Why do I get told to change my e-mail formatting?

There's a range of issues here.  Some of it is down to conforming to the etiquette of where you're posting e-mails (read a few, before you start posting to a mailing list), as well as more general (overall) e-mail etiquette (many mail programs come with some information about this, in their documentation); and understanding the limitations of e-mail systems (many mail clients will let you prepare a message that will ultimately fail to reach it's destination, and with no warning either).

NB:  The same issues apply to news groups (usenet), as well as e-mail.

Technical issues

Starting a new line with the word “from” (especially if there's a blank line preceding it) can cause some mail systems to foul up on the mail.  They can misinterpret that word as the indicator of the beginning of a new message (and, therefore, the end of the current message), then reject everything below that line, as it doesn't have any e-mail header fields.

You'd need to understand the construction of e-mail source codes to see the reason why, but simply put, mails begin with a from line, then several other mail headers, a blank line, then the message body content (the part that you read and write).  A mail spool file may have several messages, all concatenated together into the one file, and the only thing that seperates one from the next, is a blank line preceding a “from” line.  Any system that works that way, can be susceptible to this kind of problem.  Most aren't, but there's still a few around that are.

Using specialised codes, such as HTML and other text-style mark-up codes, can often make the message unreadable for other people.  Everyone will not use the same e-mail client as you, and some of the features available in yours might not be available in theirs.  And some of these extras, such as HTML mail, can make the actual message source code a lot bigger than it would have otherwise been (than if you'd just used plain text instead), as well as increase risks of accidents and virus propagation, and not to mention being difficult to read for some people.  Microsoft's Outlook Express, for instance, makes the most appalling HTML messages (they're full of unneccessary HTML elements, and badly chosen ones, too).

It's also worth noting that trying to send message with files attached, or using some other fancy features, can cause messages to be rejected by some mail systems, particularly corporate mail systems, for security reasons.  And some systems will strip out HTML and attachments, leaving just a blank message for the recipient.

How do you turn off HTML message generation?  Read the manual that comes with your mail program, and check our the configuration options available to you.  Picking “plain text,” or un-picking “HTML” or “rich text” authoring, are the sort of options that you're looking for.


Attachments to messages can represent a security problem, as it's all too easy for the attached files to be deliberately, or even accidentally, dangerous.

Some people will instantly delete mail with attachments, without even reading the message.  So it can pay to check with someone before sending them a file with a message.  Or, sending the file separately, so they may still (at least), get to read your message, even if they deleted the file without checking.  Also, they may not have anything capable of using a file that you might send them, especially things like word processor files (everyone does not have Word, or a way to read Word documents), so this is yet another reason to check before sending attachments.

Likewise, if you receive a message with a file attached, you may wish to think about what you're going to do with it.

It's usually not a good idea to open files within your mail program, unless you're absolutely sure the file will be safe.  Don't assume it's safe because it's from a friend, they may not know it was dangerous, or may not even know they sent you it (they may have been infected with a computer virus, that sent the message, unbeknownst to them); nor assume that your mail program will protect you.

A better course of action, is to save the file.  This gives your virus checker a chance to check the file contents, if you've got one set up to check files being operated on.  If your virus checker doesn't scan all files in use, in that manner, then deliberately scan the file yourself.  If you don't have any anti-virus software, then get some.

If the file is a script, or batch file, a virus scanner may not find anything wrong, even if the file is dangerous.  It's not impossible to write a script that can delete files, or reformat a hard drive, and those commands aren't something that a virus scanner would be looking for (virus scanners check for known patterns of code, in known virus programs, which also means that they won't find viruses that they don't already know about).

In these circumstances (when you've received a programming script type of file), it's a good idea to check the file in a simple text viewer (one that will simply show the contents of the file, but not perform any instructions that may be in the file).  DON'T double-click the file, to try to see the contents, that will “run” most script files.  Start a text viewer (not a word processor, like “Word”), and open the file in it; or select to “edit” the script file, instead of “opening” it.

Word processors, like “Word” can run macros (small programs) embedded inside a document, which can do dangerous things.  Simple text editors (like “Notepad” and “Edit”), will just display the contents of the file, not caring about (or acting on) what's inside the file.

If you've set your computer to hide file suffixes (i.e. to show a file called “readme.txt” as just “readme”), then you'll get no indication of what type of file, a file is.  This is a very dumb thing to do, even if you think it looks neater.  Particularly when you receive a file called something like “readme.txt.exe” which isn't really a text file, but will look like one if you're hiding file suffixes (i.e. it'll look like “readme.txt”).  This is a very common way used to fool people to run dangerous programs.

A very nasty Windows trick is to always hide .shs (ShellScrap) files, even if you deselect that option (Windows will secretly keep restoring that setting).  Meaning that some seemingly innocuous “readme.txt” file, that's really a “readme.txt.shs” file, could do anything when run (it's not a text file, at all; that's just a deceptive name that's been given to a ShellScrap file, ShellScrap scripts are small programs).


E-mail addresses need to be “letter perfect” to get to the correct destination.  Mail systems can't guess what you want.  If you address a message incorrectly, it'll either fail to be delivered (usually, but not always, you'll receive an error message back from the server), or it'll be delivered to the wrong person (if your incorrectly addressed message just happens to match someone else's real e-mail address).

E-mail systems allow a few tricks for making addresses look better, particularly if they're something like dkk432@localhost.invalid.  That sort of address gives no indication of who you are.  You can add your real name, or a nickname, or even comments, to the addressing information; as long as you do it in the correct manner.

If you enclose the address within pointy brackets, then other text on the address line can be ignored (by the server) as being nothing more than just “comments” (it will have no effect on the destination), and comments inside curved brackets will be ignored as well.

TO:  John Smith <dkk432@localhost.invalid> (home address)
TO:  John Smith <dkk432@example.com> (work address)

These examples start with a “display name” (displayed when the mail is read, but the mail server doesn't care about them), followed by the e-mail address, with some comments (in brackets).  It is the address that the mail system pays attention to, the other information is only for your benefit.

If you have problems writing to someone who has put a comma between their names, then enclose their names in quotation marks (the comma is used, in e-mail systems, to separate multiple addresses when sending one message to more than one recipient).

e.g. "Smith, John" <dkk432@localhost.invalid>

To send one message to several people, you can type all their addresses in the TO: or CC: fields (CC stands for carbon copy), separating each one with a comma.  Bearing in mind that all the recipients will see all of the addresses the message has been sent to (this may not be a good idea).

e.g. TO:  123@example.com, 456@example.com

There's a slight distinction between putting all the addresses in the TO field, or putting one address in there, and the rest in the CC field; but it's a personal rather than technical issue.  An example might be sending a letter to a politician, with copies sent to a collection of journalists.  This shows who the message was “for”, and who else got to see it.  This makes it a bit easier for anyone else to send their comments to the right person if they want to follow up on anything in it:  They can simply send it to the TO address, eliminating all the CC addresses in one go, rather than try and pick the right address out of a long list of them; as it's more than likely that everyone else does not want to see follow-ups.

To send a message to a group of people, without them knowing who else has received the message, use your mailer's BCC function (blind carbon copy) instead.  Put your own address in the TO: field (everyone can see the contents of the TO: field, and you'll receive a copy back yourself that'll will look exactly the same as the message everyone else receives), and put all the other recipients addresses in the BCC: field (the addresses written in the BCC field will usually be removed by the mail system as it handles the message; though test this first, not all of them do that).  And so that the recipients don't wonder why they've received a message addressed to you, place a suitable comment in the TO: field.


TO:  friends of John Doe <me@localhost.com>
BCC:  123@localhost.invalid.com, 456@localhost.invalid.com

Don't try this as a spamming techique, apart from it being a nasty habit, and usually against the terms of service between you and your ISP, the messages can, and will, be traced back to you.

Note that not all servers treat the BCC field as expected.  Some will strip out the addresses listed in it, before delivery (hiding the list of recipients from everyone as you want), and some won't (so everyone will be able to know who else received the message).

The point about not distributing lists of e-mail addresses, willy nilly, is to stop them being used for spamming.  Somebody on the list of addresses might be one of those cretins who collect addresses, and send spam to every address they collect.  Or, somebody on the list might have a virus that does the same thing without their knowledge.  Many people hate spam, and with the huge number of people on the net, it's easy for spam to multiply so much that a person can easily end up receiving hundreds a day, or worse, making their e-mail address useless to them.


An important thing to remember is that people can't see you (to pick up your body language), when reading a message from you.  Some things can sound just rude, when typed out.  Sticking disclaimers, and witty remarks, into your message doesn't negate any insults you may have typed in elsewhere.

If you're asking for help with something, be helpful yourself.  Just stating something doesn't work, isn't going to get you much of a useful reply as to how to resolve the problem.  Include some other information too, if it seems appropriate.

For example, if you're asking about a problem with a computer program that does actually start, but fails at a certain point, then say so.  Or, maybe provide some information about your computer, like how much RAM you have, and which version of the operating system.

Likewise, saying that something “sucks,” continuing on to denigrate it, and the person who created it, then asking for support, just isn't going get you anywhere at all.


It's good practice to quote some of a message that you're replying to, so people know what you're talking about (don't assume that they read, kept, or even remember, the original message).  Writing your responses directly below the parts of the message that you're replying to, in the manner that someone might script a dialogue (quote something, respond to it, quote the next thing, respond to that, and so on).  This way, people can tell exactly what you're responding to.  But don't quote stuff that you don't need to, such as irrelevent parts of the message, headers and signatures.  If you find it hard to avoid quoting a large portion of text, consider not quoting it, but just summarising what they talked about.

Make it clear that something is quoted text.  The most widely accepted convention for this, is to place a greater-than symbol:  > at the start of the quoted text, preferably at the start of each quoted line (with no blank spaces before them, nor between multiple quote indicators, and a blank space between the right-most quote indicator and the quoted text), and to also leave blank lines before and after the quoted text (there's examples of this, below).  Many e-mail clients will do this for you, though with varying degrees of neatness.

Some e-mail clients also insert the initials of the person you're quoting, in front of the quote indicator, but this can cause problems, such as:  Some e-mail clients no-longer recognising quoted text as being quoted text, so when someone replies, the formatting becomes scrambled, with the quote indicators splattered throughout the middle of sentences.  Even worse, is when they stick initials after the quote indicator.  And, either way, they'll also (most likely) stick the initials in front of all quoted text, even if it's text written by somebody else (e.g. quotes of a quote will be attributed to the wrong person).

A very common, and good, method of replying with quoted text:

On the 12th of May, 2000, Fred wrote about "how to quote", so I thought I'd write back…

Hi Fred,

Yesterday we were discussing something, and now I'll just blather on some more.

>> This is some text that's been quoted from two
>> two messages back (it's a quote of a quote).

You'd respond to this (the above text), here.

> This is some quoted text from the previous
> message, so you can remember what we were talking
> about the other day.

I had some further thoughts on the issue, and here they

> This is some more quoted text.  Again from the
> previous message.

And I'll write a response to this part of the message, here.


Don't forget to visit my web site, at…

The quote heading, mentions when the previous message was, and what it was about (to help jog the memory, or help find the original message.  Some of the message has been quoted, to continue the conversation more easily (so the reader doesn't have to find the previous message).  And the author has a little note that they place at the end of all their e-mails (the signature), with some little message they like to spread around, in this case an invite to look at their web site (e-mail signatures are a similar idea to how people write letters on personalised stationery).

Conforming to the norm makes it easier for people to reply to messages (usually their e-mail client will have some automatic routines for dealing with quoting text that's already quoted, but they'll only work if they can identify which parts of the text are quotes).  As well as just making it easier to read.

You may think that unimportant, but people who read a lot of mail get annoyed at badly formatted mail, and may just delete it without even reading it (rather than trying to interpret it).  And if your message was a request for something, you'll (most probably) want as many people as possible to read it.  They'll probably tell you to sort out your quoting too.

Some e-mail clients will automatically avoid quoting anything below the signature tear-off line (so don't write replies to messages underneath a signature tear-off line, or anyone replying to your message may have a difficult time trying to quote your text).  Because it's easy to remove something at the bottom of a message, this is where personalised message information is usually placed (it's a waste of bandwidth and storage space to quote irrelevent parts of a message, and annoyed people will readily point this out to you; so if your mailer doesn't automatically cut off signatures, do it yourself).

Some people reply by quoting your entire message, then typing their reply above it (simply adopting the quoting technique used by at least one Microsoft mail client), while others will intersperse comments between quoted parts of the original message (as in my example, above).

It's logically easier to read replies written the second way if you have to relate a comment to part of the original message, and it's better to not bother quoting irrelevant parts of the original message, too.  Especially when messages get quoted back and forth (just four replies later, and the message grows to rediculous lengths, and gets incredibly messy to read).  But you'll need to pay attention to the preferred way of quoting, when replying to messages in a mailing list, and do it their way.

Another common method of quoting when there are multiple participants:

Brian wrote:

>>>> Something about four messages back
>>>> (it's a quote of a quote of a quote of a quote).

And I might chip in, writing my comments, here.

John wrote:

>> Something two messages back, and here's the part of
>> it that was worth quoting.

Fred wrote:

> This is some more quoted text, from the
> previous message.

And I'll write a response to this part of the message, here.

Line length

Line length is a bone of contention too.  Most people can see up to about eighty characters across their screen, so it's conventional to use less characters than that (leaving space for quoting, scroll bars, etc, it's also easier to read text that's on shorter lines, with about sixty characters being a comfortable maximum width).

If you write longer lines than that, some people's mailers will wrap the text onto the next line, some won't (they'll have to scroll sideways to read), or it may even just chop off and discard the extras words off the edge of the page.  Some e-mail servers will trash mails with overly long lines, or the server may even crash (the SysAdmin won't be happy about this).  There is a technical limit to the lengths of lines, as the server has to parse the entire message looking for information necessary to send it to the right place, as well as identify where the end of the message is.

If you want to use very long line lengths (like when people switch off any word-wrapping of their message text), you should use a mailer that supports quoted-printable, MIME/base64, or format=flowed, encoding.  (As you write the message, and when reading it in a compatible mailer, you'll be unaware of any line breaks inserted by the encoding method.  Of course the people receiving your mail will need a mail client that supports the same encoding methods.  And messages formatted that way are often difficult to quote, when replying.)

What quoted-printable encoding will do is, wrap the message source code; so that the data which is actuallly “sent”, has short lines (this will keep mail servers happy); but it'll insert a code at the point it wrapped the line, informing the other mail reader program to ignore this (unwanted) break in the line.  The mail reader will then unwrap the text, when it's displays it (but it may still make a mess of quoting it, in any replies).  It'll also encode other characters which don't send well in a plain-text system.

Base64 encoding will encode the contents of the entire message before sending, and decode before reading it.  The structure of the message source is designed so that it will survive going through plain text systems; and, again, the way the message ends up being displayed is dependent on how it was written, not how it was sent.

Format=flowed keeps the message source line-length short, to keep plain text readers happy, but in a way that compatible mail programs can ignore (they'll re-flow the message to fit the available reading space, while still keeping line and paragraph breaks where the author typed them).

Note that using quoted-printable is more compatible with less-able readers than base64 encoding.  Quoted-printable uses just a few codes for some things, placed throughout plain text (so it's still readable, if your reader can't decode it; you'll just have to ignore a few odd little bits in it).  Base64 encodes everything, and if your reader can't decode it, you'll just see a mass of seemingly random characters.  Format=flowed should be compatible with everything, it's as close to plain text as can be, while allowing fancier mail programs to play with the contents to suit the reader.


Signatures are a footer that many people add to their messages.  They're not a signature in the hand writing sense, but play the same role as personalised stationery.  People often include links to their web sites, their alternative e-mail addresses, business contact information, etc.

There is a widely accepted method of defining the point between the end of your message and the beginning of the signature, by starting a new line, typing two hyphens, followed by a blank space, then starting another new line (so there is a single line with these three characters on it, all by themselves).  Send a message to your own address to see how your mailer works (some add the signature tear-off line themselves, even if they don't show it, or your signature; while you're typing in your message).

Some e-mail clients will draw a dividing line across the page when reading a message prepared that way.  Also, some e-mail clients will automatically avoid quoting anything below the signature tear-off line (so don't write replies to messages underneath a signature tear-off line, or anyone replying to your message may have a difficult time trying to quote your text).  And some mail clients can be configured to never show the signature, so don't put essential stuff in it.  Don't quote previous messages below a signature separator, it makes it very hard for the next respondant to quote them as well as you.

Long signatures aren't too appreciated.  In some cases, they're longer than the messages, and people who receive lots of e-mail don't like their download times being needlessly increased.  It costs them money.  If you've got a story to tell, consider publishing it on a web site, instead.

Smartarse comments aren't appreciated, so various jokes stuck in your tag-lines often provoke responses.  Some responses will be quite harsh (to put it midly).

Spam busting

If you participate in public news or mail, it's almost inevitable that you'll receive some junk mail.  Such as, from list servers that deliberately distribute people's e-mail addresses (or just make it too easy to poll it for them), or nitwits reading the messages who think you won't mind receiving some junk mail.

Never respond to junk mail, it only encourages them, and proves that someone read their message.  Likewise, don't use their (usually fake) unsubscribe instructions, for the same reason.

Reporting spammers is a task that requires properly understanding the message headers, and working out the correct service to complain to.  Because it's complex to explain for the first time, and too easy to get interpreting e-mail headers wrong, and end up complaing back to the spammer, or about some innocent third party, I'm not going to detail how to do that.  Only attempt it when you're completely familiar with how e-mail works.

You may wish to consider having more than one e-mail address, and posting from an alternative one, when participating in public mail.  This will (hopefully) keep your personal mail address free from rubbish.  Or, you can try mangling your “from” and “reply-to” addresses in some manner, so replies to it will fail (though this will raise the ire of some hotheads, who insist that all messages have valid reply addresses).  Also bear in mind that some systems won't allow the use of fake addresses, so you may need to set up an address that you don't care about, and use it as the “from” address (i.e. a “real” one, but one that you'll never be trying to use; and you won't need to try and find real mail in the middle of all the spam).

If you're mangling your address, stick .invalid onto the end of the address.  This will stop compliant mail systems from even attempting to deliver messages addressed that way.  This is important, as it's just possible that there really is an e-mail address the same as whatever you write your mangled one as, and they won't appreciate receiving unsolicited mail.  For that reason, pick a faked domain name that cannot exist as an internet address; like “localhost” (the localhost address refers to the same machine that you are using), or “example.com” (the example.com domain has been registered as an address that can be used for examples, but is not used for any real services; therefore using it any public manner that might be stored, copied, indexed, or whatever, should not interfere with a running service).

e.g. A mangled fake address like <getlost@localhost.invalid> should do the trick, rather effectively.

If you do want people to be able to reply to you directly, then write a few instructions into your signature, saying how to unmangle the address (use an address that you don't mind dumping, if you have to).

For example:


Yesterday we were discussing something, and now I'll just blather on some more.

> This is some quoted text.  From the previous
> message, so you can remember what we were talking
> about the other day

I had some further thoughts on the issue, and…


Please send any personal replies to getlost@localhost.invalid
but change "getlost" to "me" and remove ".invalid"

Although that won't stop people who're on the list (who send junk mail to addresses they read in their mail), from sending junk mail, it will probably stop automated scripts that harvest addresses from lists.  Though, as in life, there are no guarentees.

By the way, you may not think it worth the effort to reduce spam, but many of us do.  Some of us have spam free mail boxes, and want to keep it that way.  Don't spam, and don't spread our addresses around.

Main sections:
contact details
business info
personal info
eBay & trading
“sales” ads
“wanted” ads
video production
misc info
website info/help
Misc info pages
Internet related
clients & servers
e-mail issues
privacy policies
sending documents
technical terminology
PGP messaging
webpage wierdies
Video related
colour video
interlaced video
persistance of vision
VHS versus Beta
S-VHS isn't VHS
video tapes
video editing
DVD formats
E-mail info page
line length
spam busting