http://www.getpopfile.com/docs/ 2026-07-24T09:45:43+00:00 http://www.getpopfile.com/docs/ http://www.getpopfile.com/docs/lib/tpl/getpopfile-org/images/favicon.ico text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:abucket?rev=1202500160&do=diff What is a bucket? POPFile categorizes your email into two or more categories. In POPFile terms, POPFile separates your mail into two or more buckets. Thus a bucket is nothing more than a category. “Bucket” however, sounds more friendly and less academic. text/html 2011-05-03T09:58:32+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:amagnet?rev=1304416712&do=diff What is a magnet? Magnets are a POPFile feature that lets you bypass the Bayesian filtering system. For example, you can set up a magnet that will put every message from a certain sender into a certain bucket. You should use magnets with care. If you are just starting to use POPFile you may think that magnets are the way to go. But if you set up a magnet for everyone you communicate with, all those messages will bypass the Bayes engine and POPFile will never learn what a good message looks lik… text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:apop?rev=1202500160&do=diff What is APOP When your email clients connects to your mail server with a regular POP3 connection, your login and password are sent in plain text across the network (i.e. the internet). If someone wants to get at your password, he thus has a good chance of getting at it. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:aproxy?rev=1202500160&do=diff What is a proxy? A proxy is a piece of software sitting between two other pieces of software and doing something to the communication the latter to exchange. In the case of POPFile, POPFile is sitting between your mail client and your mail server. POPFile will pass all commands that your client sends on to the server. When the server sends a mail message to the client, POPFile will intercept the message, parse it, classify it, and possibly change the message by changing the subject line or add… text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:bayesian?rev=1202500160&do=diff Bayesian Bayesian, Bayes, and naive Bayes are often used in discussion of POPFile and similar email filters which usually refers to a mathematical formula. In the 1700's Thomas Bayes was a man who studied probability, his work has since become known as Bayesian Statistics. And this method has recently become popular in email filtering because it does a very good job determining between different groups of messages. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:category?rev=1202500160&do=diff Category See bucket text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:emailclient?rev=1202500160&do=diff Email Client An email client (also: mail user agent, MUA) is used to download, organize, store, and read email. Examples include Qualcomm's Eudora, Microsoft Outlook, Mozilla Thunderbird and more. See also: * Configuring Specific Mail Clients * mail server text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:falsenegative?rev=1202500160&do=diff What is a false negative? A false negative is the little kid brother of a false positive. If POPFile (or any other mail filtering software) classifies a spam message as ham (good), we have a false negative. False negatives may be irksome, but false positives can be dangerous. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:falsepositive?rev=1202500160&do=diff What is a false positive? When reviewers compare different spam filters, one of the most important aspects that is always disussed is the number of false positives. A false positive is a good piece of mail (ham) that gets mistakenly tagged as spam. In other words: a false positive might end up in your spam folder or might even be deleted (depending on the rules you set up in your email client), altough it really isn't spam. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:falsepostive?rev=1202500160&do=diff What is a false positive? When reviewers compare different spam filters, one of the most important aspects that is always disussed is the number of false positives. A false positive is a good piece of mail (ham) that gets mistakenly tagged as spam. In other words: a false positive might end up in your spam folder or might even be deleted (depending on the rules you set up in your email client), altough it really isn't spam. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:ham?rev=1202500160&do=diff What is Ham? Ham is often used as a term for good or non-spam email. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:header?rev=1202500160&do=diff Header An email header contains information about who sent an email, to whom, when, how it got there, and so on. Most of this isn't critical to everyday email reading so email clients hide it. A full email header looks like this: Return-Path: <noreply@spammer.net> Delivered-To: user@company.com Received: (qmail 4584 invoked by uid 64014); 14 Nov 2003 23:38:27 -0000 Received: from noreply@spammer.net by foo@company.com with AmikaGuardian-Server-2.0.0 (Processed in 2.524528 secs); 14 Nov 2… text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:history?rev=1202500160&do=diff What is the history? The history tab in the POPFile UI shows you the messages recently classified by POPFile. You can have a closer look at the messages here and you can reclassify single messages or complete bunches of messages. By default, POPFile will keep two days of history. You can change this setting on the configuration tab. You can safely remove messages from the history by using the remove controls. text/html 2013-03-11T14:51:14+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:loggerlevel?rev=1363013474&do=diff Logger Level Starting with POPFile v0.22, you can adjust the logging level. You can find it on the Advanced Tab; the variable's name is 'logger_level'. The default setting is 0 which will only log critical errors, 1 is much more verbose, and you get to see everything if it's set to 2. If you are debugging a problem setting to 2 is probably necessary. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:mailclient?rev=1202500160&do=diff Mail Client A mailclient is the software you use to read your email. Examples include Netscape/Mozilla, Outlook, and Eudora. See also: * Configuring Specific Mail Clients * mail server text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:mailserver?rev=1202500160&do=diff Mail Server A mail server receives and holds your email until you download or delete it. Your ISP (Internet Service Provider) or system administrator will tell you the name and/or the IP address of the server. Sometimes the same server will work for both sending (SMTP) and receiving (POP or IMAP) email and other times there will be two separate servers for these functions. Often the name is something like mail. text/html 2008-11-30T18:05:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:popfilecoreteam?rev=1228068320&do=diff Who is the member of the POPFile Core Team? Here's the list of the POPFile Core Team members: * Brian Smith * Joseph Connors * Manni Heumann * Naoki IIMURA (amatubu) * John Graham-Cumming (as Chairman) text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:pseudoword?rev=1202500160&do=diff What is a pseudoword? Usually, POPFile relies on the words it finds in email messages to do its classfications. But there is more to an email than just the words. For example, if a word is found in the subject, this should be noted. Pseudowords were created to store just this kind of meta-information. You can identify pseudowords by their special syntax: pseudoword:value. A list of all the pseudowords used by POPFile can be found on the text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:quarantine?rev=1202500160&do=diff What is the quarantine function? Usually, POPFile will classify a message, change the subject line and/or insert the X-Text-Classification header, and then hand the message on to your mail client. It's still the same message though, and it may contain web bugs or exploits that cause the preview function of your mail client to do something really nasty. For example, spammers may put some code in the text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:spam?rev=1202500160&do=diff What is Spam? Spam or junk mail is hard to define. Everyone has their own definition, but a common factor usually considered junk mail is unsolicited commercial email. For example if you buy something from eBay you want and expect a confirmation email from the seller that they have received payment and have shipped, but you probably don't want to be put on their mailing list automaticly. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:stealthmode?rev=1202500160&do=diff What is stealth mode? POPFiles acts as a proxy for POP3 connections and it displays its user interface through the browser. By default, both the POP3 proxy and the http UI will only accept connections from the local machine and not from someone somewhere on the internet. That's what we call stealth mode. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:stopword?rev=1202500160&do=diff What is a Stop Word / Ignored Word ? POPFile keeps a list of words that it will ignore if it finds them during classification. These words are likely to be used in any sort of message, such that they are not likely to be useful in classification. Now that POPFile uses a database to store and look up words it makes little if any noticable difference in speed. text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:subjectmodification?rev=1202500160&do=diff What is subject modification? POPFile can modify the subject lines of incoming messages so that they reflect POPFile's classfication. This takes the form of: * Subject: [bucketname] original subject line * Thus, “Subject:Howdy,kid?” would become text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:thecorpus?rev=1202500160&do=diff What is the corpus? When you reclassify a message, POPFile will add the words in that message to a list of words that is associated with the bucket that you reclassified the message to. Thus for each bucket there is a list of associated words. All the bucket lists together form the corpus. text/html 2009-09-30T13:34:28+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:toe?rev=1254317668&do=diff TOE (Train Only on Errors) POPFile currently works best by Training Only on Errors, commonly referred to as “TOE”. When you find POPFile has misclassified a message you reclassify it, this is most likely what you are doing now. An alternate, less popular strategy is to text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:unclassified?rev=1202500160&do=diff What is unclassified? Sometimes POPFile will not put a message into one of your buckets, but will instead mark it as 'unclassified'. In colloquial terms, this happens when POPFile is not sure about the classification of a message. In technical terms, this means that POPFile will calculate a probability score for each of your buckets for each message. Usually, the message goes to the bucket with the highest score. But when the difference of the two highest scores is less then the unclassified we… text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:wordcount?rev=1202500160&do=diff What is the (unique) word count? When you reclassify a message, POPfile will add the words in that message to the word list of the corresponding bucket. In this list, the words are stored along with a word count. Thus if POPFile has seen the word “ text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:wordsalad?rev=1202500160&do=diff What is a word salad? Word salad is a term given to a spammer technique that is supposed to fool filters into not tagging a message as spam. The technique involves hiding or attaching a list of supposedly non-spammy words to spam messages. Here's an actual example: text/html 2008-02-08T21:05:16+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:xpopfile?rev=1202504716&do=diff What is the X-POPFile-Link header? POPFile can add an extra header line to each email processed. This line has the form “X-POPFile-Link: URL”. The URL is the URL to a single message view of that message. If your mail client plays along, you can click on this text/html 2008-02-08T19:49:20+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:xtextclass?rev=1202500160&do=diff What is the X-Text-Classification header? POPFile can add a new header line to each processed email. This header line will look like X-Text-Classification: bucket where “bucket” is the bucket POPFile put this particular message in. You can use this header line to set up filtering rules in your mail client that move messages that belong to certain buckets into corresponding folders. text/html 2008-09-23T20:37:38+00:00 Anonymous (anonymous@undisclosed.example.com) http://www.getpopfile.com/docs/glossary:xtimeoutprevention?rev=1222202258&do=diff What is X-POPFile-TimeoutPrevention header? For preventing the timeout of your mail client, POPFile sends X-POPFile-TimeoutPrevention header every 60 (by default) seconds. In case of receiving a long e-mail, these headers will be added to the e-mail.