'success' appeared in the media. In each case, the report brought to mind Benjamin
Disraeli's statement "There are three kinds of lies: lies, damned lies, and statistics". In summary, a lot of 'feel good'
self-serving propaganda unsupported by reality. (See commentary immediately below about the AOL and FTC articles) ___ Complete control
of all e-mail is possible only if e-mail software professionals will  Admit that the various existing piecemeal platform-dependent
software solutions are ineffective; and  Replace them with a uniform, platform-independent, totally-effective solution. ___ All
that's needed is a major organization with enough guts to "tell the emperor he's naked" and the conviction that there is a different
and better way. ___ The EPA process IS the solution. Where's that gutsy organization???
AOL Releases Top 10 Spam List
The major portion of this article simply focuses on identifying the 10 most prevalent spam e-mail Subject Field contents
detected by AOL. That's certainly neither 'news' nor useful to AOL subscribers. The two concluding article paragraphs sets
forth two claims by AOL based on specious statistics. ___ AOL first claims to have reduced spam by 75% since 2003 based on
the number of customer complaints. That's like the US Labor Department defining the unemployment rate by excluding those who
have stopped seeking employment. 'Have stopped' implies those excluded had previously been seeking employment. Why did they
stop? If there aren't any jobs to be had, why should they look? If I complain to AOL and don't get satifactory resolution, why
complain again? ___ Also, there's the question of the quality of the underlying data on which the claim is based. If we're talking
about reducing 100 complaints in 2003 to 25 complaints in 2005, the figures are insignificant in the context of AOL's 32+ million
subscribers. ___ AOL also states it blocked an average of 1.5 billion spam e-mails per day and that this represents 80% of the traffic
reaching their gateway daily. By direct calculation, that means that 375 million non-blocked messages are delivered or about 10 per customer
per day. Doesn't sound too shabby does it? ___ However, AOL's website states that their spam detection mechanism is unreliable. AOL
advises you to check your spam folder at a frequency that increases as the level of spam filtering you choose increases. Why? Because
some e-mails you want to receive may be incorrectly classified as 'spam'. Hence, in the long run, if you want to make sure you're not
missing 'good' mail, you've got to check all arriving e-mail in both your in-box and the spam folder. ___ Also, if the delivered mail
classification is unreliable, there's a high probability that the classification of blocked mail is also unreliable. ___
My (reasonable) conclusion is that AOL's e-mail service unreliable and doesn't decrease the required user effort at all. Wouldn't
you agree? Don't you want something better???
Spam filters thwart junk mail menace: Incomplete study declares victory over spammers
A US government agency - the Federal Trade Commission (FTC) - provides us with a counterintuitive definition of 'victory' (to
go with our industry-standard counterintuitive method of filtering 'spam'). For the FTC, 'victory' means that '2 (out of how many?) ISP's achieved spam-filtering effectiveness of
95% (at best) and 78% (at worst)'. On the same basis, if 100 biological-weapon-carrying terrorists are loose in a major city,
just catch 78 or 95 of them and ignore the rest. Voila! Victory! ___ Also, assessing a software filter's effectiveness at less than
100% is an explicit acknowledgement that the assessor's definition of spam differs from the filter's. And, as long as two definitions
exist, the software can never be 100% effective! ___ However, if the assessor's definition is applied by the software filter, then the
filter will always be 100% effective. The primary question in this solution is 'how much time and effort must the assessor expend to
completely express their test criteria?'. ___ And the answer depends on the whether the filter's test criteria determine 'what is spam' or
'what is not spam'. As demonstrated by the operation of every widely-distributed filter, defining 'what is spam' criteria is a never-ending, high-cost task because each arriving e-mail has a
high probability of requiring a criteria addition. Clearly, defining 'what is not spam' is a finite, low-cost task because
each arriving e-mail has a low probability of requiring a criteria change. ___ So, why spend all the effort defining a filter
whose effectiveness depends on 'what is spam' criteria? You already know that your best effort will be much less than 100%-effective!!
* * * With the EPA process, effectiveness is always 100%! And, you can do what you want with the e-mail that EPA
keeps out of your mailbox!
* * * (Incidentally, I thought that the article's author was extremely kind in his use of the word 'incomplete' rather than 'statistically
Spam Advertising Becoming More Effective?
Page excerpt "one-third of us have clicked on a link in a spam e-mail message ... and one in 10 have actually purchased products
advertised in junk e-mail, according to a survey conducted by the security firm Mirapoint and the market research company the Radicati
Group". That's 33%+ now reading and 10% now buying - UP from the 14% and 4%previously reported survey
results. ___ In short, over the last several months, spam advertising has more than doubled it's effectiveness in reaching
potential buyers and obtaining sales revenue from these buyers. With returns like those, why would any seller NOT increase the
volume of advertising??? ___ Also, the article defines buying a product advertised in spam as "bad email behaviour". ___ Because it's
obvious that the definition of "spam" varies with every e-mail user, any study's results can only be interpreted in the context of
the "spam" definition applied. If "spam" is defined as "unsolicited bulk e-mail", then the reading rate approaches 100%. ___ Only
by defining "spam" in terms of an individual's "wants" - desires, tastes and preferences - can meaningful statistics be obtained! And with
such definition, every e-mail user reads 100% of their "wanted"/"non-spam" e-mail. ___ The "approaches that routinely achieve 90% plus
(spam) catch rates" cited in the article are still ineffective for "10% minus" billions of pieces of unwanted e-mail. The EPA approach
which ALWAYS achieves a 100% catch rate has been operating WITHOUT ERROR since NOVEMBER 2002!!!
'Major USPO Delays in Processing Software Patent Applications Impacts the US Economy!
EPA's patent application was filed in November, 2002. Our attorney has been advised that applicants can expect a 18-24 month
delay in the normal examination schedule. Their reason? Lack of qualified examiners! ___ That's bad news for the nation and
for every US software developer (from Microsoft to you and me) ___ WHY? Various international intellectual property agreements have been
structured to depend on anticipated nation application processing durations. Agreement signatory nations patent offices are interlocked
and synchronized into a relatively seamless schedule that admits little deviation. This entire schedule is slaved to expected USPO
processes. USPO failure to meet this schedule forces applicants to gamble expensive fee amounts to enter (inter)national examinations
without USPO examination results or lose international patent opportunities. ___ US NATIONAL IMPACT: A large component of the US economy
and GNP is derived from intellectual properties. This USPO logjam affords all the bright software developers around the world
opportunities to obtain the patent protection that greatly enhances product development and market exploitation. ___ REMEDIES:
PERSONALLY - talk and write to your elected representatives! BIG SOFTWARE DEVELOPERS - hold back on filing 10,000 patents a
year that don't address substantive innovation until USPO catches up. EVERYBODY - spread the word to your professional peers
and get them involved. IT'S OUR PROFESSION AND LIVELIHOOD!!!
'Inaccurate Spam Classification Can Seriously Hurt You - And DOESN'T DECREASE YOUR WORKLOAD!
Does your ISP's or ESP's e-mail service offering include an 'anti-spam' feature? What does that mail service do with e-mails
that it classifies as 'spam'? ___ AOL®saves it in a separate folder. Netscape® puts it in the same box as non-spam
and add a symbol to inform you that they think it's 'spam'. ___ And explicitly(AOL®) or implicitly (Netscape®) advise
you to check the e-mail that's classified as 'spam' so you don't miss an important but wrongly classified e-mail! ___ Now,
ask yourself _ 1) "Has my ISP significantly reduced my workload resulting from 'spam'?"_ 2) "Can I trust my ISP to filter
mail for me reliably?" _ 3) "Do I have a good control of the criteria the ISP uses to classify an e-mail as 'spam'___ Did you answer
'No', 'No' and 'No'? ___ There's an enormous amount of time, energy and money spent to provide
you with a next-to-useless service - that can actually harm you if you assign any credibility to its' capabilities! ___ Where do you
think the money comes from? Right! You pay for it with money! AND YOUR TIME AND EFFORT! ___ You NEED is anti-spam software that
is ACCURATE, RELIABLE and PERSONALIZED to your wants and needs. ___ EPA addresses both your needs and your wants
with a 100%-accurate classification of all of your e-mail. ___ Let your ISP or ESP know you're 'mad as hell and not going to take it any
longer!' You want EPA working on your e-mail!!
'Deleting Spam Costs Billions, Study Finds'
Article says "Time wasted deleting junk e-mail costs American businesses nearly $22 billion a year".
Also says that 14% of recipients
read the mail and 4% bought something advertised in 'junk' e-mail! ___ That '14%' and '4%' mean:1.) A significant portion of users have
opposing definitions of 'trash/junk/spam' and 'treasure'; and, 2.) An unsolicited bulk e-mailing has a better response/success rate than
many/most other forms of mass advertising; and, 3.) There is a high, continuing economic incentive for a merchant to use bulk e-mailing. ___
All of which means that the probability that the future volume of such mailings will decrease (or even stay constant) is
infinitesimally-small-to-zero. ___ Because it's obvious that combinations of LAWS AND AGENCIES CAN'T/WON'T CONTROL WHAT'S
TRANSMITTED to the e-mail user's mailbox, the best any E-MAIL USER CAN/MUST achieve is to CONTROL WHAT'S DELIVERED!! ___ EPA
provides the e-mail user with 100% control of 100% of their e-mail.
___ BOTH the users and the merchants will benefit when the EPA method is applied to the problem.
USERS who DO want unsolicited mail based on content and/or seller can receive the mail. USERS who DON'T want unsolicited mail
can have it disposed of as they choose before the mail gets to their mailbox. MERCHANTS can, based on a mailing's response, reduce
the volume of such mail as they can more accurately define and target their potential market audience. ___ It's an EVERYBODY WINS
situation! ___ And the key to the solution is 'Allow the users to specify what they WANT'. ___ Where in the world - against all human
cultural conditioning and experience - did anybody get the bright idea of supplying people with what they WANT by defining what they
DON'T want??? You DON'T walk into a department or hardware store and tell a clerk what you DON'T want. DO YOU??? No!! Of course not!
You say 'I WANT ...'
'Inaccurate Spam Classification CAN SERIOUSLY HURT YOU - And Doesn't Decrease Your Work Load!
I've just had a long dialogue by e-mail about, among other things, current anti-spam products' generation
of false positive classifications - i.e., it is 'spam' but not classified as such. Such classifications are,
at most, an annoyance of varying size depending on the volume of mail that must be manually sorted and deleted or
ignored. ___ More important to you are false negative classifications - i.e , it is NOT 'spam' but is so
classified. ___ To be sure that you are not missing e-mails that are important to you, you must to save mail
classified as 'spam' and sort through it manually. ___ In terms of sorting time and effort, this situation is
slightly worse than having all mail delivered to one box. ___ Advertising for currently available anti-spam
software processes claim very low false positive rates but make no mention of false negative rates. ___ What
you NEED is anti-spam software that generates absolutely NO false negatives. You'll get additional benefits
if the software generates absolutely NO false positives. ___ EPA addresses both your needs and your wants
with a 100%-accurate classification of all of your e-mail.
'CAN-SPAM Has Not Canned Much, Report Says'
Paraphrasing three quotes from the article: '77% of all 2004 e-mail traffic was unsolicted(spam)', '97% of the spam wasn't CAN-SPAM compliant',
'the FCC and the DOJ aren't going to provide significant help'! Well, DUH! Predictably, CAN-SPAM hasn't and won't significantly hinder
spam volume growth. ___ At the risk of being repetitious, 1) Spam is UNWANTED mail - not necessarily unsolicited mail; 2) Impersonal,
effective legal definition of spam is very close to impossible; 3) Legal remedies are next to impossible because of jurisdiction,
enforcement and money issues. ___ What happens to the problem if spam is defined to be 'e-mail that the addressee doesn't WANT to receive?
Then, to eliminate spam entirely, all that's necessary is to have an addresee specify what e-mails are WANTED. A software process
applies the addressee's WANT criteria to then filter mail as it arrives at the mail server. Incoming mail is classified -
ACCURATELY, WITH NO FALSE RESULTS - as WANTED or NOT WANTED. WANTED mail gets delivered. The mail server disposes of the mail that's
NOT WANTED ___ Result? 100% of (an individual addressee's definition of) spam never gets delivered. END OF SPAM PROBLEM! ___
What software process? EPA - the one described by this site that's been operational for the last 26 months.
'Md. Judge Strikes Down Anti-Spam Law'
Because, the judge says, it 'seeks to regulate commerce outside the state's borders'. In other words, no state has jurisdiction outside
it's own borders. Note that lower courts in other states have held the same opinion for the same reason and have been reversed by state
Supreme Courts upon appeal. Are there any bets that appeals to the US Supreme Courts won't reverse the state Supreme Courts? And in all
of this legal effort, has any significant amount of spam been stopped? Wasn't it apparent from the start that jurisdiction was only
one of three major stumbling blocks (spam definition and enforcement being the other two) that preclude effective legal solutions to the
spam situation? Again, it's precise software engineering that supplies the only effective solution!
'UK hosts anti-spam summit'
'Together we can defeat spam in two years'
Doesn't the victory date always seem to be 'some time in the future' as opposed to 'now'? And doesn't the thrust
of the latest battle plan seem to be 'more laws' rather than 'some other strategy which hasn't, repeatedly, been
demonstrated to be totally ineffective'? Unwanted e-mail/spam gets stopped when 1.) Bulk mailers can
get e-mail delivered to their target audience only by clearly describing their product in terms that the target audience
uses to express their interest in the mailer's product; and 2.) simultaneously, bulk mailers
with nothing to sell ('spammers) can't get their e-mails accepted and delivered by mail servers? That solution is
precisely what the EPA process provides. Users that want to see solicitations that interest them can get the
associated e-mails. E-mail with 'gibberish' subjects can't match any well-formed natural language word isn't going
to get delivered - not matter how cleverly the gibberish is formed.
'Sender Authentication Is Coming'
Once again, a massive effort is being initiated (and wasted) by many large organizations to implement
and widely deploy a quite-obviously flawed solution to the spam problem. I say 'quite-obviously' because
the article defines the counter tactics (for unimaginative spammers?) stating 'It should also be
considered that spammers will very probably start to publish their own Sender ID records, meaning the
authentication will be pointless. There's also the problem that a compromised MTA could be used to send
spam.'. Thus, more resources are going to be wasted in this 'Red Queen's Race'
for no apparent purpose while the problem continues to grow unchecked.
'Spam Wars Hit the Next Battlefield'
Another obviously ineffective solution based on imposed regulations enforced by massive bureaucracy at
enormous cost. WHY is it so difficult to see that the only effective solution is to let e-mail recipients
define what THEY want to see and exclude the rest from their inbox? - I know what I want! - You know what
you want!! - Let us have what we want!!! - Let everyone get what they want and dump the rest!!!! -- That
leaves it up to the spammers to find a way to contact their potential customers. It also removes all of
the hacker's "Ain't I so smart and annoying" incentives (unless they're only interested in filling
self-emptying, electronic "bit buckets").
'Virus writers deploy bulk mail software'
'Zombie PCs spew out 80% of spam'
Spammers using viruses for spam transmission. Virus-writing hackers using spam for virus transmission.
What awesome symmetry! And, to make the situation worse, the anti-virus software currently on the market
seems to be far more effective than the anti-spam software (except, of course, for EPA). Using the
'zombie PC' and spam as a transmission vehicle, you can expect to see the hackers 'end run' the
anti-virus software's effectiveness. Consequently, the probability that a piece of spam will contain a
virus increases significantly - combining "annoying" and "destroying" in one package. Only EPA's ability
to accurately detect and dispose of 100% of your spam at the mail server can protect you and your domain
against spam-borne viruses.
'Singapore to make spammers pay - literally'
'Two thirds of emails now spam: official'
The politicians on the other side of the fence are no smarter than those on this side. And - gee! - your
mailbox still seems to be totally trash no matter what your ISP says about their anti-spam effectiveness.
Spam stops being anything more that an Internet infrastructure bandwidth problem when it can - 100%-accurately
with no false positives and with no false negatives - be identified and disposed of at the mail server
before it can enter a domain. If that statement is true, EPA's the operational, proven solution. No
definition problems, no jurisdication problems, no legal problems and no enforcement problems. Just good
software engineering analysis and implementation to give users what THEY want and dispose of the rest!
'MSN, Hotmail fight spam using Bonded Sender'
"So, aren't there a few definition problems here that will result in legal chanllenges at
many levels in many venues? 1.) How many pieces constitute as mass mailing? 2.) Doesn't a mass
mailing from a bonded e-mailer represent 'spam' to some portion of an ISP's clients? 3.) Aren't
some ISP clients interested in mail from an unbonded mailer? 4.) Does a mailer addressing a
political subject or agenda need to be bonded? 5.) Do political mailings and self-serving letters
from governmental agencies of all levels need to be bonded? If not, why not? If so, aren't
Constitutional 1st Amendment issues involved? 6.) Doesn't the bonding agent become a self-
policing information censor driven only by it's own interest and accountable to no one? 7.)
Isn't this whole approach another information-flow restricting, horribly expensive, unnecessary
complication of the e-mail system?? *** ALSO *** PLEASE NOTE !!! The 'false positives'
referred to that make current 'aggressive filtering' methods so ineffective occur because
these methods attempt to EXCLUDE/FILTER OUT UNWANTED mail. In that test environment, a 'false positive'
is an e-mail that isn't 'spam'
but the filter 'guessed' it was and disposed of it accordingly. Similarly, a 'false megative'
is an e-mail that is 'spam' but the filter 'guessed' it wasn't and allowed it to be delivered.
A filtering process like EPA that INCLUDES/FILTER IN WANTED mail doesn't 'guess'. Instead, it
looks for equality between specfied preferences and e-mail content. There are no 'false
positives' and 'false negatives' - only 'equal/match' and 'not equal/no match'."
'Spam Thrives Despite New Law'
'Europe Pushes Stiff Antispam Law"
"No big suprises here - just continuing demonstrations of the effects of twin lacks of
both imagination and problem environment awareness. The first article confirms that a great percentage of spam
is generated well outside the jurisdiction and enforcement capability of the spam
recipient's legal venue. The second article demonstrates that European politicians and bureaucrats
are no smarter or more effective at their jobs than their opposite numbers in the
USA. Software engineering - not legislation - solves software problems! Let's institute a
directed global search-and-test to get the metrics that identify EFFECTIVE e-mail-management/spam-control software.
Let's not waste time and effort repeating ineffective strategies and methods."
'Virginia Spammer Trial'
"Predictably, the defense attorney points out the issue of jurisdiction, asking 'can an out-of-state
business be tried in a state court because the business' communication circuits pass through the state?'
If so, every web user probably violates the laws of some jurisdiction through which a network leg passes
every day! Who - in their right mind - wouild attempt to enforce such a law on just a national (forget about
'Google reads your Gmail'
and the policy is subject to change at their convenience with notification to you. When are
you notified? .. before or after the policy change? .. it doesn't say! The policy does say that Google
can send their collected information anywhere they want. The policy also says they may be compelled
to or may, in their own judgement, decide to disclose the information to third parties. Finally,
no matter what they do, the courts have held that 'you can have no reasonable expectation of
privacy when you use e-mail'. Do you wonder what other organizations might be looking at your email?
" -- Remember: 'Big Brother is watching you! ...('1984' by
'Spam messages blocked today 510,235,483 (since midnight ET)'
(04/03/04 AOL's own "Spamfighter" page)
"How many were NOT blocked? - At the least, those in MY inbox! - At the most...?? how many AOL inboxes are there?"
NEWS: 'Over 1 million members have already signed the petition... (asking legislators to enact laws that discourage
spamming)' (04/03/04 AOL's "Spamfighter" page)
"Instead of asking someone else to solve AOL's - and the other e-mail service providers' - problem, why not be
proactive? Install an effective e-mail control and filtering functionality? Eliminating spam is the ultimate discouragement to spamming
NEWS: 'US Congress Passes Anti-Spam Law' (various media sources)
"How can an organization - many of whose members are lawyers - pass a law with basic, obvious, severe
jurisdiction and enforcement deficiencies and then expect substantial results? 'passing a law' is like
'getting a judgement' - you don't 'win' until you 'collect'!"