Internet Surfer 2.0/Internet Surfer 2.0 (Wayzata Technology) (1996).iso/pc/text/mac/faqs.261

Text File | 1996-02-12 | 27.8 KB | 623 lines

Frequently Asked Questions (FAQS);faqs.261



65:  How do I get rid of the ^M junk in my Shell buffer?

  For tcsh, put this in your `.cshrc' (or `.tcshrc') file:

    if ($?EMACS) then
        if ("$EMACS" == t) then
        if ($?tcsh) unset edit
        stty nl
    endif
    endif

  Or put this in your .emacs_tcsh file:

    unset edit
    stty nl

  Alternatively, use csh in your Shell buffers instead of tcsh.  One way
  is:

    (setq explicit-shell-file-name "/bin/csh")

  and another is to do this in your .cshrc (or .tcshrc) file:

    setenv ESHELL /bin/csh

  (You must start Emacs over again with the environment variable properly
  set for this to take effect.)

66:  Why do I get `Process shell exited abnormally with code 1'?

  The most likely reason for this message is that the `env' program is not
  properly installed.  This program should be compiled (for the correct
  architecture!) and installed with execute permission for everyone in
  Emacs's program directory, which is normally /usr/local/emacs/etc.  You
  can find what this directory is at your site by inspecting the value of
  the variable exec-directory by typing "C-h v exec-directory RET".  `env'
  should also be for the correct architecture (check using `file' command).

  You should also check for other programs named `env' in your path (eg.,
  SunOS has a program named /usr/bin/env).  I don't understand why this can
  cause a failure and I don't know a general solution for working around the
  problem in this case.

  The `make clean' command will remove `env' and other vital programs, so be
  careful when using it.

  It has been reported that this sometimes happened when Emacs was started
  as an X client from an xterm window (ie. had a controlling tty) but the
  xterm was later terminated.

  See also etc/PROBLEMS for other possible causes of this message.

67:  Why can't I cut from Emacs and paste in other X programs?

  Emacs stores things you "cut" in the X "cut buffers".  It also pastes from
  the cut buffer `CUT_BUFFER0'.  This is obsolete.  Most modern X programs
  now expect to work with "selections" instead of cut buffers, although some
  like `xterm' will try to use the cut buffers if the selection is null.

  Emacs 18.58 contains a "fix" that makes xterm work by default.  This
  "fix" is that Emacs clears the `PRIMARY' selection when it stores
  something in the cut buffer.  By making the selection null, xterm will
  then fetch from the cut buffer when you try to paste.

  For versions of Emacs prior to 18.58, you can make pasting from Emacs into
  xterm work with the following X resources:

    ! Solution by Thomas Narten, should work under X11R3 and later GNU
    ! Emacs only copies to CUT_BUFFER0.  xterm by default wants to paste
    ! from the PRIMARY selection.
    XTerm*VT100.Translations: #override \
        ~Meta <Btn2Up>: insert-selection(CUT_BUFFER0,PRIMARY)

  You may have problems copying between Emacs and programs other than xterm
  that won't store cut text in the cut buffers or look in the cut buffers
  for text to paste (for backwards compatibility with obsolete applications
  like Emacs :-).  The best workaround is to use the `xcutsel' program as an
  intermediary.                                                              -

  This problem does not exist for Epoch or Lucid Emacs.

68:  Where is the termcap/terminfo entry for terminal type `emacs'?

  The termcap entry for terminal type `emacs' is ordinarily put in the
  TERMCAP environment variable of subshells.  It may help in certain
  situations (eg., using rlogin from shell buffer) to add an entry for
  `emacs' to the system-wide termcap file.  Here is a correct termcap entry
  for `emacs':

    emacs:tc=unknown:

  To make a terminfo entry for `emacs', use `tic' or `captoinfo'.  You need
  to generate /usr/lib/terminfo/e/emacs.  It may work to simply copy
  /usr/lib/terminfo/d/dumb to /usr/lib/terminfo/e/emacs.

  Having a termcap/terminfo entry will not enable the use of full screen
  programs in shell buffers.  Use M-x terminal-emulator for that instead.

  A workaround to the problem of missing termcap/terminfo entries is to
  change terminal type `emacs' to type `dumb' or `unknown' in your shell
  start up file.  `csh' users could put this in their .cshrc files:

    if ("$term" == emacs) set term=dumb

69:  Why does Emacs spontaneously start displaying `I-search:' and beeping?

  Your terminal (or something between your terminal and the computer) is
  sending C-s and C-q for flow control, and Emacs is receiving these
  characters and interpreting them as commands.  (The C-s character normally
  invokes the isearch-forward command.)  For possible solutions, see
  question 131.

70:  Why can't Emacs talk to certain hosts (or certain hostnames)?

  The problem may be that Emacs is linked with a wimpier version of
  gethostbyname than the rest of the programs on the machine.  This is often
  manifested as a message on startup of `X server not responding.  Check
  your DISPLAY environment variable.' or a message of `Unknown host' from
  open-network-stream.

  On a Sun, this may be because Emacs had to be linked with the static C
  library.  The version of gethostbyname in the static C library may only
  look in /etc/hosts and the NIS (YP) maps, while the version in the dynamic
  C library may be smart enough to check DNS in addition to or instead of
  NIS.  On a Motorola Delta running System V R3.6, the version of
  gethosbyname in the standard library works, but the one that works with
  NIS doesn't (the one you get with -linet).  Other operating systems have
  similar problems.

  Try these options:

  * Explicitly add the host you want to communicate with to /etc/hosts.

  * Relink Emacs with this line in src/config.h:

      #define LIBS_SYSTEM -lresolv

  * Replace gethostbyname and friends in libc.a with more useful versions
    such as the ones in libresolv.a.  Then relink Emacs.

  * If you are actually running NIS, make sure that `ypbind' is properly
    told to do DNS lookups with the correct command line switch.

  * Use tcp.el and tcp.c from GNUS.  This has the additional advantage that
    you can use numeric IP addresses instead of names.  open-network-stream
    currently can't handle numeric addresses.  Brian Thomson
    <[e-mail]> has a enhancement to open-network-stream to
    allow it to handle numeric addresses.

71:  Why does Emacs say `Error in init file'?

  An error occurred while loading either your .emacs file or the system-wide
  lisp/default.el file.  For information on how to debug your .emacs file,
  see question 29.

  It may be the case that you may need to load some package first, or use a
  hook that will be evaluated after the package is loaded.  A common case of
  this is explained in question 127.

72:  Why does Emacs ignore my X resources (my .Xdefaults file)?

  * Try compiling Emacs with the XBACKWARDS macro defined.  There is a bug
    in some implementations of XGetDefault, which do not correspond to the
    documentation or the header files.

  * Make sure you are either using the class name of `Emacs' (oops,
    apparently this is buggy in Emacs 18.58!) or the correct instance name.
    The instance name is normally the same as the name of the file Emacs is
    in (ie., the last part of argv[0]), but this can be overridden by -rn
    command line option or the WM_RES_NAME environment variable.

    WARNING: Reports say using the class name fails in Emacs 18.58.

    WARNING: The advice the man page gives to use `emacs' is often wrong.

    WARNING: Older versions of Emacs got the class name wrong.

  * Emacs currently ignores the -xrm command line argument.

  * Emacs does not yet handle X11R5 screen-specific resources.

  * Emacs has a bug where it ignores color specifications if running on a
    1-bit display (ie. a non-color display).

  * I don't think Emacs will use either of the application-specific resource
    files.  Thus these environment variables don't affect it: XAPPLRESDIR,
    XUSERFILESEARCHPATH, XFILESEARCHPATH.  {Correct?}

73:  Why does Emacs take 20 seconds to visit a file?

  The usual cause is that the master lock file, `!!!SuperLock!!!' has been
  left in the lock directory somehow.  Delete it.

  Mark Meuer <[e-mail]> says that NeXT NFS has a bug where an
  exclusive create succeeds but returns an error status.  This can cause the
  same problem.  Since Emacs's file locking doesn't work over NFS anyway,
  the best solution is to recompile Emacs with CLASH_DETECTION undefined.

74:  How do I edit a file with a `$' in its name?

  When entering a filename in the minibuffer, Emacs will attempt to expand
  a `$' followed by a word as an environment variable.  To suppress this
  behavior, type "$$" instead.

75:  Why does Shell mode lose track of the shell's current directory?

  Emacs has no way of knowing when the shell actually changes its directory.
  This is an intrinsic limitation of Unix.  So it tries to guess by
  recognizing `cd' commands.  If you type `cd' followed by a directory name
  with a variable reference (`cd $HOME/bin') or with a shell metacharacter
  (`cd ../lib*'), Emacs will fail to correctly guess the shell's new current
  directory.  A huge variety of fixes and enhancements to Shell mode for
  this problem have been written to handle this problem.  Check the Lisp
  Code Directory (see question 88).

76:  Why doesn't my change to load-path work?

  If you added a directory name containing a tilde (~) to your load-path,
  expecting the tilde to be interpreted as your home directory, then you
  need to do something like this:

    (setq load-path (mapcar 'expand-file-name load-path))

77:  Why does the cursor always go to the wrong column when I move up or
 down one line?

  You have inadvertently typed "C-x C-n" (set-goal-column) which sets the
  "goal column" to the column where the cursor was.  To undo this type
  "C-u C-x C-n".

  If you make this mistake frequently, you might want to unbind or disable
  this command by doing one of these two:

    (define-key ctl-x-map "\C-n" nil)
    (put 'set-goal-column 'disabled t)

78:  Why does Emacs hang with message `Unknown XMenu error' with X11R4?

  Many different X errors can produce this message.  Here is the solution
  to one problem:

  X11 Release 4 (and later, including OpenWindows) enforces some conditions
  in the X protocol that were previously allowed to pass unnoticed.  You
  need to put the X11R4 server into X11R3 bug compatibility mode for Emacs's
  Xmenu code to work.  You can do this with the command `xset bc'.

79:  Why doesn't display-time show the load average in the mode line
 anymore?

  In GNU Emacs 18.56, a change was made in the display-time code.
  Formerly, in version 18.55, Emacs used a program named `loadst' to
  notify Emacs of the change in time every minute.  loadst also sent Emacs
  the system load average if it was installed with sufficient privilege to
  get that information (or was on a system where no such privilege was
  needed).  Emacs then displayed this information in the mode line.

  In version 18.56, this code was changed to use a program named `wakeup'.
  wakeup doesn't send Emacs any information, it's only purpose is to send
  Emacs *something* every minute, thus invoking the filter function in
  Emacs once a minute.  The filter function in Emacs does all the work of
  finding the time, date, and load average.  However, getting the load
  average requires the privilege to read kernel memory on most systems.
  Since giving Emacs this privilege would destroy any security a system
  might have, for almost everyone this is not an option.  In addition,
  Emacs does not have the code built into it to get this information on
  the systems which have special system calls for this purpose, even
  though loadst had code for this.

  The solution I use is to get the files lisp/display-time.el and
  etc/loadst.c from version 18.55 and use those with 18.58.  (I have heard
  a rumor that loadst disappeared because of the legal action Unipress
  threatened against IBM.)

  WARNING:  Do not install Emacs setgid kmem unless you wish to destroy
  any security your system might have!!!!!!!!!!

  If you are using Emacs 18.55 or earlier, or already using the solution I
  describe above, read further:

  The most likely cause of the problem is that `loadst' can't read the
  special file /dev/kmem.  To properly install loadst, it should be either
  setuid to the owner of /dev/kmem, or is should be setgid to the group to
  which /dev/kmem belongs.  In either case, /dev/kmem should be readable by
  its owner or its group, respectively.  Assuming the existence of a group
  named `kmem', here is an example of how to do this:

    chgrp kmem /dev/kmem
    chmod g+r /dev/kmem
    chgrp kmem /usr/local/emacs/etc/loadst
    chmod g+s /usr/local/emacs/etc/loadst

  Another possibility is that your version of Unix doesn't have the load
  average data available in /dev/kmem.  Your version of Unix might have a
  special system call to retrieve this information (eg., inq_stats under
  UMAX), and loadst might not have been enhanced to cope with this.

80:  Why does ispell sometimes ignore the local dictionary?

  You need to update the version of Ispell to 2.0.02.  (Or you can switch to
  version 3.0 which is still in beta-testing.)  A patch is available via
  anonymous FTP:                                                             !
                                                                             !
    /archive.cis.ohio-state.edu:pub/gnu/ispell/patch2.Z                      !

  You also need to change a line in ispell.el from:

    (defconst ispell-version "2.0.01") ; Check against output of "ispell -v".

  to:

    (defconst ispell-version "2.0.02") ; Check against output of "ispell -v".

81:  Why does Ispell treat each line as a single word?

  Ispell expects to get its input one word per line.  The ispell filter,
  which is specified by the variables ispell-filter-hook and
  ispell-filter-hook-args, should output at most one word per line.

82:  Are there any security risks in GNU Emacs?

  * the `movemail' incident (No, this is not a risk.)

    Cliff Stoll in his book `The Cuckoo's Egg' describes this in chapter 4.
    The site at LBL had installed the `etc/movemail' program setuid root.
    Since `movemail' had not been designed for this situation, a security
    hole was created and users could get root priveleges.

    `movemail' has since been changed so that even if it is installed setuid
    root this security hole will not be a result.

    I have heard unverified reports that the Internet worm took advantage of
    this configuration problem.

  * the file-local-variable feature (Yes, a risk, but easy to change.)

    There is an Emacs feature that allows the setting of local values for
    variables when editing a file by including specially formatted text near
    the end of the file.  This feature also includes the ability to have
    arbitrary Emacs Lisp code evaluated when the file is visited.
    Obviously, there is a potential for Trojan horses to exploit this
    feature.

    If you set the variable inhibit-local-variables to a non-nil value,
    Emacs will display the special local variable settings of a file that
    you visit and ask you if you really want them.  This variable is not
    mentioned in the manual.

    It is wise to do this in lisp/site-init.el before building Emacs:

      (setq inhibit-local-variables t)

    If Emacs has already been built, the expression can be put in
    lisp/default.el instead, or an individual can put it in their own .emacs
    file.

    The ability to exploit this feature by sending e-mail to an RMAIL user
    was fixed sometime after Emacs 18.52.  However, any new package that
    uses find-file or find-file-noselect has to be careful about this.

    For more information, see `File Variables' in the online manual (which,
    incidentally, does not describe how to disable the feature).

    There is a new variable in Emacs 18.58 named ignore-local-eval which
    turns out to be useless as currently implemented.  Ignore it.

  * synthetic X events (Yes, a risk, use MIT-MAGIC-COOKIE-1 or better.)

    Emacs accepts synthetic X events generated by the SendEvent request as
    though they were regular events.  As a result, if you are using the
    trivial host-based authentication, other users who can open X
    connections to your X workstatation can make your Emacs process do
    anything, including run other processes with your priveleges.

    The only fix for this is to prevent other users from being able to open
    X connections.  The standard way to prevent this is to use a real
    authentication mechanism, such as MIT-MAGIC-COOKIE-1.  If using the
    `xauth' program has any effect, then you are probably using
    MIT-MAGIC-COOKIE-1.  Your site may be using a superior authentication
    method; ask your system administrator.

    If real authentication is not a possibility, you may be satisfied by
    just allowing hosts access for brief intervals while you start your X
    programs, then removing the access.  This reduces the risk somewhat by
    narrowing the time window when hostile users would have access, but DOES
    NOT ELIMINATE THE RISK.

    Lucid GNU Emacs does not accept synthetic X events unless you set a      +
    variable.                                                                +

  * autosave file permissions (Yes, a risk, hard to work around.)

    The file permissions for autosave files are determined solely by the
    Emacs process's `umask' value.  The permissions of the file being
    autosaved are not used.  The easiest workaround is to keep sensitive
    files in protected directories.  Sebastian Kremer has written an
    enhanced version of the autosave file name picking code that can avoid
    this problem by keeping autosave files in a protected directory.  {FTP
    information please?}  This problem will be fixed in Emacs 19.



Difficulties Building/Installing/Porting Emacs

83:  What should I do if I have trouble building Emacs?

  First look in the file etc/PROBLEMS to see if there is already a solution
  for your problem.  Next check the FAQ (you're reading it).  If you don't
  find a solution, then report your problem via e-mail to
  [e-mail].  Please do not post it to gnu.emacs.help or
  e-mail it to [e-mail].  For further guidelines, see
  question 8.

84:  How do I stop Emacs from failing when the executable is stripped?

  Don't do that.

  This problem has been reported on SGI Indigo machines running Irix 4.0.*
  and RS/6000 machines.  Scott Henry <[e-mail]> posted a
  patch that fixes the problem for Irix.

85:  Why does linking Emacs with -lX11 fail?                                 !
                                                                             !
  Emacs needs to be linked with the static version of the X11 library,       !
  libX11.a.  This may be missing.                                            !
                                                                             !
  Under OpenWindows, you may need to use `add_services' to add the           !
  `OpenWindows Programmers' optional software category from the CD-ROM.      -

  Under HP-UX 8.0, you may need to run `update' again to load the            -
  X11-PRG `fileset'.  This may be missing even if you specified `all
  filesets' the first time.  If libcurses.a is missing, you may need to load
  the `Berkeley Development Option' {???}.

  If you are building the MIT X11 sources, you may need to modify your       +
  `site.cf' file to get static versions of the libraries.  (Info from David  +
  Zuhn <[e-mail]>.)                                                    +
                                                                             +
  Other systems may have similar problems.  You can always define            +
  CANNOT_DUMP and link with the shared libraries instead.

  To get the Xmenu stuff to work, you need to find a copy of MIT's           +
  liboldX.a.                                                                 +

86:  Why does Emacs 18.55 say `Fatal error (6).Abort' under SunOS 4.1?

  I had hoped this question would go away after Emacs 18.57 was released,
  but people continue to compile 18.55.  Easiest solution: upgrade.

  This is a result of the SunOS localtime/tzsetwall malloc bug, which was
  (finally!) fixed in SunOS 4.1.2.  If you actually need the full
  explanation, send me e-mail.  If you absolutely must compile Emacs 18.55
  (eg., you are compiling Nemacs), the easiest workaround was to put
  `#define SYSTEM_MALLOC' in src/config.h.                                   -



Xref: bloom-picayune.mit.edu gnu.emacs.help:7401 comp.emacs:15250 news.answers:3115
Path: bloom-picayune.mit.edu!enterpoop.mit.edu!snorkelwacker.mit.edu!eff!sol.ctr.columbia.edu!caen!sdd.hp.com!think.com!news.bbn.com!bu.edu!bigbird!jbw
From: [e-mail] (Joe Wells)
Newsgroups: gnu.emacs.help,comp.emacs,news.answers
Subject: GNU Emacs FAQ (3/5, 87-124): Getting Emacs/Packages
Summary: READ BEFORE POSTING.  A regularly posted list of answers to frequently
         asked questions (FAQs) about GNU Emacs and many Emacs Lisp programs.
         Contains pointers to other resources.  Follow "References:" link for
         more metainfo.
Keywords: gnu emacs faq answers frequently asked questions periodic
Message-ID: <[e-mail]>
Date: 22 Sep 92 01:10:20 GMT
Expires: 21 Nov 92 01:10:20 GMT
References: <[e-mail]>
Sender: [e-mail]
Reply-To: [e-mail]
Followup-To: poster
Organization: GNU's Not UNIX
Lines: 810
Approved: [e-mail]
Supersedes: <[e-mail]>

Archive-Name: GNU-Emacs-FAQ/part3
Last-Modified: Tue, 22 Sep 1992 01:05:05 GMT
Last-Posted: Tue, 22 Sep 1992 01:10:20 GMT

                  GNU Emacs FAQ: Getting Emacs/Packages

If you are viewing this text in a GNU Emacs Buffer, you can type "M-2 C-x $" to
get an overview of just the questions.  Then, when you want to look at the text
of the answers, just type "C-x $".

To search for a question numbered XXX, type "M-C-s ^XXX:", followed by a C-r if
that doesn't work, then type ESC to end the search.

A `+' in the 78th column means something was inserted on the line.  A `-' means
something was deleted and a `!' means some combination of insertions and
deletions occurred.

Full instructions for getting the latest FAQ are in question 22.  Also see the
`Introduction to news.answers' posting in the `news.answers' newsgroup, or send
e-mail to `[e-mail]' with `help' on a body line, or use FTP,
WAIS, or Prospero to rtfm.mit.edu.



Finding/Getting Emacs and Related Packages

87:  Where can I get GNU Emacs on the net (or by snail mail)?

  Look in the files etc/DISTRIB and etc/FTP for information on nearby
  archive sites.  If you don't already have GNU Emacs, see question 20
  for how to get these two files.

  The latest version is always available via anonymous FTP at MIT:           !
                                                                             !
    /prep.ai.mit.edu:pub/gnu/emacs-18.58.tar.Z                               !

  See question 91.

88:  How do I find a GNU Emacs Lisp package that does XXX?

  A listing of Emacs Lisp packages, called the Lisp Code Directory, is being
  maintained by Dave Brennan <[e-mail]> and Dave Sill <[e-mail]>.
  You can search through this list to find if someone has written something
  that fits your needs.

  This list is file LCD-datafile.Z in the Emacs Lisp Archive.  (See
  question 89 for methods for getting this file.)  The files lispdir.el.Z
  and lispdir.doc.Z in the archive contain information to help you use the
  list.  Once you have installed lispdir.el and LCD-datafile, then you can
  use the "M-x lisp-dir-apropos" command to look things up in the database.
  For example, the command "M-x lisp-dir-apropos RET ange-ftp RET" produces
  this (outdated) output:

            GNU Emacs Lisp Code Apropos -- "ange-ftp"

    ange-ftp (3.112)  91-08-12
      Andy Norman, <[e-mail]>
      archive.cis.ohio-state.edu:
        /pub/gnu/emacs/elisp-archive/as-is/ange-ftp.el.Z
      transparent FTP Support for GNU Emacs

89:  Where can I get GNU Emacs Lisp packages that don't come with Emacs?

  First, check the Lisp Code Directory to find the name of the package you
  are looking for.  (See question 88).  Then check local archives and
  the Emacs Lisp Archive to find a copy of the relevant files.  Then, if
  you still haven't found it, you can send e-mail to the author asking for
  a copy.

  You can access the Emacs Lisp Archive via anonymous FTP:                   !
                                                                             !
    /archive.cis.ohio-state.edu:/pub/gnu/emacs/elisp-archive/                !

  Fetch the file README first.

  NOTE: The archive maintainers do not have time to answer individual
  requests for packages or the list of packages in the archive.  If you
  cannot use FTP or UUCP to access the archive yourself, try to find a
  friend who can, but please don't ask the maintainers.

  NOTE: Any files with names ending in `.Z' are compressed, and you should
  use `binary' mode in FTP to retrieve them.  You should also use binary
  mode whenever you retrieve any files with names ending in `.elc'.

90:  How do I submit code to the Emacs Lisp Archive?

  Submissions should be mailed to [e-mail].  The
  lispdir.el package has a function named submit-lcd-entry which will help
  you with this.  Mail messages (submissions) are automatically saved and
  periodically archived.  Urgent mail may be sent directly to Dave Sill
  <[e-mail]> or Dave Brennan <[e-mail]> or should contain the
  string `urgent' in the subject.  The incoming ftp directory is no longer
  available at the request of Ohio State.  {Is this still true?}

  However, if someone has a submission with multiple files (which would be
  archived as a tar file) or binary files, then FTP transfer is preferred
  and can be arranged via an anonymous FTP site.  This is faster than
  uudecoding, unsharing, etc., and re-packaging files.

  Before submitting anything, please read the file guidelines.Z, which is
  available in the archive.  Whenever possible, submissions should contain a
  complete LCD entry since this helps reduce administrative overhead for the
  maintainers.  You can include an entry in this format:

    ;; LCD Archive Entry:
    ;; package name|author's name|email address
    ;; |description
    ;; |date|version|archive path

  For example:

    ;; LCD Archive Entry:
    ;; tex-complete|Sebastian Kremer|[e-mail]
    ;; |Minibuffer name completion for editing [La]TeX.
    ;; |91-03-26|$Revision: 1.12 $|~/packages/tex-complete.el.Z

  Dave Brennan has software which automatically looks for data in this
  format.  The format is fairly flexible.  The entry ends when a line is
  reached with a different prefix or the seventh field terminator is
  seen.

  If you are submitting a multi-file submission you should include a file
  named "LCD-entry" which contains the archive entry, instead of placing
  it in one or more of the individual files.

91:  Where can I get other up-to-date GNU stuff?

  The most up-to-date official GNU stuff is normally kept on prep.ai.mit.edu
  and is available for anonymous FTP in the pub/gnu directory.  See the
  files etc/DISTRIB and etc/FTP for more information.  (To get copies of
  these files, see question 20.)

  The following sites are all mirror images of the GNU distribution area:

    /prep.ai.mit.edu:pub/gnu/                                                +
    /ftp.uu.net:packages/gnu/                                                +
    /src.doc.ic.ac.uk:gnu/ (available via FTP, NIFTP, FTAM)                  +
    /ftp.win.tue.nl:pub/gnu/                                                 !
    /utsun.s.u-tokyo.ac.jp:ftpsync/prep/                                     +
    /nic.funet.fi:pub/gnu/                                                   +