This document introduces and describes the various types of Internet services referred to in my faq. I broke it out of the faq file in order to streamline it and divide up the fuctions. The descriptions of services here are my own and are not official.
You subscribe to a mailing list and are sent messages by email. Mailing lists are the best way to automatically get the latest info.
Messages are sent to you either immediately when they arrive at the distribution site (reflected mode), or are sent as one big message per day (called a digest) containing all the day's messages. With a digest you don't get the messages in real time, but you also don't get a lot of mailing list messages mixed in with your other mail. Not all lists offer digests.
To subscribe, send mail to the subscription address and fill out the message appropriately. You can often send a help message to the same address. Be aware that some lists are run manually by (overworked) humans, so be polite when you write.
For lists run using listproc or listserv you usually give your real name after the list name when subscribing. For lists run using majordomo, you typically don't need to give your name or email address when subscribing, just the name of the list. For majordomo, you only need to give an email address when your preferred subscription address is different from your return address. With majordomo you can usually get information about a list by sending a message with "info list-name" in the body to the site's majordomo address.
You send email to these sites and they send prepared files back as email. Similar effect as ftp, but not as powerful a medium. Send request messages to the server listed.
Also includes email-to-fax gateways and any other services offered by email (except mailing lists which are elsewhere).
Services like ftp, gopher, and WWW (described below) are typically accessed interactively, that is, you connect to a particular service and get the info you're looking for by communicating with it directly and in real time.
However, if you have only email access to the Internet you can still get files at ftp, WWW and similar sites. See Accessing The Internet By E-Mail . (I also have a local copy of this document, but it's not necessarily current.) To get the faq by email send a message:To: mail-server@rtfm.mit.edu-nospam Enter only this line in the BODY of the note: send usenet/news.answers/internet-services/access-via-email
Of course, if you have WWW or ftp or telnet access, use it.
ftp is the Internet file transfer protocol/program. It stands for both the standard and the program used to transfer files. Use ftp to get files to your Internet-connected machine from other Internet machines. You can also use ftp to put (send, upload) files at some sites. ftp sites listed here support anonymous logins, meaning anyone can use them by giving "anonymous" for Name and your email address (not password) as Password.
Your site must be wired into the Internet to use ftp directly. If you can't ftp directly, you can ftp by email via servers, but it's a bit slow (hours to days). Still, it's better to be able to get files than not get them. Companies wired into the Internet often block ftp, telnet, and gopher access for security reasons. Ask your company Internet administrator if there is a site from which you can ftp/telnet/gopher out.
I've put some instructions for using using ftp to access my directory in file "ftp://rkba.org/how or http://rkba.org/how". Using the other ftp sites is similar. Note that ftp sites are listed in URL style, meaning:
ftp://rkba.org/research/malcom
^^site^^ ^^dir^^^ ^file
Using telnet you run a shell (terminal session) at some other remote Internet site. Logging into a remote site via telnet looks much like logging directly there. Telnet sites are reference either by site name or numeric ip address.
Gopher is a text-menu-based, Internet information service. Gopher provides a more automated way to get information from the Internet than ftp or telnet. I think of gohper as a text-based precursor to WWW. As a point of reference the number of gopher sites is growing at something like 100% a year. The number of WWW sites is growing about 9000% in the same period. A practical note: since they are purely plaintext-based (and less popular), gopher sites are sometimes faster than Web sites.
On a UNIX system you can start gopher from the command line by
typing "gopher sitename". WWW browsers also talk "gopher"
automatically. Gopher WWW URLs look like: "gopher://
WAIS stands for Wide Area Information Server. As I understand it, WAIS is a database that functions like an index to resources on the Internet. Like ftp, telnet, gopher and WWW, to use WAIS you need to run a client on your Internet machine, or you need to point your WWW browser at a WAIS server. Info on client programs can be found in the usual Usenet newsgroups such as comp.protocols.tcp-ip.ibmpc for PC versions.
WWW (World Wide Web) is an Internet standard for distributed hypertext. This means that WWW documents can have links to other documents which can be anywhere on the Internet. (Common examples of non-distributed hypertext include Hypercard stacks and Windows Help screens.) For example, if a document mentions "Constitution," clicking on that word might bring up a copy of the Constitution. Clicking on the Second Amendment might bring up a discussion of that topic, which might include links to articles at a completely different site, and so on.
Here's a little bit more background on WWW. If it seems like a lot of info, just try it. WWW is much easier to use than to describe. WWW is a revolutionary technology which I encourage you in the strongest possible terms to get.
Programs for browsing World Wide Web such as CELLO and Mosaic are available for most platforms, including Windows, Mac, Xwindows, and many more. Mosaic and CELLO are very nice GUI (Graphical User Interface) programs which allow you to point and click to navigate most of the common Internet protocols (including those described above). The browser I'm currently using is a follow-on to Mosaic called Netscape, which is available from Netscape Communications at http://home.mcom.com/home/welcome.html or file://ftp.mcom.com/netscape
You generally need an IP (Internet Protocol) network connection to use WWW directly from your machine. However, you can also use a character-based program called "lynx" from your UNIX shell account to view WWW resources. With lynx, no images are displayed, but most of the info gets across. (Images are nice, but can also slow down WWW access a lot). With lynx, the IP connection is your service provider's which is likely much faster connection than your IP connection at home.
If you want the nice graphical WWW interface on your Windows machine, various Windows TCP/IP packages exist, including solid shareware ones like Trumpet, and good supported commercial packages from NetManage, Frontier and others. Most IP packages support serial protocols for connecting directly to the Internet over a modem when you're at home or on the road. For more info see The PC-Internet Connection Update Page at: http://www.zilker.net/users/internaut/update.html or Bernard Aboba's Windows TCP/IP FAQ at: file://ftp.netcom.com/pub/mailcom/IBMTCP/ibmtcp.zip
Note that the NetManage Chameleon Sampler is included in several introductory books, including The Internet Starter Kit, and The Internet Glossary which range from $20 to $30. (The Internet Starter Kit is available in Windows and Mac versions. The Windows version comes with the Chameleon Sampler; the Mac version probably uses a version of Apple's MacTCP.) This is a simple and inexpensive way to get a solid but deliberately limited Windows TCP/IP stack and applications. The Chameleon sampler is also available via ftp at: file://ftp.netmanage.com/pub/demos/sampler/sampler.exe Be sure to see the The PC-Internet Connection Update Page above.
HTML (HyperText Markup Language) is the name of the particular hypertext language specification used by WWW. You can get information on HTML through your WWW browser. HTML itself is plaintext, editable on any text editor. Macros for Word, emacs and other editors exist, as do standalone HTML editors.
Universal Resource Locators (URLs) are pointers to WWW resources. They can be ftp sites, http (WWW) sites, mail destinations, gopher sites, telnet sites, newsgroups, -- in short all the common Internet user interface methods -- available just by pointing & clicking....
Paul Lam (homeboy@hooked.net-nospam) reports on a nice utility for searching WWW at CMU: http://lycos3.cs.cmu.edu/lycos-form.html
Usenet newsgroups are like conferences or forums on a BBS. Generally, anyone can read and post messages to a newsgroup. There are elaborate newsgroup hierarchies, starting with sci, talk, comp, alt, rec, and so on. Newsgroup postings expire after a certain amount of time regardless of whether you read them or not. Once you read a particular message, you usually don't see it again. Newsgroups have very wide readership but some have a lot of discussion.
Availability of newsgroups varies with particular sites. Ask your site administrator for more information about accessing newsgroups. You may be able to request that additional newsgroups be carried at your site. Popular UNIX newsreaders include rn, nn and tin. If your WWW browser points to a news server, you can use your browser to read these newsgroups.
Note that a Usenet Newsgroup is not the same as a BBS.
A computer Bulletin Board System/Service (BBS) is a computer with modem(s) that you dial into using your modem to get and send files and messages. Files and messages are usually grouped into conferences.
BBSes can be standalone (not connected to other BBSes), networked (usually a network of BBSes like FIDONET), and/or connected to the Internet. Though only marginally part of the Internet, BBSes are a widespread, low-cost information resource. Note that a Bulletin Board is very different from a Usenet Newsgroup.BBSes usually support 14.4k baud V.32 modems (or faster) and Zmodem file transfers. Whenever possible, use Zmodem for file transfers; it is by far the best protocol. BBS and Internet Service Provider modems usually default to 8-N-1 (8 data bits, No parity bits and 1 stop bit).