the original and Next Generation shows, but the cartoon version as well),
Lost in Space, Battlestar Galactica, the Twilight Zone, the Prisoner and
Doctor Who. There are also lists of various things related to science
fiction and an online science-fiction fanzine. 6 p.m. - 6 a.m.
Shakespeare
-----------
`atari.archive.umich.edu' The shakespeare directory contains most of the
Bard's works. A number of other sites have his works as well, but generally
as one huge mega-file. This site breaks them down into various categories
(comedies, poetry, histories, etc.) so that you can download individual plays
or sonnets.
Space
-----
`ames.arc.nasa.gov' Stores text files about space and the history of the
NASA space program in the `pub/SPACE' subdirectory. In the `pub/GIF' and
`pub/SPACE/GIF' directories, you'll find astronomy- and NASA-related GIF
files, including pictures of planets, satellites and other celestial objects.
9 p.m. - 9 a.m.
Spain
-----
`goya.dit.upm.es' This Spanish site carries an updated list of
bulletin-board systems in Spain, as well as information about European
computer networks, in the `info/doc/net' subdirectory, mostly in Spanish.
The BBS list is `bbs.Z', which means you will have to uncompress it to read
it. Available 24 hours.
TeX
---
`ftp.tex.ac.uk' in `pub/archive', `ftp.uni-stuttgart.de' in `tex-archive',
and `ftp.shsu.edu' in `soft/tex' form the CTAN (comprehensive TeX archive
network), that always has the latest TeX version (and everything that comes
with it) available. They are continuously updated, i.e. they are "mirrors"
of the primary TeX archive at Stanford University.
TV
--
`coe.montana.edu' The `pub/TV/Guides' directory has histories and other
information about dozens of TV shows. Only two anonymous-ftp log-ins are
allowed at a time, so you might have to try more than once to get in. 8 p.m.
- 8 a.m.
`ftp.cs.widener.edu' The `pub/simpsons' directory has more files than
anybody could possibly need about Bart and family. The `pub/strek' directory
has files about the original and Next Generation shows as well as the movies.
See also under Science Fiction *Note Science Fiction::.
Travel
------
`nic.stolaf.edu' Before you take that next overseas trip, you might want
to see whether the State Department has issued any kind of advisory for the
countries on your itinerary. The advisories, which cover everything from
hurricane damage to civil war, are in the `pub/travel-advisories/ advisories'
directory, arranged by country. 7 p.m. - 7 a.m.
Usenet
------
`ftp.uu.net' In the usenet directory, you'll find "frequently asked
questions" files, copied from `pit-manager.mit.edu'. The communications
directory holds programs that let MS-DOS users connect directly with UUCP
sites. In the info directory, you'll find information about ftp and ftp
sites. The inet directory contains information about Internet. Available 24
hours.
`pit-manager.mit.edu' This site contains all available FAQs "frequently
asked questions" files for Usenet newsgroups in the `pub/usenet' directory.
For easy access, get the `index' file. See under Books *Note Books:: for a
caveat in using this ftp site. 6 p.m. - 6 a.m.
Viruses
-------
`ftp.unt.edu' The antivirus directory has anti-virus programs for MS-DOS
and Macintosh computers. 7 p.m. - 7 a.m.
Weather
-------
`vmd.cso.uiuc.edu' No password needed. The wx directory contains GIF
weather images of North America. Files are updated hourly and take this
general form: `CV100222'. The first two letters tell the type of file: CV
means it is a visible-light photo taken by a weather satellite. CI images
are similar, but use infrared light. Both these are in black and white.
Files that begin with SA are color radar maps of the U.S. that show severe
weather patterns but also fronts and temperatures in major cities. The
numbers indicate the date and time (in GMT - five hours ahead of EST) of the
image: the first two numbers represent the month, the next two the date, the
last two the hour. The file `WXKEY.GIF' explains the various symbols in SA
files.
X Windows
---------
`ftp.x.org' The default server for MIT's X Window System. See under `/pub'.
`ftp.germany.eu.net' Germany's backbone site located at the University of
Dortmund, in the European part of the Internet; the so-called EUnet. It's
also Germany's default server for X window system releases, and also
"mirrors" several important sites; e.g. in `pub/packages/gnu' the GNU
project's default server. Furthermore you'll find "mirrors" of `386BSD',
`NetBSD', and `Linux'. Available 24 hours.
=====================
* You get a "host unavailable" message. The ftp site is down for some
reason. Try again later.
* You get a "host unknown" message. Check your spelling of the site name.
* You misspell "anonymous" when logging in and get a message telling you a
password is required for whatever you typed in. Type something in, hit
enter, type bye, hit enter, and try again.
FYI:
====
Liberal use of archie will help you find specific files or documents. For
information on new or interesting ftp sites, try the `comp.archives'
newsgroup on Usenet. You can also look in the `comp.misc',
`comp.sources.wanted' or `news.answers' newsgroups on Usenet for lists of ftp
sites posted every month by TOM CZARNIK and JON GRANROSE.
The `comp.archives' newsgroup carries news of new ftp sites and
interesting new files on existing sites.
In the `comp.virus' newsgroup on Usenet, look for postings that list ftp
sites carrying anti-viral software for Amiga, MS-DOS, Macintosh, Atari and
other computers.
The `comp.sys.ibm.pc.digest' and `comp.sys.mac.digest' newsgroups provide
information about new MS-DOS and Macintosh programs as well as answers to
questions from users of those computers.
*"Welch ein Ort zum Pluendern!"
(What a place to plunder!)*
-- General Gebhard Leberecht von Bluecher
*************************************
Even with tools like Hytelnet and archie, telnet and ftp can still be
frustrating. There are all those telnet and ftp addresses to remember.
Telnet services often have their own unique commands. And, oh, those weird
directory and file names!
But now that the Net has become a rich repository of information, people
are looking at ways to make it far easier to find all that data. Gophers and
Wide-Area Information Servers (WAISs) are two programs that could ultimately
make the Internet as easy to navigate as commercial networks like CompuServe
or Prodigy.
Both programs essentially take a request for information and then scan the
Net for it, so you don't have to. Both also work through menus - instead of
typing in some long sequence of characters, you just move a cursor to your
choice and hit enter. Newer gophers even let you select files and programs
from ftp sites this way.
Let's look at gophers first.
Many public-access sites now have gophers online. To use one, type
gopher
at the command line and hit enter. If you know your site does not have a
gopher, or if nothing happens when you type that, telnet to
consultant.micro.umn.edu
At the log-in prompt, type
gopher
and hit enter. You'll be asked what type of terminal emulation you're
using, after which you'll see something like this:
Internet Gopher Information Client v1.03
Root gopher server: gopher.micro.umn.edu
--> 1. Information About Gopher/
2. Computer Information/
3. Discussion Groups/
4. Fun & Games/
5. Internet file server (ftp) sites/
6. Libraries/
7. News/
8. Other Gopher and Information Servers/
9. Phone Books/
10. Search lots of places at the U of M <?>
11. University of Minnesota Campus Information/
Press ? for Help, q to Quit, u to go up a menu Page: 1/1
Gophers are great for exploring. Just keep making choices to see what
pops up. Play with it; see where it takes you. Some choices will be
documents. When you read one of these and either come to the end or hit a
lower-case `q' to quit reading it, you'll be given the choice of saving a
copy to your home directory or e-mailing it to yourself. Other choices are
simple databases that let you enter a word to look for in a particular
database.
Notice that one of your choices is "Internet file server (ftp) sites."
Choose this, and you'll be connected to a modified archie program - an archie
with a difference. When you search for a file through a gopher archie,
you'll get a menu of sites that have the file you're looking for, just as
with the old archie. Only now, instead of having to write down or remember
an ftp address and directory, all you have to do is position the cursor next
to one of the numbers in the menu and hit enter. You'll be connected to the
ftp site, from which you can then choose the file you want, again just by
making a choice in a menu.
You'll be asked for a name in your home directory to use for the file,
after which the file will be copied to your home system. Unfortunately, this
file-transfer process does not yet work with all public-access sites for
computer programs and compressed files. If it doesn't work with yours, you'll
have to get the file the old-fashioned way, via ftp.
The letter u is an important one to remember while navigating a gopher -
it moves you back up a gopher directory tree, much like cd .. on an ftp site.
In addition to ftp sites, there are now scores of databases and libraries
around the world accessible through gophers. There is not yet a common
gopher interface for library catalogs, so be prepared to follow the online
directions more closely when you use gopher to connect to one.
Some gopher menu choices will end with a <?>. This means that if you
select it, you'll be starting up a simple database that can search through the
given service by keyword.
So many services are now available through gophers, that finding what you
want has become difficult. Fortunately, you can use veronica, a laboriously
constructed acronym that does for "gopherspace" what archie (there is no
betty, yet) did for files. You'll usually find veronicas (there are now
several) under "Other gopher and information services." When you call up a
veronica, tell her (it?) the keyword or words you're interested in, and she/it
will search all available databases for it. For example, say you want to
impress company tonight and make cherries flambe. If you were to type in
"flambe" after calling up veronica, you would soon get a menu listing several
flambe recipes, including one called "dessert flambe." Put your cursor on
that line of the menu and hit enter, and you'll find it's a menu for cherries
flambe. Then hit your q key to quit, and gopher will ask you if you want to
save the file in your home directory on your public-access site or whether you
want to e-mail it somewhere.
========
Veronica is a "Very Easy Rodent-Oriented Net-wide Index to Computerized
Archives." Veronica's name is a play on the concepts of both gopher and
archie. (Remember the comic book couple Archie and Veronica? Veronica does
for gopher what archie does for anonymous FTP.) Veronica searches through
hundreds of gopher holes looking for anything that matches a keyword supplied
by the user, and assembles a list of gopher servers that contain items of
interest. Note: veronica checks *titles* of gopher items only, not their
contents.
At present, there are no veronica clients; veronica is a gopher tool. An
informal veronica FAQ is posted regularly in `comp.infosystems.gopher' and
archived on `veronica.scs.unr.edu' as `veronica/veronica-faq'.
=============================
Now you know there are hundreds of databases and library catalogs you can
search through. But as you look, you begin to realize that each seems to
have its own unique method for searching. If you connect to several, this
can become a pain. Gophers reduce this problem somewhat.
Wide-area information servers promise another way to zero in on
information hidden on the Net. In a WAIS, the user sees only one interface -
the program worries about how to access information on dozens, even hundreds,
of different databases. You tell give a WAIS a word and it scours the net
looking for places where it's mentioned. You get a menu of documents, each
ranked according to how relevant to your search the WAIS thinks it is.
Like gophers, WAIS "client" programs can already be found on many
public-access Internet sites. If it does, type
swais
at the command line and hit enter (the "s" stands for "simple"). If it
doesn't, telnet to bbs.oit.unc.edu, which is run by the University of North
Carolina At the "login:" prompt, type
bbs
and hit enter. You'll be asked to register and will then get a list of
"bulletins," which are various files explaining how the system works. When
done with those, hit your Q key and you'll get another menu. Hit 4 for the
"simple WAIS client," and you'll see something like this:
SWAIS Source Selection Sources: 23#
Server Source Cost
001: [ archie.au] aarnet-resource-guide Free
002: [ archive.orst.edu] aeronautics Free
003: [nostromo.oes.orst.ed] agricultural-market-news Free
004: [sun-wais.oit.unc.edu] alt-sys-sun Free
005: [ archive.orst.edu] alt.drugs Free
006: [ wais.oit.unc.edu] alt.gopher Free
007: [sun-wais.oit.unc.edu] alt.sys.sun Free
008: [ wais.oit.unc.edu] alt.wais Free
009: [ archive.orst.edu] archie-orst.edu Free
010: [ archie.au] archie.au-amiga-readmes Free
011: [ archie.au] archie.au-ls-lRt Free
012: [ archie.au] archie.au-mac-readmes Free
013: [ archie.au] archie.au-pc-readmes Free
014: [ pc2.pc.maricopa.edu] ascd-education Free
015: [ archie.au] au-directory-of-servers Free
016: [ cirm2.univ-mrs.fr] bib-cirm Free
017: [ cmns-sun.think.com] bible Free
018: [ zenon.inria.fr] bibs-zenon-inria-fr Free
Keywords:
selects, w for keywords, arrows move, searches, q quits, ?
Each line represents a different database (the .au at the end of some of
them means they are in Australia; the .fr on the last line represents a
database in France). And this is just the first page! If you type a capital
K, you'll go to the next page (there are several pages). Hitting a capital J
will move you back a page.
The first thing you want to do is tell the WAIS program which databases
you want searched. To select a database, move the cursor bar over the line
you want (using your down and up arrow keys) and hit your space bar. An
asterisk will appear next to the line number. Repeat this until you've
selected all of the databases you want searched. Then hit your W key, after
which you'll be prompted for the key words you're looking for. You can type
in an entire line of these words - separate each with a space, not a comma.
Hit return, and the search begins.
Let's say you're utterly fascinated with wheat. So you might select
agricultural-market-news to find its current world price. But you also want
to see if it has any religious implications, so you choose the Bible and the
Book of Mormon. What do you do with the stuff? Select recipes and
usenet-cookbook. Are there any recent Supreme Court decisions involving the
plant? Chose supreme-court. How about synonyms? Try roget-thesaurus and
just plain thesaurus.
Now hit w and type in wheat. Hit enter, and the WAIS program begins its
search. As it looks, it tells you whether any of the databases are offline,
and if so, when they might be ready for a search. In about a minute, the
program tells you how many hits it's found. Then you get a new menu, that
looks something like this:
Keywords:
# Score SourceTitleLines
001: [1000] (roget-thesaurus) #465. [results of comparison. 1] Di 19
002: [1000] (roget-thesaurus) #609. Choice. -- N. choice, option; 36
003: [1000] (roget-thesaurus) #465. [results of comparison. 1] Di 19
004: [1000] (roget-thesaurus) #609. Choice. -- N. choice, option; 36
005: [1000] (recipes) aem@mthvax Re: MONTHLY: Rec.Food.Recipes 425
006: [1000] ( Book_of_Mormon) Mosiah 9:96
007: [1000] ( Book_of_Mormon) 3 Nephi 18:185
008: [1000] (agricultural-ma) Re: JO GR115, WEEKLY GRAIN82
009: [ 822] (agricultural-ma) Re: WA CB351 PROSPECTIVE PLANTINGS 552
010: [ 800] ( recipes) kms@apss.a Re: REQUEST: Wheat-free, Suga 35
011: [ 750] (agricultural-ma) Re: WA CB101 CROP PRODUCTION258
012: [ 643] (agricultural-ma) Re: SJ GR850 DAILY NAT GRN SUM72
013: [ 400] ( recipes) pat@jaamer Re: VEGAN: Honey Granola63
014: [ 400] ( recipes) jrtrint@pa Re: OVO-LACTO: Sourdough/Trit 142
Each of these represents an article or citing that contains the word wheat,
or some related word. Move the cursor bar (with the down and up arrow keys)
to the one you want to see, hit enter, and it will begin to appear on your
screen. The "score" is a WAIS attempt to gauge how closely the citing
matches your request. Doesn't look like the Supreme Court has had anything
to say about the plant of late!
Now think of how much time you would have spent logging onto various
databases just to find these relatively trivial examples. But as more
databases are added to WAIS programs, a problem arises that is similar to the
one WAISs were supposed to solve: how do you find the specific databases you
want? Scrolling through page after page of database listings becomes rather
tedious rather quickly and you could wind up missing the one database you
really need. That's the next step in WAIS research.
==============
Developed by researchers at CERN, the European Particle Physics Laboratory
in Geneva, the World-Wide Web project is somewhat similar to a WAIS. But
it's designed on a system known as hypertext. Words in one document are
"linked" to other documents. It's sort of like sitting with an encyclopedia
- you're reading one article, see a reference that intrigues you and so you
flip the pages to look up that reference.
To take a walk on The Web try the WWW's (or W3 in the project's own
jargon), default telnet site:
info.cern.ch
No log in is needed. When you connect, you'll see:
Overview of the Web
GENERAL OVERVIEW
There is no "top" to the World-Wide Web. You can look at it from many points
of view. If you have no other bias, here are some places to start:
by Subject[1] A classification by subject of interest. Incomplete
but easiest to use.
by Type[2] Looking by type of service (access protocol, etc) may
allow to find things if you know what you are looking
for.
About WWW[3] About the World-Wide Web global information sharing
project
Starting somewhere else
To use a different default page, perhaps one representing your field of
interest, see "customizing your home page"[4].
What happened to CERN?
1-6, Up, for more, Quit, or Help:
Ok. Now type `3', and get the following screen:
The World Wide Web project
WORLD WIDE WEB
The WorldWideWeb (W3) is a wide-area hypermedia[1] information retrieval
initiative aiming to give universal access to a large universe of documents.
Everything there is online about W3 is linked directly or indirectly to this
document, including an executive summary[2] of the project, an illustrated
talk[3] , Mailing lists[4] , Policy[5] and Conditions[6] , May's W3 news[7]
, Frequently Asked Questions[8] .
What's out there?[9] Pointers to the world's online information,
subjects[10] , W3 servers[11] , etc.
WWW Software Products[12]
What there is and how to get it: clients, servers and
tools.
Technical[13] Details of protocols, formats, program internals etc
Bibliography[14] Paper documentation on W3 and references. Also:
manuals[15] .
1-20, Back, Up, for more, Quit, or Help:
You navigate the web by typing the number next to a given reference. So
if you want to know more about the web, hit 2. This is another system that
bears playing with.
=======
If you are used to plain-vanilla Unix or MS-DOS, then the way these
gophers and WAISs work seems quite straightforward. But if you're used to a
computer with a graphical interface, such as a Macintosh, an IBM compatible
with Windows or a Next, you'll probably regard their interfaces as somewhat
primitive.
There are, however, ways to integrate these services into your graphical
user interface. In fact, there are now ways to tie into the Internet
directly, rather than relying on whatever interface your public-access system
uses.
There is now a growing number of these "client" programs for everything
from ftp to gopher. PSI of Reston, Va., which offers nationwide Internet
access, in fact, requires its customers to use these programs.
Using protocols known as SLIP and PPP, these programs communicate with the
Net using the same basic data packets as much larger computers online.
Beyond integration with your own computer's "desktop," client programs let
you do more than one thing at once on the net - while your downloading a
large file in one window, you can be chatting with a friend through an
Internet chat program in another.
These client programs have a couple of disadvantages. One is that you'll
need a 9600-baud modem - while it is possible to connect to the Net with them
at lower speeds, you will likely find them painfully slow. Not all
public-access sites are set up to allow such connections. And those that are
usually charge far more for them.
Your system administrator can give you more information on setting up one
of these connections.
FYI:
====
See the Usenet newsgroups `comp.infosystems.*':
`comp.infosystems.gopher', `comp.infosystems.wais', and
`comp.infosystems.www' are places to go for technical discussions about
Gopher, WAISs, and the World-Wide Web project respectively. Moreover there
are `comp.infosystems' for more general discussion of related issues. The
group `comp.infosystems.gis' relates to Geographic Information Systems, and
thus is more specialized on this subject.
There even exists a Gopher service to read Usenet news: `gopher
gopher.msu.edu 4320'. But, the lines behind this service are few, and thus
it's likely that you get the following message, when trying to enter:
We are sorry, but our Usenet News gateway limits the number of
simultaneous connections. If you were attempting to read news and were
instead directed to this file, all of those connections are in use. We
offer this gateway as a "last resort" for people who have no other
access for reading Usenet. We do not have the capacity to serve as the
Usenet gateway for large numbers of users around the Internet.
Individuals who like this style of access should ask their Internet
service providers to offer the same sort of gateway on their local
Gopher server. Individuals and campuses should consider installing
local news feeds and local news readers (such as RN, NN, TIN, or
Trumpet) so that users can read and post to Usenet newsgroups
conveniently.
For system administrators: the software we use to implement this gateway
is the go4gw Gopher gateway software from Roland Schemers of Stanford
University. This software should be available by anonymous ftp from
boombox.micro.umn.edu, somewhere under /pub/gopher.
-- The Michigan State University Gopher Team
*"Reliable information is the basis of successful planning."*
-- Christoph Columbus
***************
E-mail by itself is a powerful tool, and by now you may be sending e-mail
messages all over the place. You might even be on a mailing list or two. But
there is a lot more to e-mail than just sending messages. If your host
system does not have access to ftp, or it doesn't have access to every ftp
site on the Net, you can have programs and files sent right to your mailbox.
And using some simple techniques, you can use e-mail to send data files such
as spreadsheets, or even whole programs, to friends and colleagues around the
world.
A key to both is a set of programs known as encoders and decoders. For
all its basic power, Net e-mail has a big problem: it can't handle graphics
characters or the control codes found in even the simplest of computer
programs. Encoders however, can translate these into forms usable in e-mail,
while decoders turn them back into a form that you can actually use. If you
are using a Unix-based host system, chances are it already has an encoder and
decoder online that you can use. These programs will also let you use
programs posted in several Usenet newsgroups, such as `comp.binaries.ibm.pc'.
To help people without ftp access, a number of ftp sites have set up mail
servers (also known as archive servers) that allow you to get files via
e-mail. You send a request to one of these machines and they send back the
file you want. As with ftp, you'll be able to find everything from
historical documents to software (but please note that if you do have access
to ftp, that method is always quicker and ties up fewer resources than using
e-mail).
Some interesting or useful mail servers include:
Files of "frequently asked questions" related to Usenet; state-by-state
lists of U.S. representatives and Senators and their addresses and
office phone numbers.
Information about the Electronic Frontier Foundation; documents about
legal issues on the Net.
Back copies of the Computer Underground Digest and every possible fact
you could want to know about "The Simpsons."
Programs for many types of personal computers; archives of past postings
from many Usenet newsgroups.
Space-related text and graphics (GIF-format) files.
Detailed information about Internet.
Most mail servers work pretty much the same - you send an e-mail message
that tells them what file you want and how you want it sent to you. The most
important command is "send," which tells the computer you want it to send you
a particular file.
First, though, you'll need to know where the mail server stores that file,
because you have to tell it which directory or sub-directory it's in. There
are a couple of ways to do this. You can send an e-mail message to the
archive-server that consists of one line:
index
The server will then send you a directory listing of its main, or root
directory. You'll then have to send a second message to the archive server
with one line:
index directory/subdirectory
where that is the directory or directory path for which you want a
listing. An alternative is to send an e-mail message to our old friend
archie, which should send you back the file's exact location on the
archive-server (along with similar listings for all the other sites that may
have the file, however)
Once you have the file name and its directory path, compose a message to
the archive server like this:
send directory/subdirectory/file
Send off the message and, anywhere from a few minutes to a couple of days
later, you'll find a new message in your mailbox: a copy of the file you
requested. The exact time it will take a file to get to you depends on a
variety of factors, including how many requests are in line before yours
(mail servers can only process so many requests at a time) and the state of
the connections between the server and you.
Seems simple enough. It gets a little more complicated when you request a
program rather than a document. Programs or other files that contain unusual
characters or lines longer than 130 characters (graphics files, for example)
require special processing by both the mail server to ensure they are
transmitted via e-mail. Then you'll have to run them through at least one
converter program to put them in a form you can actually use. To ensure that
a program or other "non-mailable" file actually gets to you, include another
line in your e-mail message to the server:
encoder
This converts the file into an encoded form. To decode it, you'll first
have to transfer the file message into a file in your home directory. If you
are using the simple mail program, go into mail and type
w # file.name
where `#' is the number of the message you want to transfer and file.name
is what you want to call the resulting file. In pine, call up the message
and hit your `O' key and then `E'. You'll then be asked for a file name. In
elm, call up the message and hit your `S' key. You'll get something that
looks like this:
=file.request
Type a new file name and hit enter (if you hit enter without typing a file
name, the message will be saved to another mail folder, not your home
directory). Exit mail to return to your host system's command line. Because
the file has been encoded for mail delivery, you now have to run a decoder.
At the command line, type
uudecode file.name
where `file.name' is the file you created while in mail. Uudecode will
create a new, uncompressed file. In some cases, you may have to run it
through some other programs (for example, if it is in "tar" form), but
generally it should now be ready for you to download to your own computer.
One further complication comes when you request a particularly long file.
Many Net sites can only handle so much mail at a time. To make sure you get
the entire file, tell the mail server to break it up into smaller pieces,
with another line in your e-mail request like this:
size 100000
This gives the mail server the maximum size, in bytes, of each file
segment. This particular size is good for UUCP sites. Internet and Bitnet
sites can generally go up to 300000. When you get all of these files in
mail, transfer them to your home directory. Exit mail and call up each file
in your host system's text processor and delete each one's entire header and
footer (or "signature" at the end). When done with this, at your host
system's command line, type
cat file1 file2 >bigfile
where file1 is the first file, file2 the second file, and so on. The >
tells your host system to combine them into a new megafile called bigfile (or
whatever you want to call it). You can then run uudecode, tar, etc. One word
of caution, though: if the file you want is long enough that it has to be
broken into pieces, think of how much time it's going to take you to download
the whole thing - especially if you're using a 2400-baud modem!
There are a number of other mail servers. To get a list, send an e-mail
message to:
send usenet/comp.sources.wanted/How_to_find_sources_(READ_THIS_BEFORE_POSTING)
You'll have to spell it exactly as listed above. Some mail servers use
different software, which will require slightly different commands than the
ones listed here. In general, if you send a message to a mail server that
says only
help
you should get back a file detailing all of its commands.
But what if the file you want is not on one of these mail servers? That's
where ftpmail comes in. Run by Digital Equipment Corp. in California, this
service can connect to almost any ftp site in the world, get the file you
want and then mail it to you. Using it is fairly simple - you send an e-mail
message to ftpmail that includes a series of commands telling the system
where to find the file you want and how to format it to mail to you.
Compose an e-mail message to
ftpmail@decwrl.dec.com
Leave the "subject:" line blank. Inside the message, there are several
commands you can give. The first line should be
reply address
where "address" is your e-mail address. The next line should be
connect host
where "host" is the system that has the file you want (for example:
wuarchive.wustl.edu). Other commands you should consider using are "binary"
(required for program files); "compress" (reduces the file size for quicker
transmission) and "uuencode" (which encodes the file so you can do something
with it when it arrives). The last line of your message should be the word
"quit".
Let's say you want a copy of the U.S. constitution. Using archie, you've
found a file called, surprise!, `constitution', at the ftp site
`archive.cis.ohio-state.edu', in the `/pub/firearms/politics/ rkba'
directory. You'd send a message to that looks like
this:
reply adamg@world.std.com
connect archive.cis.ohio-state.edu
binary
compress
uuencode
get pub/firearms/politics/rkba/constitution
quit
When you get the file in your mailbox, use the above procedure for copying
it to a file. Run it through uudecode. Then type
uncompress file.name
to make it usable.
Since this was a text file, you could have changed the "binary" to "ascii"
and then eliminated the "uuencode" file. For programs, though, you'll want
to keep these lines.
=======================================
The uuencode and uudecode programs will also come in handy if you ever
want to send your own files to somebody else.
If both you and your intended recipient communicate via Unix-based host
systems, then it's pretty easy, because almost all Unix host systems will
have encoder/decoder programs online.
First, upload the file you want to send to your friend to your host site.
Ask your system administrator how to upload a file to your name or "home"
directory. Then type
uuencode file file >file.uu
and hit enter. "File" is the name of the file you want to prepare for
mailing, and yes, you have to type the name twice! The > is a Unix command
that tells the system to call the "encoded" file "file.uu" (you could
actually call it anything you want).
Now to get it into a mail message. The quick and dirty way is to type
mail friend
where "friend" is your friend's address. At the subject line, tell her
the name of the enclosed file. When you get the blank line, type
~r file.uu
or whatever you called the file, and hit enter. (on some systems, the `~'
may not work; if so, ask your system administrator what to use). This
inserts the file into your mail message. Hit control-D, and your file is on
its way!
On the other end, when your friend goes into his mailbox, she should
transfer it to her home directory. Then your friend should type
uudecode file.name
and hit enter. This creates a new file in her name directory with
whatever name you originally gave it. She can then download it to her own
computer. Before she can actually use it, though, she'll have to open it up
with a text processor and delete the mail header that has been "stamped" on
it. If you use a mailer program that automatically appends a "signature,"
tell her about that so she can delete that as well.
But what if your friend only connects with a non-Unix system, such as
CompuServe or MCIMail? There are programs available for MS-DOS, Apple and
Amiga computers that will encode and decode files. Of course, since you
can't send one of these programs to them via e-mail (how would they un-encode
it?), you'll have to mail or give them a diskette with the program on it
first. Then, they can get their message, run it through a text editor to
delete the header, and finally decode the file. If they want to send you
files in return, they'll also want an encoder
For MS-DOS machines, you'll want to get `uunecode.com' and `uudecode.com'.
Both can be found through anonymous ftp at `wuarchive.wustl.edu' in the
`/mirrors/msdos/starter' directory. The MS-DOS version is as easy to use as
the Unix one: Just type
uudecode filename.ext
and hit enter.
Mac users should get a program called uutool, which can be found in the
`info-mac/util' directory on `sumex-aim.stanford.edu'.
Once again, be careful with large files. Although large sites connected
directly to the Internet can probably handle mega-files, many smaller systems
cannot. Some commercial systems, such as CompuServe and MCIMail limit the
size of mail messages their users can receive. Fidonet doesn't even allow
encoded messages. In general, a file size of 30,000 or so bytes is a safe
upper limit for non-Internet systems.
One other thing you can do through e-mail is consult with the Usenet
Oracle. You can ask the Oracle anything at all and get back an answer
(whether you like the answer is another question).
First, you'll want to get instructions on how to address the Oracle (he,
or she, or it, is very particular about such things and likes being addressed
in august, solemn and particularly sycophantic tones). Start an e-mail
message to
oracle@iuvax.cs.indiana.edu
In the "subject:" line, type
help
and hit enter. You don't actually have to say anything in the message
itself - at least not yet. Hit control-D to send off your request for help.
Within a few hours, the Oracle will mail you back detailed instructions.
It's a fairly long file, so before you start reading it, turn on your
communications software's logging function, to save it to your computer (or
save the message to a file on your host system's home directory and then
download the file). After you've digested it, you can compose your question
to the Oracle. Mail it to the above address, only this time with a subject
line that describes your question. Expect an answer within a couple of days.
And don't be surprised if you also find a question in your mailbox - the
Oracle extracts payment by making seekers of knowledge answer questions as
well!
*"If just one piece of mail gets lost, well, they'll just think they
forgot to send it. But if *two* pieces of mail get lost, hell, they'll
just think the other guy hasn't gotten around to answering his mail.
And if *fifty* pieces of mail get lost, can you imagine it, if *fifty*
pieces of mail get lost, why they'll think someone *else* is broken!
And if 1GB of mail gets lost, they'll just *know* that Arpa is down and
think it's a conspiracy to keep them from their God given right to receive Net Mail ..."*
-- Leith `Casey' Leedom
*****************
Usenet "newsgroups" can be something of a misnomer. They may be
interesting, informative and educational, but they are often not news, at
least, not what you'd think of as news. But there are several sources of
news, sports and weather on the Net.
One of the largest is *Clarinet*, a company in Cupertino, Calf., that
distributes wire-service news and columns, along with a news service devoted
to computers, in Usenet form.
*USA Today* also has a presence on the Net, through the Cleveland Free-Net
system, and we'll show you how to get news of eastern Europe and Brazil as
well.
Distributed in Usenet form, Clarinet stories and columns are organized
into more than 100 newsgroups (in this case, a truly appropriate name), some
of them with an extremely narrow focus, for example, `clari.news.gov.taxes'.
The general news and sports come from United Press International; the
computer news from the NewsBytes service; the features from several
syndicates.
Because Clarinet charges for its service, not all host systems carry its
dispatches. Those that do carry them as Usenet groups starting with
`clari.*' As with other Usenet hierarchies, these are named starting with
broad area and ending with more specific categories. Some of these include
business news (`clari.biz'); general national and foreign news, politics and
the like (`clari.news'), sports (`clari.sports'); columns by MIKE ROYKO, MISS
MANNERS, DAVE BARRY and others (`clari.feature'); and NewsBytes computer and
telecommunications reports (`clari.nb'). Because Clarinet started in Canada,
there is a separate set of `clari.canada' newsgroups.
The clari.nb newsgroups are divided into specific computer types
(`clari.nb.apple', for example).
Clari news groups feature stories updated around the clock. There are
even a couple of "bulletin" newsgroups for breaking stories:
`clari.news.bulletin' and `clari.news.urgent'. Clarinet also sets up new
newsgroups for breaking stories that become ongoing ones (such as major
natural disasters, coups in large countries and the like).
Occasionally, you will see stories in clari newsgroups that just don't
seem to belong there. Stories about former Washington, D.C. mayor MARION
BARRY, for example, often wind interspersed among columns by DAVE BARRY.
This happens because of the way wire services work. UPI uses three-letter
codes to route its stories to the newspapers and radio stations that make up
most of its clientele, and harried editors on deadline sometimes punch in the
wrong code.
=========
If your host system doesn't carry the clari newsgroups, you might be able
to keep up with the news a different way over the Net. USA Today has been
something of an online newspaper pioneer, selling its stories to
bulletin-board and online systems across the country. Cleveland Free-Net
provides the online version of USA Today (along with all its other services)
for free. Currently, the paper only publishes five days a week, so you'll
have to get your weekend news fix elsewhere.
Telnet: `freenet-in-a.cwru.edu' or `freenet-in-b.cwru.edu'
After you connect and log in, look for this menu entry: NPTN/USA TODAY
HEADLINE NEWS. Type the number next to it and hit enter. You'll then get a
menu listing a series of broad categories, such as sports and
telecommunications. Choose one, and you'll get a yet another menu, listing
the ten most recent dates of publication. Each of these contains
one-paragraph summaries of the day's news in that particular subject.
===============
Radio Free Europe and Radio Liberty are American radio stations that
broadcast to the former Communist countries of eastern Europe. Every day,
their news departments prepare a summary of news in those countries, which is
then disseminated via the Net. To subscribe, send an e-mail message to
listserv@ubvm.cc.buffalo.edu
Leave the subject line blank, and as a message, write:
subscribe rferl-l Your Name
Daily Brazilian news updates are available (in Portuguese) from the
University of Sao Paulo. Use anonymous ftp to connect to
uspif.if.usp.br
Use cd to switch to the whois directory. The news summaries are stored in
files with this form: `NEWS.23OCT92;1'. But to get them, leave off the
semicolon and the one, and don't capitalize anything, for example:
get news.23oct92
FYI:
====
The `clari.net.newusers' newsgroup on Usenet provides a number of articles
about Clarinet and ways of finding news stories of interest to you.
*"Be it true or false, so it be news."*
-- Ben Johnson, "News from the World"
*"In a medium in which a News piece takes a minute
and an `In-Depth' piece takes two minutes,
the Simple will drive out the Complex."*
-- Frank Mankiewicz
************************************************************
Many Net systems provide access to a series of interactive services that
let you hold live "chats" or play online games with people around the world.
To find out if your host system offers these, you can ask your system
administrator or just try them - if nothing happens, then your system does
Lost in Space, Battlestar Galactica, the Twilight Zone, the Prisoner and
Doctor Who. There are also lists of various things related to science
fiction and an online science-fiction fanzine. 6 p.m. - 6 a.m.
Shakespeare
-----------
`atari.archive.umich.edu' The shakespeare directory contains most of the
Bard's works. A number of other sites have his works as well, but generally
as one huge mega-file. This site breaks them down into various categories
(comedies, poetry, histories, etc.) so that you can download individual plays
or sonnets.
Space
-----
`ames.arc.nasa.gov' Stores text files about space and the history of the
NASA space program in the `pub/SPACE' subdirectory. In the `pub/GIF' and
`pub/SPACE/GIF' directories, you'll find astronomy- and NASA-related GIF
files, including pictures of planets, satellites and other celestial objects.
9 p.m. - 9 a.m.
Spain
-----
`goya.dit.upm.es' This Spanish site carries an updated list of
bulletin-board systems in Spain, as well as information about European
computer networks, in the `info/doc/net' subdirectory, mostly in Spanish.
The BBS list is `bbs.Z', which means you will have to uncompress it to read
it. Available 24 hours.
TeX
---
`ftp.tex.ac.uk' in `pub/archive', `ftp.uni-stuttgart.de' in `tex-archive',
and `ftp.shsu.edu' in `soft/tex' form the CTAN (comprehensive TeX archive
network), that always has the latest TeX version (and everything that comes
with it) available. They are continuously updated, i.e. they are "mirrors"
of the primary TeX archive at Stanford University.
TV
--
`coe.montana.edu' The `pub/TV/Guides' directory has histories and other
information about dozens of TV shows. Only two anonymous-ftp log-ins are
allowed at a time, so you might have to try more than once to get in. 8 p.m.
- 8 a.m.
`ftp.cs.widener.edu' The `pub/simpsons' directory has more files than
anybody could possibly need about Bart and family. The `pub/strek' directory
has files about the original and Next Generation shows as well as the movies.
See also under Science Fiction *Note Science Fiction::.
Travel
------
`nic.stolaf.edu' Before you take that next overseas trip, you might want
to see whether the State Department has issued any kind of advisory for the
countries on your itinerary. The advisories, which cover everything from
hurricane damage to civil war, are in the `pub/travel-advisories/ advisories'
directory, arranged by country. 7 p.m. - 7 a.m.
Usenet
------
`ftp.uu.net' In the usenet directory, you'll find "frequently asked
questions" files, copied from `pit-manager.mit.edu'. The communications
directory holds programs that let MS-DOS users connect directly with UUCP
sites. In the info directory, you'll find information about ftp and ftp
sites. The inet directory contains information about Internet. Available 24
hours.
`pit-manager.mit.edu' This site contains all available FAQs "frequently
asked questions" files for Usenet newsgroups in the `pub/usenet' directory.
For easy access, get the `index' file. See under Books *Note Books:: for a
caveat in using this ftp site. 6 p.m. - 6 a.m.
Viruses
-------
`ftp.unt.edu' The antivirus directory has anti-virus programs for MS-DOS
and Macintosh computers. 7 p.m. - 7 a.m.
Weather
-------
`vmd.cso.uiuc.edu' No password needed. The wx directory contains GIF
weather images of North America. Files are updated hourly and take this
general form: `CV100222'. The first two letters tell the type of file: CV
means it is a visible-light photo taken by a weather satellite. CI images
are similar, but use infrared light. Both these are in black and white.
Files that begin with SA are color radar maps of the U.S. that show severe
weather patterns but also fronts and temperatures in major cities. The
numbers indicate the date and time (in GMT - five hours ahead of EST) of the
image: the first two numbers represent the month, the next two the date, the
last two the hour. The file `WXKEY.GIF' explains the various symbols in SA
files.
X Windows
---------
`ftp.x.org' The default server for MIT's X Window System. See under `/pub'.
`ftp.germany.eu.net' Germany's backbone site located at the University of
Dortmund, in the European part of the Internet; the so-called EUnet. It's
also Germany's default server for X window system releases, and also
"mirrors" several important sites; e.g. in `pub/packages/gnu' the GNU
project's default server. Furthermore you'll find "mirrors" of `386BSD',
`NetBSD', and `Linux'. Available 24 hours.
=====================
* You get a "host unavailable" message. The ftp site is down for some
reason. Try again later.
* You get a "host unknown" message. Check your spelling of the site name.
* You misspell "anonymous" when logging in and get a message telling you a
password is required for whatever you typed in. Type something in, hit
enter, type bye, hit enter, and try again.
FYI:
====
Liberal use of archie will help you find specific files or documents. For
information on new or interesting ftp sites, try the `comp.archives'
newsgroup on Usenet. You can also look in the `comp.misc',
`comp.sources.wanted' or `news.answers' newsgroups on Usenet for lists of ftp
sites posted every month by TOM CZARNIK and JON GRANROSE.
The `comp.archives' newsgroup carries news of new ftp sites and
interesting new files on existing sites.
In the `comp.virus' newsgroup on Usenet, look for postings that list ftp
sites carrying anti-viral software for Amiga, MS-DOS, Macintosh, Atari and
other computers.
The `comp.sys.ibm.pc.digest' and `comp.sys.mac.digest' newsgroups provide
information about new MS-DOS and Macintosh programs as well as answers to
questions from users of those computers.
*"Welch ein Ort zum Pluendern!"
(What a place to plunder!)*
-- General Gebhard Leberecht von Bluecher
*************************************
Even with tools like Hytelnet and archie, telnet and ftp can still be
frustrating. There are all those telnet and ftp addresses to remember.
Telnet services often have their own unique commands. And, oh, those weird
directory and file names!
But now that the Net has become a rich repository of information, people
are looking at ways to make it far easier to find all that data. Gophers and
Wide-Area Information Servers (WAISs) are two programs that could ultimately
make the Internet as easy to navigate as commercial networks like CompuServe
or Prodigy.
Both programs essentially take a request for information and then scan the
Net for it, so you don't have to. Both also work through menus - instead of
typing in some long sequence of characters, you just move a cursor to your
choice and hit enter. Newer gophers even let you select files and programs
from ftp sites this way.
Let's look at gophers first.
Many public-access sites now have gophers online. To use one, type
gopher
at the command line and hit enter. If you know your site does not have a
gopher, or if nothing happens when you type that, telnet to
consultant.micro.umn.edu
At the log-in prompt, type
gopher
and hit enter. You'll be asked what type of terminal emulation you're
using, after which you'll see something like this:
Internet Gopher Information Client v1.03
Root gopher server: gopher.micro.umn.edu
--> 1. Information About Gopher/
2. Computer Information/
3. Discussion Groups/
4. Fun & Games/
5. Internet file server (ftp) sites/
6. Libraries/
7. News/
8. Other Gopher and Information Servers/
9. Phone Books/
10. Search lots of places at the U of M <?>
11. University of Minnesota Campus Information/
Press ? for Help, q to Quit, u to go up a menu Page: 1/1
Gophers are great for exploring. Just keep making choices to see what
pops up. Play with it; see where it takes you. Some choices will be
documents. When you read one of these and either come to the end or hit a
lower-case `q' to quit reading it, you'll be given the choice of saving a
copy to your home directory or e-mailing it to yourself. Other choices are
simple databases that let you enter a word to look for in a particular
database.
Notice that one of your choices is "Internet file server (ftp) sites."
Choose this, and you'll be connected to a modified archie program - an archie
with a difference. When you search for a file through a gopher archie,
you'll get a menu of sites that have the file you're looking for, just as
with the old archie. Only now, instead of having to write down or remember
an ftp address and directory, all you have to do is position the cursor next
to one of the numbers in the menu and hit enter. You'll be connected to the
ftp site, from which you can then choose the file you want, again just by
making a choice in a menu.
You'll be asked for a name in your home directory to use for the file,
after which the file will be copied to your home system. Unfortunately, this
file-transfer process does not yet work with all public-access sites for
computer programs and compressed files. If it doesn't work with yours, you'll
have to get the file the old-fashioned way, via ftp.
The letter u is an important one to remember while navigating a gopher -
it moves you back up a gopher directory tree, much like cd .. on an ftp site.
In addition to ftp sites, there are now scores of databases and libraries
around the world accessible through gophers. There is not yet a common
gopher interface for library catalogs, so be prepared to follow the online
directions more closely when you use gopher to connect to one.
Some gopher menu choices will end with a <?>. This means that if you
select it, you'll be starting up a simple database that can search through the
given service by keyword.
So many services are now available through gophers, that finding what you
want has become difficult. Fortunately, you can use veronica, a laboriously
constructed acronym that does for "gopherspace" what archie (there is no
betty, yet) did for files. You'll usually find veronicas (there are now
several) under "Other gopher and information services." When you call up a
veronica, tell her (it?) the keyword or words you're interested in, and she/it
will search all available databases for it. For example, say you want to
impress company tonight and make cherries flambe. If you were to type in
"flambe" after calling up veronica, you would soon get a menu listing several
flambe recipes, including one called "dessert flambe." Put your cursor on
that line of the menu and hit enter, and you'll find it's a menu for cherries
flambe. Then hit your q key to quit, and gopher will ask you if you want to
save the file in your home directory on your public-access site or whether you
want to e-mail it somewhere.
========
Veronica is a "Very Easy Rodent-Oriented Net-wide Index to Computerized
Archives." Veronica's name is a play on the concepts of both gopher and
archie. (Remember the comic book couple Archie and Veronica? Veronica does
for gopher what archie does for anonymous FTP.) Veronica searches through
hundreds of gopher holes looking for anything that matches a keyword supplied
by the user, and assembles a list of gopher servers that contain items of
interest. Note: veronica checks *titles* of gopher items only, not their
contents.
At present, there are no veronica clients; veronica is a gopher tool. An
informal veronica FAQ is posted regularly in `comp.infosystems.gopher' and
archived on `veronica.scs.unr.edu' as `veronica/veronica-faq'.
=============================
Now you know there are hundreds of databases and library catalogs you can
search through. But as you look, you begin to realize that each seems to
have its own unique method for searching. If you connect to several, this
can become a pain. Gophers reduce this problem somewhat.
Wide-area information servers promise another way to zero in on
information hidden on the Net. In a WAIS, the user sees only one interface -
the program worries about how to access information on dozens, even hundreds,
of different databases. You tell give a WAIS a word and it scours the net
looking for places where it's mentioned. You get a menu of documents, each
ranked according to how relevant to your search the WAIS thinks it is.
Like gophers, WAIS "client" programs can already be found on many
public-access Internet sites. If it does, type
swais
at the command line and hit enter (the "s" stands for "simple"). If it
doesn't, telnet to bbs.oit.unc.edu, which is run by the University of North
Carolina At the "login:" prompt, type
bbs
and hit enter. You'll be asked to register and will then get a list of
"bulletins," which are various files explaining how the system works. When
done with those, hit your Q key and you'll get another menu. Hit 4 for the
"simple WAIS client," and you'll see something like this:
SWAIS Source Selection Sources: 23#
Server Source Cost
001: [ archie.au] aarnet-resource-guide Free
002: [ archive.orst.edu] aeronautics Free
003: [nostromo.oes.orst.ed] agricultural-market-news Free
004: [sun-wais.oit.unc.edu] alt-sys-sun Free
005: [ archive.orst.edu] alt.drugs Free
006: [ wais.oit.unc.edu] alt.gopher Free
007: [sun-wais.oit.unc.edu] alt.sys.sun Free
008: [ wais.oit.unc.edu] alt.wais Free
009: [ archive.orst.edu] archie-orst.edu Free
010: [ archie.au] archie.au-amiga-readmes Free
011: [ archie.au] archie.au-ls-lRt Free
012: [ archie.au] archie.au-mac-readmes Free
013: [ archie.au] archie.au-pc-readmes Free
014: [ pc2.pc.maricopa.edu] ascd-education Free
015: [ archie.au] au-directory-of-servers Free
016: [ cirm2.univ-mrs.fr] bib-cirm Free
017: [ cmns-sun.think.com] bible Free
018: [ zenon.inria.fr] bibs-zenon-inria-fr Free
Keywords:
Each line represents a different database (the .au at the end of some of
them means they are in Australia; the .fr on the last line represents a
database in France). And this is just the first page! If you type a capital
K, you'll go to the next page (there are several pages). Hitting a capital J
will move you back a page.
The first thing you want to do is tell the WAIS program which databases
you want searched. To select a database, move the cursor bar over the line
you want (using your down and up arrow keys) and hit your space bar. An
asterisk will appear next to the line number. Repeat this until you've
selected all of the databases you want searched. Then hit your W key, after
which you'll be prompted for the key words you're looking for. You can type
in an entire line of these words - separate each with a space, not a comma.
Hit return, and the search begins.
Let's say you're utterly fascinated with wheat. So you might select
agricultural-market-news to find its current world price. But you also want
to see if it has any religious implications, so you choose the Bible and the
Book of Mormon. What do you do with the stuff? Select recipes and
usenet-cookbook. Are there any recent Supreme Court decisions involving the
plant? Chose supreme-court. How about synonyms? Try roget-thesaurus and
just plain thesaurus.
Now hit w and type in wheat. Hit enter, and the WAIS program begins its
search. As it looks, it tells you whether any of the databases are offline,
and if so, when they might be ready for a search. In about a minute, the
program tells you how many hits it's found. Then you get a new menu, that
looks something like this:
Keywords:
# Score SourceTitleLines
001: [1000] (roget-thesaurus) #465. [results of comparison. 1] Di 19
002: [1000] (roget-thesaurus) #609. Choice. -- N. choice, option; 36
003: [1000] (roget-thesaurus) #465. [results of comparison. 1] Di 19
004: [1000] (roget-thesaurus) #609. Choice. -- N. choice, option; 36
005: [1000] (recipes) aem@mthvax Re: MONTHLY: Rec.Food.Recipes 425
006: [1000] ( Book_of_Mormon) Mosiah 9:96
007: [1000] ( Book_of_Mormon) 3 Nephi 18:185
008: [1000] (agricultural-ma) Re: JO GR115, WEEKLY GRAIN82
009: [ 822] (agricultural-ma) Re: WA CB351 PROSPECTIVE PLANTINGS 552
010: [ 800] ( recipes) kms@apss.a Re: REQUEST: Wheat-free, Suga 35
011: [ 750] (agricultural-ma) Re: WA CB101 CROP PRODUCTION258
012: [ 643] (agricultural-ma) Re: SJ GR850 DAILY NAT GRN SUM72
013: [ 400] ( recipes) pat@jaamer Re: VEGAN: Honey Granola63
014: [ 400] ( recipes) jrtrint@pa Re: OVO-LACTO: Sourdough/Trit 142
Each of these represents an article or citing that contains the word wheat,
or some related word. Move the cursor bar (with the down and up arrow keys)
to the one you want to see, hit enter, and it will begin to appear on your
screen. The "score" is a WAIS attempt to gauge how closely the citing
matches your request. Doesn't look like the Supreme Court has had anything
to say about the plant of late!
Now think of how much time you would have spent logging onto various
databases just to find these relatively trivial examples. But as more
databases are added to WAIS programs, a problem arises that is similar to the
one WAISs were supposed to solve: how do you find the specific databases you
want? Scrolling through page after page of database listings becomes rather
tedious rather quickly and you could wind up missing the one database you
really need. That's the next step in WAIS research.
==============
Developed by researchers at CERN, the European Particle Physics Laboratory
in Geneva, the World-Wide Web project is somewhat similar to a WAIS. But
it's designed on a system known as hypertext. Words in one document are
"linked" to other documents. It's sort of like sitting with an encyclopedia
- you're reading one article, see a reference that intrigues you and so you
flip the pages to look up that reference.
To take a walk on The Web try the WWW's (or W3 in the project's own
jargon), default telnet site:
info.cern.ch
No log in is needed. When you connect, you'll see:
Overview of the Web
GENERAL OVERVIEW
There is no "top" to the World-Wide Web. You can look at it from many points
of view. If you have no other bias, here are some places to start:
by Subject[1] A classification by subject of interest. Incomplete
but easiest to use.
by Type[2] Looking by type of service (access protocol, etc) may
allow to find things if you know what you are looking
for.
About WWW[3] About the World-Wide Web global information sharing
project
Starting somewhere else
To use a different default page, perhaps one representing your field of
interest, see "customizing your home page"[4].
What happened to CERN?
1-6, Up,
Ok. Now type `3', and get the following screen:
The World Wide Web project
WORLD WIDE WEB
The WorldWideWeb (W3) is a wide-area hypermedia[1] information retrieval
initiative aiming to give universal access to a large universe of documents.
Everything there is online about W3 is linked directly or indirectly to this
document, including an executive summary[2] of the project, an illustrated
talk[3] , Mailing lists[4] , Policy[5] and Conditions[6] , May's W3 news[7]
, Frequently Asked Questions[8] .
What's out there?[9] Pointers to the world's online information,
subjects[10] , W3 servers[11] , etc.
WWW Software Products[12]
What there is and how to get it: clients, servers and
tools.
Technical[13] Details of protocols, formats, program internals etc
Bibliography[14] Paper documentation on W3 and references. Also:
manuals[15] .
1-20, Back, Up,
You navigate the web by typing the number next to a given reference. So
if you want to know more about the web, hit 2. This is another system that
bears playing with.
=======
If you are used to plain-vanilla Unix or MS-DOS, then the way these
gophers and WAISs work seems quite straightforward. But if you're used to a
computer with a graphical interface, such as a Macintosh, an IBM compatible
with Windows or a Next, you'll probably regard their interfaces as somewhat
primitive.
There are, however, ways to integrate these services into your graphical
user interface. In fact, there are now ways to tie into the Internet
directly, rather than relying on whatever interface your public-access system
uses.
There is now a growing number of these "client" programs for everything
from ftp to gopher. PSI of Reston, Va., which offers nationwide Internet
access, in fact, requires its customers to use these programs.
Using protocols known as SLIP and PPP, these programs communicate with the
Net using the same basic data packets as much larger computers online.
Beyond integration with your own computer's "desktop," client programs let
you do more than one thing at once on the net - while your downloading a
large file in one window, you can be chatting with a friend through an
Internet chat program in another.
These client programs have a couple of disadvantages. One is that you'll
need a 9600-baud modem - while it is possible to connect to the Net with them
at lower speeds, you will likely find them painfully slow. Not all
public-access sites are set up to allow such connections. And those that are
usually charge far more for them.
Your system administrator can give you more information on setting up one
of these connections.
FYI:
====
See the Usenet newsgroups `comp.infosystems.*':
`comp.infosystems.gopher', `comp.infosystems.wais', and
`comp.infosystems.www' are places to go for technical discussions about
Gopher, WAISs, and the World-Wide Web project respectively. Moreover there
are `comp.infosystems' for more general discussion of related issues. The
group `comp.infosystems.gis' relates to Geographic Information Systems, and
thus is more specialized on this subject.
There even exists a Gopher service to read Usenet news: `gopher
gopher.msu.edu 4320'. But, the lines behind this service are few, and thus
it's likely that you get the following message, when trying to enter:
We are sorry, but our Usenet News gateway limits the number of
simultaneous connections. If you were attempting to read news and were
instead directed to this file, all of those connections are in use. We
offer this gateway as a "last resort" for people who have no other
access for reading Usenet. We do not have the capacity to serve as the
Usenet gateway for large numbers of users around the Internet.
Individuals who like this style of access should ask their Internet
service providers to offer the same sort of gateway on their local
Gopher server. Individuals and campuses should consider installing
local news feeds and local news readers (such as RN, NN, TIN, or
Trumpet) so that users can read and post to Usenet newsgroups
conveniently.
For system administrators: the software we use to implement this gateway
is the go4gw Gopher gateway software from Roland Schemers of Stanford
University. This software should be available by anonymous ftp from
boombox.micro.umn.edu, somewhere under /pub/gopher.
-- The Michigan State University Gopher Team
*"Reliable information is the basis of successful planning."*
-- Christoph Columbus
***************
E-mail by itself is a powerful tool, and by now you may be sending e-mail
messages all over the place. You might even be on a mailing list or two. But
there is a lot more to e-mail than just sending messages. If your host
system does not have access to ftp, or it doesn't have access to every ftp
site on the Net, you can have programs and files sent right to your mailbox.
And using some simple techniques, you can use e-mail to send data files such
as spreadsheets, or even whole programs, to friends and colleagues around the
world.
A key to both is a set of programs known as encoders and decoders. For
all its basic power, Net e-mail has a big problem: it can't handle graphics
characters or the control codes found in even the simplest of computer
programs. Encoders however, can translate these into forms usable in e-mail,
while decoders turn them back into a form that you can actually use. If you
are using a Unix-based host system, chances are it already has an encoder and
decoder online that you can use. These programs will also let you use
programs posted in several Usenet newsgroups, such as `comp.binaries.ibm.pc'.
To help people without ftp access, a number of ftp sites have set up mail
servers (also known as archive servers) that allow you to get files via
e-mail. You send a request to one of these machines and they send back the
file you want. As with ftp, you'll be able to find everything from
historical documents to software (but please note that if you do have access
to ftp, that method is always quicker and ties up fewer resources than using
e-mail).
Some interesting or useful mail servers include:
Files of "frequently asked questions" related to Usenet; state-by-state
lists of U.S. representatives and Senators and their addresses and
office phone numbers.
Information about the Electronic Frontier Foundation; documents about
legal issues on the Net.
Back copies of the Computer Underground Digest and every possible fact
you could want to know about "The Simpsons."
Programs for many types of personal computers; archives of past postings
from many Usenet newsgroups.
Space-related text and graphics (GIF-format) files.
Detailed information about Internet.
Most mail servers work pretty much the same - you send an e-mail message
that tells them what file you want and how you want it sent to you. The most
important command is "send," which tells the computer you want it to send you
a particular file.
First, though, you'll need to know where the mail server stores that file,
because you have to tell it which directory or sub-directory it's in. There
are a couple of ways to do this. You can send an e-mail message to the
archive-server that consists of one line:
index
The server will then send you a directory listing of its main, or root
directory. You'll then have to send a second message to the archive server
with one line:
index directory/subdirectory
where that is the directory or directory path for which you want a
listing. An alternative is to send an e-mail message to our old friend
archie, which should send you back the file's exact location on the
archive-server (along with similar listings for all the other sites that may
have the file, however)
Once you have the file name and its directory path, compose a message to
the archive server like this:
send directory/subdirectory/file
Send off the message and, anywhere from a few minutes to a couple of days
later, you'll find a new message in your mailbox: a copy of the file you
requested. The exact time it will take a file to get to you depends on a
variety of factors, including how many requests are in line before yours
(mail servers can only process so many requests at a time) and the state of
the connections between the server and you.
Seems simple enough. It gets a little more complicated when you request a
program rather than a document. Programs or other files that contain unusual
characters or lines longer than 130 characters (graphics files, for example)
require special processing by both the mail server to ensure they are
transmitted via e-mail. Then you'll have to run them through at least one
converter program to put them in a form you can actually use. To ensure that
a program or other "non-mailable" file actually gets to you, include another
line in your e-mail message to the server:
encoder
This converts the file into an encoded form. To decode it, you'll first
have to transfer the file message into a file in your home directory. If you
are using the simple mail program, go into mail and type
w # file.name
where `#' is the number of the message you want to transfer and file.name
is what you want to call the resulting file. In pine, call up the message
and hit your `O' key and then `E'. You'll then be asked for a file name. In
elm, call up the message and hit your `S' key. You'll get something that
looks like this:
=file.request
Type a new file name and hit enter (if you hit enter without typing a file
name, the message will be saved to another mail folder, not your home
directory). Exit mail to return to your host system's command line. Because
the file has been encoded for mail delivery, you now have to run a decoder.
At the command line, type
uudecode file.name
where `file.name' is the file you created while in mail. Uudecode will
create a new, uncompressed file. In some cases, you may have to run it
through some other programs (for example, if it is in "tar" form), but
generally it should now be ready for you to download to your own computer.
One further complication comes when you request a particularly long file.
Many Net sites can only handle so much mail at a time. To make sure you get
the entire file, tell the mail server to break it up into smaller pieces,
with another line in your e-mail request like this:
size 100000
This gives the mail server the maximum size, in bytes, of each file
segment. This particular size is good for UUCP sites. Internet and Bitnet
sites can generally go up to 300000. When you get all of these files in
mail, transfer them to your home directory. Exit mail and call up each file
in your host system's text processor and delete each one's entire header and
footer (or "signature" at the end). When done with this, at your host
system's command line, type
cat file1 file2 >bigfile
where file1 is the first file, file2 the second file, and so on. The >
tells your host system to combine them into a new megafile called bigfile (or
whatever you want to call it). You can then run uudecode, tar, etc. One word
of caution, though: if the file you want is long enough that it has to be
broken into pieces, think of how much time it's going to take you to download
the whole thing - especially if you're using a 2400-baud modem!
There are a number of other mail servers. To get a list, send an e-mail
message to:
send usenet/comp.sources.wanted/How_to_find_sources_(READ_THIS_BEFORE_POSTING)
You'll have to spell it exactly as listed above. Some mail servers use
different software, which will require slightly different commands than the
ones listed here. In general, if you send a message to a mail server that
says only
help
you should get back a file detailing all of its commands.
But what if the file you want is not on one of these mail servers? That's
where ftpmail comes in. Run by Digital Equipment Corp. in California, this
service can connect to almost any ftp site in the world, get the file you
want and then mail it to you. Using it is fairly simple - you send an e-mail
message to ftpmail that includes a series of commands telling the system
where to find the file you want and how to format it to mail to you.
Compose an e-mail message to
ftpmail@decwrl.dec.com
Leave the "subject:" line blank. Inside the message, there are several
commands you can give. The first line should be
reply address
where "address" is your e-mail address. The next line should be
connect host
where "host" is the system that has the file you want (for example:
wuarchive.wustl.edu). Other commands you should consider using are "binary"
(required for program files); "compress" (reduces the file size for quicker
transmission) and "uuencode" (which encodes the file so you can do something
with it when it arrives). The last line of your message should be the word
"quit".
Let's say you want a copy of the U.S. constitution. Using archie, you've
found a file called, surprise!, `constitution', at the ftp site
`archive.cis.ohio-state.edu', in the `/pub/firearms/politics/ rkba'
directory. You'd send a message to
this:
reply adamg@world.std.com
connect archive.cis.ohio-state.edu
binary
compress
uuencode
get pub/firearms/politics/rkba/constitution
quit
When you get the file in your mailbox, use the above procedure for copying
it to a file. Run it through uudecode. Then type
uncompress file.name
to make it usable.
Since this was a text file, you could have changed the "binary" to "ascii"
and then eliminated the "uuencode" file. For programs, though, you'll want
to keep these lines.
=======================================
The uuencode and uudecode programs will also come in handy if you ever
want to send your own files to somebody else.
If both you and your intended recipient communicate via Unix-based host
systems, then it's pretty easy, because almost all Unix host systems will
have encoder/decoder programs online.
First, upload the file you want to send to your friend to your host site.
Ask your system administrator how to upload a file to your name or "home"
directory. Then type
uuencode file file >file.uu
and hit enter. "File" is the name of the file you want to prepare for
mailing, and yes, you have to type the name twice! The > is a Unix command
that tells the system to call the "encoded" file "file.uu" (you could
actually call it anything you want).
Now to get it into a mail message. The quick and dirty way is to type
mail friend
where "friend" is your friend's address. At the subject line, tell her
the name of the enclosed file. When you get the blank line, type
~r file.uu
or whatever you called the file, and hit enter. (on some systems, the `~'
may not work; if so, ask your system administrator what to use). This
inserts the file into your mail message. Hit control-D, and your file is on
its way!
On the other end, when your friend goes into his mailbox, she should
transfer it to her home directory. Then your friend should type
uudecode file.name
and hit enter. This creates a new file in her name directory with
whatever name you originally gave it. She can then download it to her own
computer. Before she can actually use it, though, she'll have to open it up
with a text processor and delete the mail header that has been "stamped" on
it. If you use a mailer program that automatically appends a "signature,"
tell her about that so she can delete that as well.
But what if your friend only connects with a non-Unix system, such as
CompuServe or MCIMail? There are programs available for MS-DOS, Apple and
Amiga computers that will encode and decode files. Of course, since you
can't send one of these programs to them via e-mail (how would they un-encode
it?), you'll have to mail or give them a diskette with the program on it
first. Then, they can get their message, run it through a text editor to
delete the header, and finally decode the file. If they want to send you
files in return, they'll also want an encoder
For MS-DOS machines, you'll want to get `uunecode.com' and `uudecode.com'.
Both can be found through anonymous ftp at `wuarchive.wustl.edu' in the
`/mirrors/msdos/starter' directory. The MS-DOS version is as easy to use as
the Unix one: Just type
uudecode filename.ext
and hit enter.
Mac users should get a program called uutool, which can be found in the
`info-mac/util' directory on `sumex-aim.stanford.edu'.
Once again, be careful with large files. Although large sites connected
directly to the Internet can probably handle mega-files, many smaller systems
cannot. Some commercial systems, such as CompuServe and MCIMail limit the
size of mail messages their users can receive. Fidonet doesn't even allow
encoded messages. In general, a file size of 30,000 or so bytes is a safe
upper limit for non-Internet systems.
One other thing you can do through e-mail is consult with the Usenet
Oracle. You can ask the Oracle anything at all and get back an answer
(whether you like the answer is another question).
First, you'll want to get instructions on how to address the Oracle (he,
or she, or it, is very particular about such things and likes being addressed
in august, solemn and particularly sycophantic tones). Start an e-mail
message to
oracle@iuvax.cs.indiana.edu
In the "subject:" line, type
help
and hit enter. You don't actually have to say anything in the message
itself - at least not yet. Hit control-D to send off your request for help.
Within a few hours, the Oracle will mail you back detailed instructions.
It's a fairly long file, so before you start reading it, turn on your
communications software's logging function, to save it to your computer (or
save the message to a file on your host system's home directory and then
download the file). After you've digested it, you can compose your question
to the Oracle. Mail it to the above address, only this time with a subject
line that describes your question. Expect an answer within a couple of days.
And don't be surprised if you also find a question in your mailbox - the
Oracle extracts payment by making seekers of knowledge answer questions as
well!
*"If just one piece of mail gets lost, well, they'll just think they
forgot to send it. But if *two* pieces of mail get lost, hell, they'll
just think the other guy hasn't gotten around to answering his mail.
And if *fifty* pieces of mail get lost, can you imagine it, if *fifty*
pieces of mail get lost, why they'll think someone *else* is broken!
And if 1GB of mail gets lost, they'll just *know* that Arpa is down and
think it's a conspiracy to keep them from their God given right to receive Net Mail ..."*
-- Leith `Casey' Leedom
*****************
Usenet "newsgroups" can be something of a misnomer. They may be
interesting, informative and educational, but they are often not news, at
least, not what you'd think of as news. But there are several sources of
news, sports and weather on the Net.
One of the largest is *Clarinet*, a company in Cupertino, Calf., that
distributes wire-service news and columns, along with a news service devoted
to computers, in Usenet form.
*USA Today* also has a presence on the Net, through the Cleveland Free-Net
system, and we'll show you how to get news of eastern Europe and Brazil as
well.
Distributed in Usenet form, Clarinet stories and columns are organized
into more than 100 newsgroups (in this case, a truly appropriate name), some
of them with an extremely narrow focus, for example, `clari.news.gov.taxes'.
The general news and sports come from United Press International; the
computer news from the NewsBytes service; the features from several
syndicates.
Because Clarinet charges for its service, not all host systems carry its
dispatches. Those that do carry them as Usenet groups starting with
`clari.*' As with other Usenet hierarchies, these are named starting with
broad area and ending with more specific categories. Some of these include
business news (`clari.biz'); general national and foreign news, politics and
the like (`clari.news'), sports (`clari.sports'); columns by MIKE ROYKO, MISS
MANNERS, DAVE BARRY and others (`clari.feature'); and NewsBytes computer and
telecommunications reports (`clari.nb'). Because Clarinet started in Canada,
there is a separate set of `clari.canada' newsgroups.
The clari.nb newsgroups are divided into specific computer types
(`clari.nb.apple', for example).
Clari news groups feature stories updated around the clock. There are
even a couple of "bulletin" newsgroups for breaking stories:
`clari.news.bulletin' and `clari.news.urgent'. Clarinet also sets up new
newsgroups for breaking stories that become ongoing ones (such as major
natural disasters, coups in large countries and the like).
Occasionally, you will see stories in clari newsgroups that just don't
seem to belong there. Stories about former Washington, D.C. mayor MARION
BARRY, for example, often wind interspersed among columns by DAVE BARRY.
This happens because of the way wire services work. UPI uses three-letter
codes to route its stories to the newspapers and radio stations that make up
most of its clientele, and harried editors on deadline sometimes punch in the
wrong code.
=========
If your host system doesn't carry the clari newsgroups, you might be able
to keep up with the news a different way over the Net. USA Today has been
something of an online newspaper pioneer, selling its stories to
bulletin-board and online systems across the country. Cleveland Free-Net
provides the online version of USA Today (along with all its other services)
for free. Currently, the paper only publishes five days a week, so you'll
have to get your weekend news fix elsewhere.
Telnet: `freenet-in-a.cwru.edu' or `freenet-in-b.cwru.edu'
After you connect and log in, look for this menu entry: NPTN/USA TODAY
HEADLINE NEWS. Type the number next to it and hit enter. You'll then get a
menu listing a series of broad categories, such as sports and
telecommunications. Choose one, and you'll get a yet another menu, listing
the ten most recent dates of publication. Each of these contains
one-paragraph summaries of the day's news in that particular subject.
===============
Radio Free Europe and Radio Liberty are American radio stations that
broadcast to the former Communist countries of eastern Europe. Every day,
their news departments prepare a summary of news in those countries, which is
then disseminated via the Net. To subscribe, send an e-mail message to
listserv@ubvm.cc.buffalo.edu
Leave the subject line blank, and as a message, write:
subscribe rferl-l Your Name
Daily Brazilian news updates are available (in Portuguese) from the
University of Sao Paulo. Use anonymous ftp to connect to
uspif.if.usp.br
Use cd to switch to the whois directory. The news summaries are stored in
files with this form: `NEWS.23OCT92;1'. But to get them, leave off the
semicolon and the one, and don't capitalize anything, for example:
get news.23oct92
FYI:
====
The `clari.net.newusers' newsgroup on Usenet provides a number of articles
about Clarinet and ways of finding news stories of interest to you.
*"Be it true or false, so it be news."*
-- Ben Johnson, "News from the World"
*"In a medium in which a News piece takes a minute
and an `In-Depth' piece takes two minutes,
the Simple will drive out the Complex."*
-- Frank Mankiewicz
************************************************************
Many Net systems provide access to a series of interactive services that
let you hold live "chats" or play online games with people around the world.
To find out if your host system offers these, you can ask your system
administrator or just try them - if nothing happens, then your system does