Search the FAQ Archives

3 - A - B - C - D - E - F - G - H - I - J - K - L - M
N - O - P - Q - R - S - T - U - V - W - X - Y - Z - Internet FAQ Archives FAQ - General info

( Part1 - Part2 - Part3 )
[ Usenet FAQs | Web FAQs | Documents | RFC Index | Schools ]
Archive-name: pictures-faq/part2
Last-modified: 05 March 1993

See reader questions & answers on this topic! - Help others by sharing your knowledge
	This is part 2 of the FAQ for the* hierarchy.  
    This part of the FAQ contains "general", or
    operating-system independent information.  It answers (hopefully) all 
    the questions you may have about the pictures newsgroups, decoding and 
    encoding techniques, and picture formats.

      For information on issues of etiquette and posting policy and/or 
    suggestions, consult part 1 of this posting.

      For information on your particular system and on specific utilities, 
    consult part 3 of this posting.

    Before posting to these groups for the first time, please check the FAQ
    list (this posting - including parts 2 and 3), and also read the newsgroup 
    news.announce.newusers, which contains many answers to questions about 
    UseNet in general.

    If you've read previous versions of this FAQ, you'll probably only want
    to read anything that has changed since the last distribution.  These
    changes appear both in this document and in the accompanying "Changes to
    the FAQ".  Note that this is a "live" document, and 
    is always getting important information added or updated.


This file is intended to be a general introduction to the pictures
newsgroups, answering some common questions concerning pictures posted 
in those newsgroups, namely how to decode and view them.  It is not, of 
course, possible to cover everything, but I will try to to get as much 
as I can into this file.  If you feel something important has been 
omitted and you know the subject well, please write me so I can 
include the info for future releases.  E-mail should be sent to for these purposes.

Before you miss an important detail contained in this file, let me
"pre-repeat" that many of the programs mentioned in this document are
available for anonymous ftp at (, in 
the gifstuff directory.  Also: there are NO GIF files of any kind at 
this site!  Save your time and don't bother looking for them!

OK... on to the real reason you're reading this document...



This FAQ is posted every other Monday to the 
newsgroups and to news.answers.  It is also available by anonymous FTP, 
from UUCP, or through e-mail by using the services available from a couple 
of mail servers.  For anonymous FTP access, you can look on either [] in /pub/usenet/news.answers/pictures-faq
in files "part1.Z", "part2.Z", or "part3.Z", on 
[] in /pub/NEWS.ANSWERS/pictures-faq for "part1", "part2", or 
"part3", on [,, or] in 
/usenet/news.answers/pictures-faq as the files "part1.Z", "part2.Z", or 
You can get the FAQ via UUCP by retrieving the appropriate part from
"uunet!/archive/usenet/news.answers/pictures-faq/part2", or
    For e-mail access, send a message to 
with the mail body "send usenet/news.answers/pictures-faq/part1" to get the 
first part, "send usenet/news.answers/pictures-faq/part2" for the second, 
and "send usenet/news.answers/pictures-faq/part3" for the third, or e-mail 
to with "send NEWS.ANSWERS/pictures-faq/part1",
"send NEWS.ANSWERS/pictures-faq/part2", and/or 
"send NEWS.ANSWERS/pictures-faq/part3" in the body of the message.


	Basic checklist:		Alternate checklist:
	----------------		--------------------
	News reader			News reader (optional in some cases)
	Text file editor		"Super-decoder"

By far the most common method of posting files to the pictures 
newsgroups is the UUENCODE standard.  This program, shipped standard 
with most implementations of UNIX, converts binary files into plain-text 
ASCII files which can be handled by the mail system.  You will need a 
version of UUDECODE before anything else in order to view anything 
downloaded from the net.  If your system does not have a version of 
UUDECODE available, you can get one via anonymous ftp from, in the gifstuff/uutools directory.

The format of a uuencoded file consists of an optional "table specification",
which consists of the word "table" alone on a line, followed by one or more
lines containing the characters that will be used in the remaining encoded
data.  Following this, the standard requires the line containing only the text 
"begin <permissions> <filename>" (where "<permissions>" is a three-character 
numeric string, and "<filename>" notes the name of the decoded file - for 
example "begin 640 myfile.gif").  This "begin" line is then followed by 
several lines of approximately 61 characters, all beginning with a capital 
"M", and containing any non-lower-case printing character (and very rarely 
resembles anything but absolute gibberish).  Optionally, one to two lines 
may be blank or contain less than the normal number of characters if those 
lines are immediately before the line containing the "end" notation (in this
case, these shorter lines will NOT begin with "M").  The "end" text alone on 
a line marks the conclusion of the uuencoded data.  Any information that does
not fit into the above classifications are termed as either "headers" or 
"trailers", and are not intended to be included in the information to be 
decoded.  For example, the following represents a valid uuencoded file 
(although it contains no useful information - don't bother decoding it!):

begin 666 bogus.file


Most decoders are smart enough to ignore anything before the "begin" line 
and after the "end" line.
The first step is to save the file you want to view... in most versions
of the newsreader, this is done by pressing 's' followed immediately (no
spaces usually, although some versions don't care) by a file name.
You will usually be asked if you want to save it in mailbox format;
you should answer 'n'.  When saving an article to a file in
mailbox format, the article is sometimes changed in a subtle
way, making it impossible to decode.

In the case of a single-part file, you can now uudecode the file, 
which will create whatever output file is encoded.  You can usually 
tell if it's a single-part file by looking on the subject line; 
standard netiquette is to make something like [03/06] part of the 
subject line, which indicates you're on part 3 of a 6-part file.  If no
numbers are there, you can usually assume it is a 1-part file.  If 
not, feel free to write the poster (directly... please don't waste 
bandwidth by posting) and request that he/she put this info in the 
subject line.  Be nice about it!  Another way to determine if a file 
is a single-parter is if both the uuencode "begin" and "end" lines 
(as outlined above in the description of the uuencode format) are 
included in the file.

For multi-part files, life is a little more difficult.  If all you 
have is a standard UUDECODE program (as opposed to a "smart decoder"), 
you will need to trim the headers and trailers out from the rest of the 
information.  You can either do this by saving each part in its own file 
and editting them separately, then concatenate the editted files together 
to make one big file (this might be your only choice if your editor can't 
handle large files!), or you can save each part in order into one big 
file and then edit all the headers and trailers out from that file.  
Either way, you'll need to run the result through UUDECODE.  You can use 
your favorite text editor to strip out header and trailer information.

There are several "smart decoders" out there that will handle all of
the header/trailer stripping and decoding for you (some will even make
sure that the pieces are in order!) - see part 3 of this posting for 

Some articles are actually posted with easy decoding in mind, and contain
UNIX shell script headers/trailers that facilitate easier decoding.  This
is often very helpful, as it saves you a lot of work, and can also provide
error checking not available in a "normal" uuencoded posting.  These
postings nearly always contain instructions on their use, so I won't
attempt to explain all the details here.  There's no set "standard" for
this type of posting anyway - except for MIME.  MIME, the Multipurpose 
Internet Mail Extensions, proposes a standard for the posting and mailing
of multi-media articles (postings may include pictures, sounds, movies, 
or other media types - which may be combined in one article).  Public-
domain packages using MIME are available (Metamail, for example).  For 
more information on MIME and Metamail, contact

Some news readers have an "extract" capability that greatly simplifies
life by automatically decoding articles - this means you don't have to
go to the hassle of saving to a file and then decoding.  Newer versions 
of rn, nn, and trn can handle this - check the "man" page or ask your 
news administrator to find out if you can let your news reader do the 
work for you!

If you're going to download the decoded picture file to a home machine, 
or move it around a network, remember that most decoded file outputs are 
going to be BINARY files, so set your transfer protocol accordingly.  
If you are moving around just the uuencoded data, an ASCII transfer will 
work just fine, however (you'll have to decode it eventually, of course).  
Note that if you *don't* transfer the decoded file in BINARY mode, 
everything will appear to work just fine - until you try to view the 
picture.  Then you'll get all sorts of undefined results...


	Basic checklist:		Alternate checklist:
	----------------		--------------------
	GIF viewer			Multi-format viewer
	Format conversion tool(s)	Format conversion tool(s)
					Image manipulation tool(s)

OK.  Now you've got this great picture file from downloading it and 
running it through UUDECODE.  What is it, and what do you do with it?

The most common type of picture is the GIF format (which usually has
a .GIF or .gif file suffix).  GIF stands for Graphic Interchange Format, 
and is a standard format for images that was developed by CompuServe to 
be a device-independent method of storing pictures.  It includes 
Lempel-Ziv-Welch (LZW) compression, which makes the files fairly small.

JPEG is another standardized image compression mechanism, which stands 
for Joint Photographic Experts Group (the original name of the committee 
that wrote the standard).  It seems more and more common that JPEG-type 
pictures (.JPG or .jpg file suffix, usually) are getting posted to the 
net.  Some claim that JPEG is destined to overtake GIF format in popularity, 
because it is the most compact method to store 24-bit data, but mostly due 
to the fact that it uses much less space to store the same picture (this is, 
in fact, true - I have seen many examples of this phenomenon).  This may be 
an accurate assessment, but this will probably take a while to happen, as 
most people HAVE GIF software/viewers, but lack JPEG equivalents.  
Undoubtedly, however, this too shall change, but at this point, JPEG is 
recognized as still being in its infancy.  But, if you prefer to be on the 
leading (bleeding?) edge, it is possible to get software both to view JPEG
pictures, and to convert JPEG to and from other formats, as detailed in 
part 3.

The latest and greatest info about JPEG is included in the Tom Lane's
"JPEG image compression: Frequently Asked Questions" (archive name is
"jpeg-faq"), posted on a regular basis to the,,,,
and news.answers newsgroups.

Of course, to view a picture of a particular type, you will need a viewer 
that supports that type (again, for specifics on viewers for your 
particular configuration, see part 3 of this posting).

There are other types of single-picture files posted to the net, 
although they are not as common as GIF or JPEG files.  Other than the 
difference in the viewing software, the downloading/decoding and 
encoding/uploading procedures are identical as for other types of pictures.
Platform-dependent picture types and conversion programs are discussed
in part 3 of this posting.

Occasionally people get into an argument about which standard is best. 
I think the answer is:  WHO CARES?!?  The only thing I have to say 
about this matter is that almost every machine under the sun already 
has a program written for it to view GIF files, and if yours doesn't, 
shareware or PD source code is available almost everywhere.

Commonly people post files to the net with a .GL extension.  These 
files are actually animated picture-shows that can be viewed on a small
number of system types.

Usually, GL files are huge, so people often compress them with one of
several popular compression/archiving packages.  Perhaps the most
common is the PC family's PKZIP package.  If a GL file is posted with
a .ZIP extension, you know it's been ZIP'ed.  Similarly, if it has a
.Z extension, it's been compressed with the UNIX `compress' utility. 
"Uncompression" tools of either type are available for various types of
systems - part 3 has the necessary details.

Files of a .DL extension are also sometimes posted.  These are very
similar to GL files, except in format, so of course it takes different 
software to view them (this software is also discussed in part 3).

Then there's FLI - yet another GL/DL type of file.  FLI's are generally
considered poorer quality than either GL or DL, however.

The table below lists many of the common file types for pictures or 
compression formats for different systems.  This information may be useful 
if you download a tool and then don't know how to decompress it into a 
usable form, or as a "quick reference" of file types.   Decompressors or
viewers of "unlike" system types exist on some systems - see the particular 
system information for details on this aspect.

	File extension		File  type
	--------------		----------
	ARC			ARChive (many OS's support) - compressed file(s)
	ARJ			Yet another archive format - compressed file(s)
	BMP			Windows and OS/2 BitMaP picture file
	CPT			Macintosh CompactPro compressed file.
	DIB			Windows and OS/2 BitMaP picture file
	DL			Animated picture file (system independent, for
				those with viewers)
	FLI			Animated picture file (system independent, for
				those with viewers)
	GIF			Graphics Interchange Format -
				system independent picture file
	GL			Animated picture file (system independent, for
				those with viewers)
	IMG			IMaGe - ? picture file
	JPG (JPEG)		Joint Photography experts Group - system 
				independent picture file
	LZH			Amiga LZH - compressed file(s) - LHarc output
	MAC (MACP)		Macintosh MacPaint - Macintosh picture file
	HQX			Macintosh BinHex - encoded file
	IFF			Amiga Interchangeable File Format - Amiga 
				file interchange (used for many types of binary
				data).  If it contains a picture file, then
				the picture is either an ILBM (InterLeaved 
                                BitMap), HAM (Hold-And-Modify), DHAM (DynaHAM),
				or SHAM (Sliced HAM).
	IM8 (RAST)		Sun RASTer file - Sun picture file
	PCX			IBM PC Paintbrush - IBM picture file
	PICT			Macintosh QuickDraw PICTure - Macintosh picture
	PS (PSID)		Encapsulated PostScript/PostScript Image Data - 
				printer-ready text/picture file
	RAW			RAW RGB - 24-bit system independent picture file
	SEA			Macintosh Self-Extracting Archive
	SHK			Macintosh Shrinkit - compressed file(s)
	SIT			Macintosh StuffIt - compressed file(s)
	TGA			TrueVision TarGA file - ? picture file
	TIFF			Tagged Image Format File - 24-bit system 
				independent picture file
	UUE			UNIX UUEncoding - encoded file
	XBM			X windows Bit Map - UNIX/X windows picture file
	Z			UNIX LZW "compress" - compressed file(s)
	ZIP			MS-DOS ZIP - compressed file(s)
	ZOO			MS-DOS ZOO - compressed file(s)


	Basic checklist:		Alternate checklist:
	----------------		--------------------
	UUENCODEr			"Auto-posting" tool(s)
	Editor or file splitter
	News posting software

First things first:  before you do any sort of posting, be sure you've
read and understand the a.b.p* netiquette as outlined in part 1 of this 
FAQ.  This will save you from countless flamings!

OK.  You need to UUENCODE the file.  Find an encoder and encode it!  
If the output file is particularly large (i.e. more than 60 KB), it 
would be wise to split up the encoded file into smaller parts (<= 60 KB) 
and then post those.  You can split the file with a text editor if you 
like, or check part 3 for more specifics on splitting utilities.

Now post the files... and remember to include the neat info mentioned 
in part 1, like subject lines that mean something, descriptions, 
checksums, "Cut Here" lines, etc...  

There are some very nice "super posting" utilities out there that will 
handle all the lower-level details for you.  See part 3 for more info
on these utilities.  If you don't use one, you'll obviously need to do 
all the uuencoding, splitting, and the posting of each split part
yourself - which can become quite a tedious process!  Another benefit of
the "super posters" is that they enforce some standardization on the way 
posts look - making an auto-decoder's job much easier in the process!


	Basic checklist:		Alternate checklist:
	----------------		--------------------
	Direct Internet access		E-mail software
	FTP software
	"archie" access

The pictures newsgroups are certainly not the only source for pictures, 
nor are GIF files the only types available (see section III).  The most 
likely place you are to find other pictures is in an archive that is 
reachable via FTP.  FTP stands for File Transfer Protocol, and is a 
program for transmitting files over the network.  To use FTP, you will 
need access to a computer with the FTP program, and a network connection.
Be aware that files on FTP sites will probably NOT be UUENCODED, so 
remember to transfer in binary when getting non-text files.
For the greatest level of detail on FTP and finding sources in general, you 
should refer to the posting "How to find sources (READ THIS BEFORE POSTING)",
which is periodically posted to comp.sources.wanted, alt.sources.wanted, and 
news.answers OR you can execute either 'finger' or 
'finger' to get a quick tutorial.  You can also get the
"finding sources" FAQ via anonymous FTP, available on either [] in /pub/usenet/news.answers as the file
"finding-sources.Z", on [] in /pub/NEWS.ANSWERS
as "finding-sources", on [,, or]
in /usenet/news.answers as file "finding-sources.Z".  UUCP access is done by
retrieving the file "uunet!/archive/usenet/news.answers/finding-sources". 
Lastly, you can get this FAQ by sending a message to either of with the mail body 
"send usenet/news.answers/finding-sources", or to with 
"send NEWS.ANSWERS/finding-sources" in the body of the message.
One of the useful things detailed in the "finding sources" posting mentioned 
above involves the use of the "archie" facility, which makes it very easy to 
find a program if you know its name (or just part of its name if you specify 
the "set search sub" option).  You can do this either directly by logging 
into an archie server or via e-mail.  It may take a small amount of effort - 
but it's a heck of a lot easier and faster than asking the entire 

Additionally, it is possible to get files from anonymous FTP sites via 
e-mail.  For details on this wonderful facility, send an e-mail containing 
the text "help" to  For those of you on BITNET, 
send an e-mail containing the text "help" to  
Now you too can get all sorts of great utilities from anonymous FTP sites 
using an e-mail proxy!

Due to popular demand, an anonymous FTP site list of pictures-related 
"stuff" has now been compiled and is available from bongo in 
/gifstuff/ftpsites.  This list is by no means guaranteed to be accurate
or comprehensive, but hopefully most of the information is valid.  BTW,
this list is a condensed and supplemented version of the Jan. 20, 1990 
revision of Jon Granrose's ( "List of Hosts that 
Accept Anonymous FTP Requests", which is posted regularly to comp.misc, 
comp.sources.wanted, and alt.sources.wanted, and also available via 
anonymous FTP from (  Any additions or 
corrections would be most welcome and appreciated!

Most ftp programs will allow you to enter something like 
which will connect you with the mighty SIMTEL-20 archives at the White 
Sands Missile Range.  Occasionally, you will encounter an ftp program 
that is old enough or slothful enough that it does not recognize 
internet-style addresses like the one above.  In that case, you'll 
need to know the computer's numeric address; for SIMTEL-20
you would enter

Once you're connected, you'll have to tell the computer at the other 
end that you want to log in, by entering USER (some machines save you 
this step by *assuming* you want to log in.  What else would you want 
to do?)  When you are prompted for an account name, enter
When it asks you for a password, enter *your* internet address.

Often the machine to which you are trying to connect will be busy 
(i.e. too many anonymous users), in which case the machine will inform 
you of this and throw you off.  Try again later.

Now you're in.  What do you do?  Well, you need to know where the 
files are stored that you want.  If you know this, just 
  cd directory-name
to the directory in question.  Then you can do a DIR to find out
what is in it.  

So you see a file called CRSH+BRN.GIF and you want it for yourself.  
What do you do?  Well, the first thing is to tell the computer on the 
other end that you want it to transmit a binary file.  On most FTP 
servers, entering the magic word TENEX will do this.  If the machine 
doesn't recognize TENEX, try BINARY, or if all else fails, you can 
  TYPE L 8
Be sure to do this for GIF files or you'll get garbage when you try 
to view them!
The difference between TENEX and BINARY is in translation of data type 
sizes - if your machine type has different data type sizes than the one
you're downloading from, use TENEX, otherwise use BINARY.  If you're not
sure, try TENEX first (if the command isn't recognized, you're probably 
OK).  On some VAX platforms, the keyword "IMAGE" is also sometimes used
to denote binary files.

Now you're ready to grab the files you want.  You have two options: 
you can type 
  get filename
  mget wildcard
where wildcard is any UNIX-style wildcard.  MGET will get all files 
that satisfy the specification.

When you're done grabbing files, type QUIT or BYE to log off the remote 
machine and return to yours.  Now you're ready to view the picture -
no decoding step necessary (neat, eh?)!

Most of the non-erotica pictures that appear in postings to the* hierarchy are available from anonymous FTP sites 
(again, see bongo's "ftpsites" list), but this is of course not guaranteed.  

The other most common method for obtaining files is from an archival 
file server.  Most of these work in the following way: you send mail 
to the server's address, with one-line commands in your message, like
   directory \pictures\gif\family-oriented
   send \pictures\gif\family-oriented\CRSH+BRN.GIF
and the requested info is sent back to you at some later time, when 
the server has time to get around to it.

The first step when you discover a server system is to send a HELP
command so you can learn what the commands are for that server.  
However, most servers operate with commands basically similar to those 
listed above.


	Basic checklist:
	At least one clue
	Some small level of intelligence

Well, you've downloaded the file, tried to view it, and got garbage.  
What went wrong?

The two most likely places for something to go wrong are both in the 
transmission of the file.  The first is this:  when you downloaded the 
file to your home computer, did you remember to tell the modem- 
transfer software that you're sending a binary file?

The second-most likely is that you forgot to say TENEX before you 
grabbed the file via FTP.

Either of these will result in mangled files that are unviewable by 
anything known to man.

Also: did you remember to trim off the header and trailer information if
you are/were using a "simple" uudecoder?  The symptom of forgetting to
do this is usually a message something like "short file" from your GIF
viewer.  There could also be the problem where blank lines are left 
between parts (or anywhere for that matter) within the 'begin' and 'end' 
lines of the uuencoded file.  Uudecode will get through them fine, but some
GIF viewers will choke on the results.  The only blank line I've seen
get by is the one just before the 'end' statement.  Beware of taking
too much or not enough off of the headers and trailers.

Another common problem is this one:  IBM mainframes often use an 
EBCDIC character set (yes, there's more than one EBCDIC set!) instead 
of the ASCII set used by everyone else.  This wouldn't be a problem except 
that most ASCII-EBCDIC converters have a bug which mungs the translation 
of several characters, including ^ { } and a few others.  Even this 
wouldn't be a problem except that the particular munging it does is to 
map several of these characters onto the *same* wrong character.  Ooops.

The way around this is not to use uuencode to transfer these files,
but to use xx-encode, which produces files which look almost exactly
like uu-encoded files, but they use a character set which is
IBM-proof.  If you are using an IBM mainframe as your host computer
and you're having trouble decoding files, this is most likely your
problem.  Solution: 1) find a kind soul who is willing to uudecode the
files, xxencode them and send them to you, 2) get the files via FTP,
which should be EBCDIC-proof, or 3) get a better computer that uses
everybody else's character set. :-) 

Sometimes, you need to run the "bilf" utility on a file in order to
fix it.  The "bilf" utility changes carriage-return/line-feed sets
into just carriage-return (or vice-versa).  MSDOS uses CR/LF at the
end of lines to indicate end-of-record, UNIX and VAXen use only CR.  
You might want to try running bilf before unzipping, compressing, etc.
if you are running into problems.  bilf also comes in handy when you 
are using Kermit to transmit to/from Unix and VMS.

Almost all of the problems described above can be checked by using
GIFTEST to check the GIF file's integrity on your host machine before
you download it.  I have recently added the source code for GIFTEST to
the archive at bongo.  I highly recommend that you get a copy of this,
even if you only occasionally have problems with your GIF files; it
runs in only a few seconds, and has the potential to save you hours of
download time!

The last and least likely problem is that some mailer somewhere 
actually munged the file.  It happens.  Fortunately, it doesn't happen 
all that often.  When it does (and please check all of the other 
problems *FIRST*), it's time to ask for a re-post, as detailed in part 1.


Bottom line:  It's OK to copy something (electronically or otherwise) for
your own personal use.  It's NOT OK to re-distribute that copy, whether or
not you make any money doing it.


That's about it for the "general" information.  System-specific 
information is continued in part 3 of this FAQ.  If you have any 
suggestions for things to include in future versions, don't hesitate 
to let me know...

    Jim Howard *** ***
     Author, "The Internet Voyeur" ( 
          (^:=             Flames cheerfully ignored.             =:^)
"Death is life's way of telling you you've been fired." -- R. Geis

User Contributions:

Comment about this article, ask questions, or add new information about this topic:

Part1 - Part2 - Part3

[ Usenet FAQs | Web FAQs | Documents | RFC Index ]

Send corrections/additions to the FAQ Maintainer:
Jim Howard <>

Last Update March 27 2014 @ 02:12 PM