Third National Flag of the Confederate States of America |
P. A. Stonemann, CSS Dixieland |
National Jack of the Confederate States Navy |
CSS Dixieland
Probing the depths of knowledge
These essays by P. A. Stonemann, CSS Dixieland, cover a wide range of
historical, philosophical, scientifical and technical subjects. Each page
deals with a particular topic, divided into sections and explained by itself.
Every page shows at its top hyper links to every other page. The Start page
also has short descriptions of the other pages. CSS Dixieland expresses
gratitude to the readers that make this work meaningful.
This Web document has been tested with KDE Konqueror, graphic HTML interpreter
for Linux. It may not be rendered correctly by other graphic HTML interpreters.
It will probably be correct when rendered by text-only HTML interpreters (visual,
aural, or Braille tactile interpreters), but if feasible, please use KDE Konqueror.
Uniform Resource Locator:
Internet Gopher Protocol page
The most simple and uniform hyper text protocol, free of commercialism
Sections in this page
Introduction, chronology of hyper text
Technical note: In languages other than English or Latin, but which use mainly
Latin characters, some characters are taken from other alphabets, or some Latin
characters are modified with diacritic marks for representing different phonemic
sounds or other orthographic conventions of those languages. Those characters,
when used in this document, have been encoded as entities of Hyper Text Mark-up
Language or sometimes in Unicode UTF-8. Therefore computers using other character
encodings may render some characters inaccurately, but hopefully, it will still
be possible to read non-English words without too much difficulty.
Introduction, chronology of hyper text
The idea of hyper text is surprisingly simple: to facilitate the location and
retrieval of any information or piece of knowledge. A forerunner of the hyper
text idea is the standard reference or citation. On paper this can be done by
indicating, for instance, the book, chapter, section... of a text that bears
relation to the subject that is being discussed, or by pointing to the page
and line of a specific edition of that book.
With computers, that reference or citation can be done as well. The pointing
link and the pointed resource may be both located inside a single isolated
computer, or in a small local area network with a few computers, or in a wide
area network with many computers, or in the entire planet with millions of
computers or other connected devices. Even in outer space, by means of
communications to or from astronautical vehicles.
The efficiency of the process will depend on various factors, such as the
reliability of the computers and of their connections, the interoperability
of the methods being used, the expertise of those who write the hyper links
and of those who read them, the sustained storage of linked resources in a
stable location, and other factors. In ideal circumstances, computers can
reach a needed information in an instant.
Internet is since the late XX century and as of 2016 the most common medium.
All kinds of text, even full books, can be read through Internet using
different protocols: Wide Area Information System Protocol, File Transfer
Protocol, Gopher Protocol, Hyper Text Transfer Protocol and its Hyper Text
Mark-up Language (or other standard languages of the World Wide Web), or by
some other protocols. Of those, only Gopher and HTTP-HTML (WWW) are forms of
hyper text generally available. WAIS or FTP are not hyper text, properly
speaking, although they can of course indicate the location of a resource and
the method for retrieving a copy of that resource.
Chronology of hyper text
About 1945 Doctor Vannevar Bush (Massachussetts Institute of Technology)
published the essay 'As We May Think', where he describes his vision for a
computer aided hyper text system that he named 'Memex'. His description of
browsing the Memex for finding linked information includes the ability of
easily inserting new information by its different users, thus adding to the
growing knowledge base of Memex, as the hyper text system does today in the
Gopher Protocol, or in the Hyper Text Transfer Protocol and Mark-up Language
used by the World Wide Web.
In 1963 Douglas Engelbart (Stanford Research Institute, Director of the
Bootstrap Institute) created with Joseph Licklider the Augmentation Research
Center. He later helped to develop hyper text.
In 1967 Ted Nelson coined the term 'hyper-text' in his book 'Literary
Machines and Dream Machines', which described hyper media and advocated for a
global hyper text system that he named 'Xanadu'.
In 1969 began the distribution of 'Request for Comments', or RFC, method
developed by S. Crocker (U. C. L. A.) for interchange of ideas and proposals
among researchers. Initially distributed by the physical postal service, the
Requests For Comments became commonly distributed in later years through File
Transfer Protocol. Since the 1990's, they can also be accessed as hyper text
documents via World Wide Web. They are edited and coordinated by Jon Postel
(Stanford Research Institute), and have become the technical standard on
which Internet is based.
On 2nd September 1969 Leonard Kleinrock performed the first transmission of a
short distance message between two computers, both inside U. C. L. A. and
connected by a 5 metre cable. The computers interchanged meaningless data
while about twenty people watched the historical event, which MARKS THE START
OF INTERNET. Because of its success, the Centre for Network Measuring of
Leonard Kleinrock at U. C. L. A. was chosen for the installation of the first
Interface Message Processor of Heart and Newman, THUS CREATING THE FIRST
COMPUTER HOST-SERVER. Days later, the Human Intellect project of Douglas
Engelbart (that included NLS, a forerunner of hyper text) was installed at
the Stanford Research Institute, becoming the second computer host-server.
Stanford maintained the Host Name list for the mapping of directory and
addresses of Request For Comments.
In July 1972 Lawrence Roberts added to electronic post the functions 'List',
'Read selectively', 'Save', 'Forward' and 'Reply', making the SMTP and POP
Protocols the most important Arpanet and Internet applications until the mid
1990's, with File Transfer Protocol and Telnet Protocol following suit, all
of them overpassed since the mid 1990's only by Hyper Text Transfer Protocol.
In 1981 Ted Nelson made Xanadu operational, an early hyper text system.
In 1985 appeared 'Xerox Note Cards', a hyper text system based on Lisp
programming language.
In 1986 'Owl Guide', professional hyper text system for large scale
applications, inspired on Xerox Note Cards.
In 1987 'Hyper Card', programme to create graphical hyper text documents, by
Bill Atkinson (Macintosh). Mister Atkinson was also known for 'Mac Paint',
bitmap painting programme. Hyper Card featured bitmapped graphics, form
fields, scripting, and full text search. Hyper Card spawned imitators such as
'Asymmetrix Toolbook', that created drawn graphics and was executable in the
IBM Personal Computer.
Also in 1987 there was a workshop on hyper text systems in Chapel Hill, North
Carolina, helping to form Siglink in 1989 (later renamed ACM Sig Web),
organisation that has for many years centred attention with its annual
conferences on most of the academic research on hyper text.
In 1989 came the proposal of a World Wide Web, system of linked information
able to work with different kinds of computers, by Tim Berners-Lee and Robert
Cailliau (Centre d'Etudes sur la Recherche Nucleaire, Genevre, Switzerland).
Most scientists of C. E. R. N. used TeX or Post Script for their documents at
that time, others used Unics Troff, a few used Standard Generalised Mark-up
Language. Mister Berners-Lee realised that something simple was needed to cope
with dumb terminals through high end graphical X Window Unics work stations.
His proposal was thus a simple Hyper Text Mark-up Language based on SGML, with
an also simple network protocol that he named Hyper Text Transfer Protocol.
In 1990 began operations Archie search engine, for File Transfer Protocol.
In late 1990 came the public release of the HTTP-HTML 'World Wide Web' by
Mister Tim Berners-Lee (Centre d'Etudes sur la Recherche Nucleaire, Genevre,
Switzerland), and of WWW-Talk posting list.
In 1991 came the release of Gopher Protocol (University of Minnesota). Gopher
is mainly used for transmission of text, although some Gopher clients present
a graphic interface. Gopher went into a slow continuous decline after the
spread of the HTTP-HTML 'World Wide Web' of Mister Berners-Lee. In terms of
traffic (packet exchange) the Web overpassed FTP, Gopher and most other
Internet protocols about 1995, but as of 2016 there are still over a hundred
active Gopher servers. Gopher presents a list of menus from which a document
or another resource can be chosen.
In 1992 was published the Hyper Text Mark-up Language, a proposal by Tim
Berners-Lee and Robert Cailliau, loosely based on SGML. It was never official
(there was never an HTML 1.0), but the CERN-Convex-Atrium specification of
June 1993 is now known as 'HTML 2.0'.
In November 1992 appeared Veronica, search engine for Gopher Protocol, and in
1993 Jughead, a more limited search engine for a single server of Gopher
Protocol.
In 1993 appeared Lynx, text-only user agent for many network protocols,
working originally in VMS or Unics operating systems, later with ports to DOS,
Macintosh, Windows, BSD, Linux and other platforms and systems. Among the
many protocols natively supported by Lynx are Gopher and HTTP-HTML of WWW.
History: past and present of Gopher
In the late 1980's there were hundreds of centralised computer networks or
other computer connections belonging to public or private organisations of
various sorts, such as government, academic, commercial, recreational, or
other entities. Some were big, like AOL, Compuserve, Genie, Fidonet, Netcom
or Prestel, others had a scientific or technical importance, such as Arpanet,
NSFnet, Bitnet, Janet, EARnet or Vnet, others were communication protocols, as
SMTP-POP, FTP or NNTP Usenet, but most of them were simply small centralised
connections known as 'bulletin boards', similar in functionality to the Web
fora that are more common as of 2016. Communication was not done in real time,
instead, messages were sent to the bulletin board, which made then available
for other persons to read or copy.
A bulletin board had a temporary connection of only a few hours per day
(usually nocturnal, in the location of the central computer), and it was
often managed by one or two individuals called 'system operators'. Client
computers communicated to the central computer by dial-up using a telephone
line and a modulator-demodulator that converted digital instructions into
analogue signals, and viceversa. The bulletin board might have several
telephone lines, for several incoming computer connections. Big bulletin
boards had sections on different topics, small ones tended to be specific on
some particular subject of interest, perhaps a very specialised subject. All
had a set of rules on what was acceptable behaviour in that particular board.
Most bulletin boards or other networks were not a part of Internet, although
some of the biggest ones were connected to Arpanet and later to the backbones
of the National Science Foundation that became Internet. There were networks
such as UUCP (Unix to Unix Copy, for computers using one of the many variants
of Unics operating systems), where the location of an address or resource is
indicated by a path composed of exclamation marks and, optionally, also by
curly brackets for mentioning alternative hosts, as in the fictional example:
The vast majority of those bulletin boards or proprietary networks either
disappeared, or were absorbed by Internet in the mid 1990's. Some of them
survived for years as independent networks, such as Fidonet, or may even have
survived until today in 2016, albeit certainly in a limited form, without the
importance that they once had. Most of those still in operation are now part
of Internet. In Internet proper, exist or have existed various protocols for
communication among computers. The most important protocols are or were those
that make possible the fundamental functions of a computer network:
-Transmission Control Protocol (1975), which in January 1983 replaced the
Network Control Protocol (1969-1982). The Transmission Control Protocol was
implemented for sharing resources of any network, not only of Arpanet, while
the Network Control Protocol had been specific to Arpanet.
-Internet Protocol and User Datagramme Protocol (1978), for separating the
function of routing packets of data from various other network functions.
Besides the above, other protocols were devised for particular purposes:
-Telnet (1969) and later TN3270, for emulation of terminal, a working session
with a remote computer.
-Simple Mail Transfer Protocol and Post Office Protocol (1970 for the first
version, version 3 is current in 2016), for messages from one origin to one
or a few destinations. SMTP-POP was also much used as a gateway to many other
protocols or services (see in this page the sub-section 'Gopher, Veronica,
Jughead by electronic mail', in the section 'Resources available for Gopher').
-File Transfer Protocol (1972), for downloading all kinds of data sets, such
as images, sounds, text documents or software. It inspired creation of Gopher.
-News Network Transfer Protocol 'Usenet' (1980), for reading or writing public
messages. Created by Steve Bellovin, Jim Ellis, Tom Truscott and Steve Daniel
(Duke University). Usenet was initially carried by UUCP (Unix to Unix Copy),
but in the early 1990's most servers passed to Internet. Usenet was a very
common resource for collective programming projects, or for other forms of
interaction among computer enthusiasts, during all of the 1980's and most of
the 1990's. Unfortunately, in the mid and late 1990's many ignorami got easy
access to Usenet, and since most news groups were not monitored against abuse,
they gradually fell into spam, flamage, 'holy wars', and harsh language.
-Finger, for finding all potential destinations of a message, receiving in
reply an optional .plan data set with informative text stored at destination.
-CSO 'ph qi' (names, telephones and addresses of individuals, mostly academic).
-Wide Area Indexed Server (a kind of enlarged campus-wide information system).
Other protocols existed, but those named were perhaps the most commonly used.
Communications were done mainly by physical teletype or telephone land lines
(wireless only used for long distance communication among the main nodes),
and the terminals often featured a teleprinter or a typewriter for input, a
paper printer or a screen for output, perhaps a few other peripheral devices.
1989-1991: The introduction of hyper text
The File Transfer Protocol, FTP, was necessary for collaborational work among
academics, researchers, programmers, or other persons involved in developing
Internet or any of its parts. A common method of access to a remote computer
by FTP was known as 'anonymous log-in', and data sets open to public download
were usually placed in a directory called 'pub' or similar name. Anyone with
the necessary hardware and software could obtain whatever might be available
in that public directory. But that was not hyper text, FTP required a log-in
into the storing computer, although it were an anonymous log-in. Because the
concept of hyper text already existed (as it has been shown in the Chronology
above), it was a matter of converting it into a general Internet protocol.
First the Centre d'Etudes sur la Recherche Nucleaire (in Genevre, Switzerland)
and then the University of Minnesota, gave the earliest successful steps in
that direction. In Switzerland, a team presented in 1989 a proposal for a
Hyper Text Transfer Protocol, rendered by a Hyper Text Mark-up Language based
on Standard Generalised Mark-up Language. It was released in late 1990 with
the name of 'World Wide Web'. In Minnesota, another team engaged in hard work
during 1991 into developing a protocol that could accomplish all that FTP
accomplished, but without need of log-in. It was released in late 1991 with
the name of 'Gopher' (a gopher is a small burrowing mammal typical of that
region, mascot of the University and of the State of Minnesota, as well as of
a sports team called 'The Golden Gophers'). Gopher was immediately acclaimed
by the press as an impressive feat, a notorious enhancement over FTP. In very
short time, thousands of public or private organisations had created their
own Gopher documents, and placed them into a growing number of Gopher hosts,
whence the contents could be served to client computers just by activating a
hyper text selector, or by Telnet. FTP continued to exist until today, but
Gopher knew a golden era, lasting from late 1991 to late 1994 or early 1995.
Most other protocols have not been so fortunate as Gopher and FTP, which have
both survived until today. Protocols such as Finger, WAIS, CSO, TN3270, and
many others, are for all practical purposes gone as of 2016, or they are now
little more than historical (a few Unics enthusiasts still keep in their home
directory a .plan data set, readable by the Finger Protocol). NNTP Usenet and
Telnet still exist, but they are only shadows of their former selves. Usenet
can today be accessed from Google Groups via World Wide Web, having increased
irrelevant content posted by idiots who do not even know of the existence of
NNTP. They think of it as part of Google and of the Web, much to the anger of
long-time Usenetters, who already had a high noise to low signal ratio before
Google (in September 1993, America On Line permitted thousands of stupids to
post to Usenet, seriously damaging the average quality of Usenet messages).
As for SMTP, POP3 or IMAP they continue in regular use, but nowadays they are
very often accessed via World Wide Web. Electronic Mail clients are dwindling.
Thus, the Hyper Text Transfer Protocol of the Web has become a hungry monster
affected by an insatiable 'protocolphagia', having devoured many of the other
protocols into itself. Many browsers (intended for ignorant consumers) assume
an Internet address to begin with 'http://', it is not even necessary to give
the prefix of the protocol. Donkeys know nothing about protocols.
This has been the case not only with Internet protocols, but also with most
other forms of communication, which have been negatively affected by the Web
or by other resources that have grown very rapidly in few years: video, mobile
telephone, instant message systems. Newspapers and other publications printed
on paper still exist, but they are dwindling. Short distance communication
resources like teletext or videotext (which were often carried by television
signals) knew some success but are now practically gone. Analogue wireless
radio or television has been in many areas substituted by cable services or
also by digital signals, especially television. Teletype (teleprinter) and
traditional telegramme are today almost exclusively used by ships at sea, or
in places where no other resources exist. Even long distance, trans-oceanic
communications like Short Wave (High Frequency) wireless radio, have known a
dramatic shrinking with the spread of the Web. Many international broadcasts
have recently moved to stream audio in the Web, or have been seriously limited
or completely closed down in the last years, mainly in West European countries
(Radio Portugal, Radio Exterior de Espana, Radio Netherland, Radio France
Internationale, BBC World Service, Deutsche Welle, Yle Radio Finland, RAI
Italiana, Radio Poland, Radio Moscow 'The Voice of Russia', and many others).
Fewer ionospheric broadcasts still remain for the surviving enthusiasts of DX.
Korea, China, Turkey, Romania, Czechia, Slovakia, and some private or semi
official North American stations, are among the few that still invest in
trans-oceanic Short Wave. Most of the existing Short Wave is now intended for
regional audiences. Long Wave radio is almost gone (except for a few utility
stations), and even Medium Wave has almost disappeared from most advanced
countries, first in the competition against local Frequency Modulation (of
shorter distance, but with better sound quality), and now also against the
Web, which offers the best audio and video to those fortunates who have a
suitable computer and Internet connection. For similar reasons, many public
cinematographs have become forced to attend only a limited segment of the
public, or they have been absorbed by official institutions, or they have
closed down and been converted into commercial centres, exhibition areas,
television studios, churches, or so on. We remember the days when we enjoyed
reading a 'pulp' book or magazine, when we watched an 'old fashion' motion
picture, or when we listened the faint sound of a remote Short Wave station
in a thermionic valve wireless (vacuum tube radio) receiver, or of a Long or
Medium Wave station in a piezo-electric (crystal) receiver built at home.
Likewise with a long list of things that were common at one time, but now are
very rare or completely gone: phonograph cylinders, gramophone discs, magnetic
tapes (wide format, eight track cartridge, or cassette), cellulose or acetate
films for chemical Photography or Cinematography, and endless of other things.
Such is the evolution of technical standards. Old or uncommon standards may
see themselves swallowed or displaced due to a number of causes, not always
necessarily because of 'being bad', but because of the complex interaction of
various factors, especially commercial pressure. The present essay proposes
to analyse some of the causes that provoked the rise and fall of the Gopher
Protocol, to offer to the reader a reasoned explanation of that process, and
to advocate Gopher as still a perfectly valid protocol, efficient until today.
A simple and uniform hyper text protocol, free of commercialism.
1991-1994: The Golden Years of Gopher
The University of Minnesota was initially thinking of a campus-wide system of
shared information. Besides the protocols mentioned above (and of others not
mentioned), the 1980's saw the development of various systems of distributed
data sets, such as the Network Filing System or the Andrew File System. Some
of them became generally used, even commercial (sigh), while others had more
limited use or remained in the experimental stage. Those systems inspired the
University into creation of a new campus-wide information system, with some
ideas taken from a number of those predecessors, and with other ideas evolved
inside the University itself, by dedicated teams of scientists and engineers.
Two options presented themselves: or one mainframe computer with many dumb
terminals connected to it, or one or more host computers that could serve
information to any client computer that requested it. Still under the strong
influence of 'Big Blue thinking' (affectionate name given to IBM Corporation),
the adherents of the mainframe option were winning the debate, but those who
were convinced of the feasibility of the host-client option took the bold step
of declaring that they could make their solution work in a short time. The
University gave them a chance to prove their optimistic assertions. And good
that the University took that historical decision. It was the year 1991.
The optimists were these gentlemen: Farhad Anklesaria, Robert Alberti, Paul
Linder, Mark MacCahill and Daniel Torry. The concoction of their privileged
brains included programmes acting server-side, those acting client-side, and
the protocol for communication between them. The client computer could get
the desired information by one of two methods: or by using a Gopher programme
installed in the client, or by starting a remote work session via Telnet and
using a Gopher programme installed in the host. Either way, the client then
received a Gopher document, whose first page was typically a menu listing
other menus or directories, text documents, searchable data bases, protocols
such as Telnet or WAIS, or other resources (later, even images or sounds).
All text was encoded in 7-bit ASCII (American Standard Code for Information
Interchange), but a document of text was formatted differently from a menu,
using numerical ciphers from '0' to '9', some alphabetic letters, tabulation
characters, commas, periods, slashes or other formatting marks. Needless to
say, all was done by command line, as it should be. Graphic interfaces with
funny icons already existed, but were not used. Everything was fine and dandy
for those computer operators who are always too lazy or too stupid for serious
study of technicalities, because by activating a Gopher selector the requested
resource appeared on their screen like if by magic, without needing to bother
about paths, names and types, data set systems, operating systems, softwares,
platforms, architectures, or other aspects of Internet too 'cryptic' for them.
Therefore, nearly from night to morning the Gopher Protocol and the University
of Minnesota became famous. Headlines, reports and interviews abounded. Major
newspapers and other public media praised the initiative of the Minnesotans.
A new era had begun for Computing. Thus the original idea of a campus system
disappeared in a puff of smoke, it was perfectly clear that Gopher was much
too powerful for being confined within the narrow limits of academic campi.
So, from late 1991 to late 1994 were the three Golden Years of Gopher, when
lauding comments on the protocol were in every mouth, newspaper, magazine,
wireless or television station. When the Sun shone in Minnesota, when the
gracious symbol of a gopher illuminated every computer screen. But in the
midst of all that ephemeral glory, the inextricable paths of History still
had a long trek to be walked. The unforeseeable future was biding its time.
The Norns of Destiny were already weaving the oblivion of Gopher...
The apparent 'coincidences' of History
In the realm of scientific or technical research, discoveries or inventions
no rare have appeared in more than one place, in form apparently coincidental
on identical or very similar fields of study, and almost at the same time, by
two or more researchers or teams of researchers who were unaware of the work
of one another. Dimitri Mendeliev was not the first one to classify chemical
elements into a periodic table, Lothar Meyer had already done that, though in
a less developed manner. The self-propelled heavier-than-air flying machine
(the aeroplane, different from the balloon, the airship, or the glider) was
invented by either Santos Dumont or by the Wright Brothers, both claims count
with stern defenders until today (not forgetting the important contribution
of Clement Ader and the pioneer effort of Otto Lilienthal). The infinitesimal
calculus (differential and integral calculus) came from the genius of Wilhelm
Leibnitz or from the not lesser genius of Isaac Newton, albeit using different
methods (which proves that one mathematician did not plagiarise the other).
There is a list of names of those who invented various photographic processes
and who could almost equally claim to have been the 'inventor' of Photography:
Thomas Wedgewood, W. H. Fox Talbot, Nicephore Niepce, J. L. M. Daguerre,
Hyppolite Bayard, Hercules Florence... So with Computing, apart from the
pioneer work of Charles Babbage, there is Konrad Zuse, J. V. Atanasoff, J. V.
Neumann, P. Eckert and J. Mauchly, Alan Turing, and others. The explosion
engine is named 'Otto' in Germany and 'Beau de Rochas' in France, after its
alleged inventors. The telephone has Elisha Gray and A. G. Bell as separate
inventors. So with Cinematography. So with the submersible ship (submarine).
So with a surprising amount of other examples. For those of us who love the
History of Science, and who despise chauvinistic claims, those coincidences
prove that more than one mind is focused on a given problem at a given time.
Of course, those 'coincidences' are more apparent than real. T. A. Edison once
said that invention is one percent genius and ninety-nine percent hard work.
That statement gives a clue to the 'mystery' of such coincidences. What happens
is that the composing elements of a discovery or invention already exist, but
separated. The enviromental conditions that make possible that development
are already present, but not obvious to everyone. Society is ready to make
use of the new gadget, but that social acceptance cannot be seen (obviously)
until the gadget in question be offered in the streets. Thus one team worked
in Minnesota and the other in Switzerland, both keen on developing hyper text,
and without knowledge of what the other team was doing at the other side of
the Atlantic. The former team was Gopher, the latter was the World Wide Web.
1990: The competition enters the scene
In truth, it is a common error to think of Gopher as 'older' than the Web.
The proposal of Messieurs Berners-Lee and Cailliau for the hyper text system
that they named 'World Wide Web' came in 1989 (see the Chronology above), its
public release in late 1990, together with WWW-Talk posting list. The first
'proposed' HTML specification was published in 1992, and the first 'official'
HTML specification finally saw the light in June 1993 (now called HTML 2.0).
By comparison, Gopher was born in late 1991, by the frenetic work of a small
team of specialists. It did not exist at all before 1991 (unless, perhaps, as
a project in the mind of some of those specialists). Therefore, the Web is
older. Also, the first implementations of the World Wide Web were not very
different from Gopher, as of user interface. Everything was a 'wall of text',
with no images or sounds, in Gopher or in the Web. Images and sounds came
later, for both. The Web developed those visual and aural stimuli to their
current maximum, while Gopher remained in a comparatively 'primitive' stage.
Louis Pasteur once declared that 'Science knows no national boundaries,
although the individual scientist may have a national loyalty'. He himself
was an ardent French patriot. But one thing is patriotism, and quite another
is chauvinism, provincialism, parochialism, and narrow mindedness. In the
Centre d'Etudes sur la Recherche Nucleaire of Genevre there was (and there
is) a sizeable number of brilliant minds from all over Europe, who were (and
are) in daily contact with the rest of the World. Scientists and engineers
of many nationalities, by necessity fluent polyglots (capable of reading,
writing, listening or speaking in several languages), who were engaged into
some of the most advanced research that History has seen, and who were
supported by many institutional agencies and governments through the Conseil
Europeen pour la Recherche Nucleaire. In 1990, the year before the birth of
Gopher, the CERN enthusiastically endorsed the proposal of Messieurs
Berners-Lee and Cailliau, and began to use the World Wide Web inside the
heterogeneus network of European computer systems.
An international Web against a provincial Gopher
As opposed to the professed internationalism of the Web, Gopher began as an
internal information system for a single academic campus, where language
barriers were not an issue at all (everyone in the University spoke English,
and interacted with the others ONLY in English), and where the heterogeneity
of computer systems was a relatively less important issue in Minnesota, than
it was in Europe. Then, in spite of the huge boom that Gopher saw between 1991
and 1994, the mindset of most of those people who were responsible for the
development of Gopher, continued as essentially a provincial mindset. Several
examples can be cited in support of that hard accusation of provincialism.
One is the symbol. The Minnesotans chose their most typical animal, a gopher,
nearly unkown outside North America except by those who are interested in
Zoology. The Europeans chose (later) a globe representing the whole Earth,
much in the vein of the centuries-old Tradition of voyage and exploration
that had begun with Phoenicians, Carthaginians, Greeks, Romans, and some
other ancient peoples, and had continued in Europe with Scandinavian Vikings
and Varangians, Normans, Portuguese, Castillian, Basque, Catalan-Aragonese,
French, Breton, Dutch, Flemish, English, Scotch, Venetian, Genovese, Prussian,
Austrian, Russian, and other European peoples, even from minor countries.
In a similar vein of provincialism, Gopher did not seriously consider the use
of human languages other than English. The mostly monolingual Minnesotans saw
no need of bothering themselves with language support. There is a history of
Scandinavian migration in Minnesota, but nearly everyone speaks English today.
The Gopher team 'assumed' that everyone in the World can at least understand
English. While this is partly true for those who are used to foreign travel,
who are in frequent contact with foreigners, or who enjoy languages, it is by
no means the case with a vast majority of the population in regions where,
like in Minnesota, culture is closed into itself. There are millions in huge
areas of Central and South America, Asia, Africa, and even in the South and
the East of Europe, who cannot understand more than a dozen words in English.
A really international language fluently spoken by everyone has never existed.
Accadian, Persian, Greek, Latin, Arabic, Turkish, Italian, Portuguese, Spanish,
French or German have had their high tide as languages of diplomacy, commerce,
culture or science, but only for a relative minority outside the regions where
those languages are or were commonly spoken. Efforts of promoting artificial
languages such as Volapuk, Esperanto or Interlingua have seen limited results.
By contrast, the Web was born in a polyglot environment, and this fact was
reflected from the start. Initially there was support for diacritic marks on
Latin letters, which are standard for accented characters, distinct phonemic
sounds, or orthographic conventions in many languages where some variant of
the Latin Alphabet is in use. Later, the World Wide Web Consortium embraced
Unicode. This is a 16-bit character code intended to cover all of the world's
writing systems, including Latin, Greek, Cyrillic, Chinese, Japanese hiragana,
katakana or kanji, Devanagari, Thai, Laotian and many other scripts, even with
plans for future support of the fictional Elvish languages of Middle Earth,
from the immortal literary work of J. R. R. Tolkien (Angerthas Runic Alphabet
of the Dwarfs and Tengwar Cursive Alphabet of the Elves). With a set of up to
65 536 characters (2^16), Unicode can perfectly cover all those scripts. Not
all Web software supports the whole Unicode, but many Web servers or clients
implement at least the Latin, Greek and Cyrillic subsets of Unicode, serving
the needs of almost every European language in current use, and of many other
languages, not in current use or not European, which also use those alphabets
or variations of them.
There is little of that for Gopher. Motsognir Gopher server (see below in this
page) supports 8-bit ASCII or UTF-8 characters, but Motsognir is a remarkable
exception of recent introduction into Gopher. Until now, publishing in Gopher
in a language other than English meant that certain characters would not be
correctly displayed by every computer terminal. There are several major groups
of character encodings. One is EBCDIC, started in the early 1960's by IBM. It
has at least six varieties, but today it is only used by mainframe computers.
Another encoding is ASCII, the most important at present, whose 7-bit standard
(128 characters) began in 1961 and is now used by almost all computers. 8-bit
Extended ASCII (256 characters) came later in two main varieties: IBM ASCII
(used for example in DOS) and ANSI ASCII (a standard as ISO-8859-1 Latin 1).
There is an EUC encoding for Japanese, Korean and other mainly Asian tongues,
with over 10 000 characters. Finally there is Unicode ISO 10646, with 65 536
single characters, but capable of combinations giving millions of characters.
Unicode can be variously encoded, the most common encodings are UCS-2 and
UTF-8, less common are UTF-16LE, UTF-16BE, UTF-32LE, UTF-32BE (LE is 'Little
Endian' and BE is 'Big Endian', in reference to the order of bits). Unicode
is supported by most operating systems of recent implementation (not by DOS),
and by the Web, but not by Gopher (except in Motsognir, and only for UTF-8).
More information on this encoding of the highest importance at:
Unicode, Universal Encoding
Gopher disdains voluntary help
Another example of provincialism is the lack of a board, forum, posting list,
or another resource for interchange of ideas on the development of Gopher.
The Web understood the importance of such interchange, and it began regular
interaction with interested persons almost from its inception, with WWW-Talk
posting list. By contrast, Gopher did NOTHING of the sort. Anklesaria et alii
released their brilliant work, and that was all. Period. It did not occur to
them that there might exist what Mister Berners-Lee called 'evolvability' for
the Web, that other persons outside the University might help in the further
development of Gopher, and that their enthusiam had to be mustered through
some common channel (like Source Forge now is for development of software).
Mister Anklesaria and his associates later published 'requests for comments'
for Gopher, in 1993, when the Web was already well established and announcing
itself as a strong competitor to Gopher. But in factual reality, RFC 1436
(The Internet Gopher Protocol) and RFC 4266 (The Gopher URI Scheme) were both
merely informative documents, without any serious purpose of interaction with
other programmers who might be able and willing to help with the growth of
Gopher into maturity. Thus the cute little Gopher remained a child, for ever.
As of 2016 there are lots of resources for the Web, but only two resources
for those who are interested in Gopher and wish to interchange ideas. One is
a posting list, the other is a Usenet news group. Unfortunately, many of the
messages seen in those two Gopher resources are not really useful, or they
are not relevant at all. A number of messages are of the nostalgic variety:
they weep over the loss of Gopher, or they complain on the causes that have
made Gopher almost disappear, or they blame (fairly or unfairly) the guilty
parties who have brought Gopher to its nearly demise, but they do not offer
any workable solution. Other messages are of the brainless 'modernising'
variety: they label Gopher as an 'archaic' protocol, and are glad to see
Gopher almost gone. There is also spam, especially in Usenet. A few Gopher
enthusiasts, in particular a tiny group of Gopher programmers, still post
relevant information, but finding that information is difficult, it means
looking at tons of uninteresting garbage that pollutes the posting list and
the Usenet group. Not surprisingly, the very few who are serious about Gopher
remain largely invidual heroes. No significant collective efforts have been
done for many years. Most of the general public do not even know what is the
Gopher Protocol, and they are completely unaware that Gopher ever existed.
These are the two resources that currently exist for Gopher development:
The Gopher Mailing List (accessible by SMTP, POP3 or IMAP protocols):
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/gopher-project
The Gopher Usenet News Group (accessible by News Network Transfer Protocol):
comp.infosystems.gopher
The Gopher Mailing List is hosted at a Uniform Resource Locator of the World
Wide Web, and served thence to electronic addresses of interested persons.
The Gopher Usenet News Group can also be accessed via Web, through the
gateway services provided by Google Groups. Unfortunately, none of the two
resources can be accessed directly from Gopher.
There are in RFC 1436 and RFC 4266 strong indications of extensions that had
been planned to the Gopher Protocol, but which were never implemented. One
example is the requirement of a dot 'standing in a line of its own', which
must be included by the server (it does not need to exist in the source).
After having carefully studied the Gopher specification, the purpose of that
solitary dot was so mysterious, that P. A. Stonemann was impelled to consult
the well known Gopher expert Mister Mateusz Viste. The expert confirmed that,
in effect, the dot had been ruled in preparation of an intended feature that
never became a reality. The small Gopher team at the University of Minnesota
was limited in man power. So, however intelligent and dedicated, that handful
of geniuses could never effectively compete against the uncountable crowds of
Mongolian hordes that were massing behind the World Wide Web. The Gopher team
might have followed the strategy of proposing extensions to Gopher and then
recruiting volunteers for the project, of brain-storming Gopher enthusiasts
for new ideas, but the team chose not to. It had a short-sighted tactics.
The mysterious 'dot on a single line' only affects the termination of Gopher
item type zero (a text document) or Gopher item type one (a directory menu),
it does not affect other Gopher item types. Gopher item types zero or one
having any line starting with a dot (a period), must have another dot added
by the server (for preventing confusion with a termination indicator), then
the said dot must be removed by the client. Nice complication. Request For
Comments 1436 is specific about all this rigmarole:
'Lines beginning with periods must be prepended with an extra period to
ensure that the transmission is not terminated early. The client should
strip extra periods at the beginning of the line'.
The intention of Mister Anklesaria and his Council of the Wise was double:
First, they were considering implementing a 'persistent connection' feature,
as seen in Hyper Text Transfer Protocol version 1.1, which would allow a
Gopher client to fetch several Gopher documents within a single Transfer
Control Protocol connection (instead of having to establish a separate TCP
connection for each resource), thus slightly decreasing the lag of the
protocol, possibly perceived on hyper links with a very high latency. This
might be interesting, but the added complexity versus the supposed benefit
makes the 'solution' highly ineffective. Of course, computer speeds and
Internet connections of the early 1990's were much slower than they are in
2016, but even so, the difficulties do not seem to compensate the results.
Second, the terminating dot that appears solitary in its own line is also an
easy way for the Gopher client to know that effectively it has received the
entire content of the Gopher text document (type zero) or Gopher directory
menu (type one). Otherwise, the content may have been truncated at some point
during the transmission of packets of information. This verification should
really be done by monitoring the correct ending of the TCP connection, with a
FIN/ACK - FIN/ACK exchange, but such verification is comparatively more
complex for the Gopher client than it is the trivial checking of whether or
not a single dot trailer exist in the received Gopher document or directory.
This is an unfortunate feature, because it encourages lazy programming of the
client software. Besides, the Gopher Protocol as defined in RFC 1436 is rather
inconsistent in this approach, because it does not require the solitary
terminating dot for any other Gopher type (such as an executable), which means
that a client programmed in the lazy way will fail detecting corruptions of
all other resources (unless using a check sum or some other verification).
In sum, the Geniuses of Minnesota were overloading themselves with hard work,
not really knowing whether their ideas would find a continuation or not. In a
scenario of software development by the collective efforts of many minds,
this tentative 'enhancement' would have been hardly the case, because there
is always someone who turns the attention of the others to the advantages or
disadvantages of this or that proposal. But in the holy shrine of a reduced
group with a well established hierarchy, of uncontested authority, there is a
risk of a 'proposal from above' being taken by the acolytes as an 'order from
above'. No one dare point to possible problems that might arise, then or in
the future. Result: the proposal goes directly into official specifications,
without any further discussion. No wonder that, outside the sacred recint,
'heretic' branches begin to appear, more or less chaotic, more or less in
tone with the original 'orthodox' dogma. However well intended they might be,
any deviations in the form of 'enhancements' or 'extensions' pose problems of
lack of uniformity, which may seriously endanger general acceptability.
The 'extensions to Gopher' known as Gopher Plus and Hyper Gopher, or the more
recent implementations such as Gopher Virtual Reality, never enjoyed a wide
acceptance. Gopher VR is workable, but it is almost unknown, except to Mister
Cameron Kaiser and a relatively reduced number of Gopher enthusiasts. Gopher
+ or Hy-Gopher are now in the trunk of old memories, together with so many
other reminiscences that Time forgot, of a past that 'might have been'. The
Gopher Conference of 1992 (Gopher Con'92) was the high tide for Gopher, with
about fifty delegates attending and with important new ideas coming from it.
The original Veronica search engine for Gopher released in November 1992, and
Jughead in 1993, were two other boosts for Gopher. And so were other efforts,
which the University could very well have guided and helped. Instead, the
University chose to let time fly by, without any significant input. P. A.
Stonemann, author of this essay, contends that the University of Minnesota
bears a huge responsibility for the near demise in which Gopher is as of 2016.
Gopher demands his pocket money
The 'coup de grace' came when the University made public the possibility of
exacting payment of money for legal royalties, from any 'commercial' uses of
Gopher, though not from 'educational' uses. We all agree that commercialism
is nothing short of repugnant, but it may be difficult to determine who is
'commercial' and who is not (albeit often it is quite clear), and besides, we
voluntary programmers help a software project because we love the concept,
but we go immediately out if we even suspect of someone making money from it.
The solution would not have been to impose royalties, but to enact legal
licences where all software development must be Free Use and Open Source,
free for everyone to use or to modify, with minimal limitations. Only with
the requirements of keeping it free, of acknowledging the original authors,
of labelling derivative works or distributions as being the responsibility of
other persons, and little else. The model is perfectly represented by the GNU
Free Software Foundation of Mister Richard Stallman. Other free licences might
have been devised by the University, instead of demanding money for legal
royalties. The demand for money was never enforced, but it had already caused
damage, spoiling the feeling of trust that many had initially held for Gopher.
The guiltiness of the University could be argued, especially by those who
were involved at that time, but what is incontestable is that the doubtful
idea of the royalties was a serious blow against a still infant Gopher. This
is bluntly declared by the foremost competitor to Gopher, by the Father of
the Web, Mister Tim Berners-Lee. In an interview of August 1996 (when the Web
had already surpassed Gopher by far), Mister Berners-Lee made it clear that:
So, the inventor of the Web recognised that Gopher was preferable in terms of
simplicity, but that the lust for money signalled a massive escapade of those
who, if they had stayed pampering Gopher, would have made the child grow into
an adult. Today there would be two mature and efficient hyper text protocols
standing on equal terms to each other, instead of a monopolistic Web that is
overstuffed with commercialism, and an alive and efficient, but also marginal
and underground Gopher, whose existence is only known to a select few. Mister
Cameron Kaiser intelligently makes the case for Gopher, stating clearly that
Gopher is not just a reminiscence for the nostalgic of times past, but that
Gopher can perfectly co-exist with the Web even today. Wishful thinking, sadly.
Well aware he is that Gopher has about equal probabilities of co-existing in
fair relation with the commercialised Web, as the probabilities that tomorrow
every motorcar be voluntarily abandoned by its owner, driver, and passengers,
who will start using horse, bicycle, or public transport, in order to stop
polluting the natural environment and to diminish the number of accidents.
The all too human tendency for predominant fashion and gregarious socialising
will continue using the Web, or television, or mass consumption, as the human
desire for 'comfort' will keep burning petrol in millions of motorcars or
other motor vehicles world-wide. Only firm restrictions enacted at the level
of international organisations and of governments, can counter-march the
destruction of this planet. Dreams will not counter-march the chaos. Surely
the final catastrophe will, but we shall better be dead than survive it.
1993: Gopher begins to go underground
In 1993 Marc Andreessen and Eric Bina of the National Center for Supercomputer
Applications at Urbana, Illinois (part of the National Science Foundation)
released Mosaic, the most flexible and advanced Internet user agent so far
created. It was based on graphic interface (rather than command line, sadly),
with different ports installable in X-11 Window for Unics systems, Apple
Macintosh, or Microsoft Windows 3.x (based on DOS, but not for DOS as a
stand-alone system). Mosaic featured many resources, including search, cache
and book-mark, and supported various protocols such as FTP, NNTP Usenet, WAIS,
HTTP-HTML WWW, and Gopher. It provoked a craze not inferior to the craze that
Gopher had provoked two years earlier, with thousands of Mosaic copies being
downloaded every day. For millions of people, Internet had until then been a
kind of mysterious limbo that few durst traverse, because they felt awe at
the technical 'difficulties' of the cryptic command line of dedicated clients,
or because they had not the resources of hardware, software, or reliable
connections. But with Mosaic, even the most stupid could now access Internet.
And well known it is, that stupids are the most common rabble under the Sun.
So, the stupids won the day. Gopher relies on external indexing through menus
separated from content, putting the work load for maintenance of resources on
hosting administrators rather than on document authors. The Web is exactly the
opposite, it relies on internal linking by in-lined anchors, with content and
hyper links together in the same page, thus authors bear the work of correct
document structure and updated links, while hosting administrators can attend
to other tasks of server maintenance. At first glance it might seem that
relying on thousands (and soon millions) of document authors is an error, in
that most of them would never be sufficiently technical for creating correct
documents. That is true, but it is compensated by browsers that 'forgive'
errors in the HTML code, by Web creation software of the type 'What You See
Is What You Get', by on-line or off-line HTML validators, and by plenty of
other tools. Little of that has ever been available for Gopher. A Web Master
is helped by innumerable aids of all sorts, while a Gopher Master is very
much on his own. Unavoidably, little by little Gopher became relegated to a
minority of sophisticated Computing enthusiasts, the ones with deep know-how.
Shortly after the appearance of Mosaic in 1993, plenty of ignorants believed
(and they believe until today) in the myth of considering Gopher older than
the Web, while the opposite is really the case. The Web appeared in 1990 at
CERN, and Gopher in 1991 at the University of Minnesota. The cause of the
error is that Gopher became popular immediately, while the Web only became
popular in 1993, with Mosaic. Before that year, the Web was largely confined
to a minority of knowledgeable souls. After that year, the reverse was true,
the knowledgeable souls remained in Gopher while the mass of donkeys flooded
to the Web. Intrinsically, it cannot in good faith be affirmed that Gopher be
better than the Web, or the Web better than Gopher. They are simply different,
with better or worse aspects largely depending on personal notions of what
constitutes 'better' or 'worse', in regard to Internet protocols and to their
ancillary components or their supporting tools. What it can be said is that
the Web played very well the game of presenting its cause as a world-wide,
cutting-edge technique for global communication, while Gopher presented its
as an improvement over FTP, a local information system, a wide area network
just a little wider than the others, a cute 'go pher this and go pher that'.
The result of that provincialism soon began to tell in the mind of the public.
1995: Gopher permanently moves underground
In mid 1995 the Web had surpassed Gopher in Internet traffic. Shortly later,
nearly all commercial corporations (who are always after grabbing the damned
buck) began dismantling their Gopher servers and moving to the Web. Rather
hesitatingly at first, more decidedly later, most official agencies gradually
did that move as well, with plenty of libraries and of academic institutions
following suit. Even the Fatherland of Gopher, the now infamous University of
Minnesota, surrendered to the new trend and decided to leave Gopher an orphan.
Charitable souls adopted the orphan. We take care of him until today, because
we are convinced that he offers a valuable choice, a workable alternative to
the Web. Amongst the many failures of the University, one was the failure to
make the improvement of Gopher an open communications standard and an open
source software project. In 1994 CERN had begun the construction of a really
huge project, the Hadron Collider Accelerator, for the study of sub-nuclear
particles. That milestone project absorbed all the energies of CERN, to the
point that the World Wide Web could not efficiently be attended any longer.
Thus, CERN intelligently chose to waive all claims to the property of the Web,
and released it to the public, this is, to that minority of the public who is
able and willing to work on further development of the Web.
Minnesota saw things differently. The University insisted on being the
protecting 'tutor' of Gopher, on keeping all development efforts inside the
walls of the campus buildings, or at least approved and closely watched by
the zealous University authorities, like a naughty child who must be
safe-guarded from bad acquaintances and evil places. That mindset of 'teacher
of little boys' did not help Gopher. It alienated him from the surrounding
world. He responded by following his instincts, by burrowing deep into the
underground, and staying there. There he is today. The small team at the
University could not possibly match the myriads of volunteers who helped to
develop the Web. For Gopher, it was not permitted. Gopher had to be under the
watching protection of his creators. One result of that short-sighted
approach was that 'unapproved' Gopher variants began to appear. Sure,
variants for the Web also appeared, but the World Wide Web Consortium knew
how to harness them. And besides, the Web already stood on a firmer footing
than the still infant Gopher. Projects that otherwise would have been a
valuable nurturing assistance for Gopher, came and were quickly gone without
the approval of the Minnesotans. An example was Project Panda, an advanced
set of tools for Gopher that would have been the envy of the Web, but which
only lasted from late 1992 to mid 1995, then disappearing in the blue. Gopher
suffered some examples of forking, most of it rather chaotic, and most of it
gone today. Some remaining Gopher resources are listed below.
Resources available for Gopher
Gopher was initially developed by the University of Minnesota in 1991, but
years later that university abandoned its own protocol. Gopher is today
maintained by enthusiasts of the protocol. It is the intelligence and the
dedication of those enthusiasts that keep Gopher an alive and efficient
protocol until today. So be it for many years to come.
Free host servers of Gopher documents
The following hyper link points to one of the few public servers of Gopher
documents, Super Dimension Fortress, which will host the Gopher document for
free during one year or two, then expecting a small one-time contribution.
Besides its excellent Gopher hosting, Super Dimension Fortress also offers a
plethora of other services for sophisticated Computing enthusiasts, such as
Telnet for real time communication with other members, participative games,
Unics technical informations, and programming advice.
Super Dimension Fortress Europe. World Wide Web hyper link
User agents without capability for direct access to Gopher Protocol, but with
capability for Hyper Text Transfer Protocol and Hyper Text Mark-up Language
(or another standard language of the World Wide Web), may use the services of
a Web to Gopher gateway (also known as a Web to Gopher proxy). This can be
done mainly by one of four methods:
-First, by entering the Gopher Uniform Resource Locator into the Web interface
of the gateway (as a command line or address line is entered into a browser).
Of the two Web to Gopher gateways hyper linked below (via World Wide Web),
Meulie offers the first, second and fourth methods, while Floodgap offers the
third and fourth methods. Thus, the two gateways provide access to Veronica-2
search engine, by which means a query delves into a data base containing the
full text of all known Gopher documents world-wide.
Web to Gopher gateways
The two hyper links below are gateways from the World Wide Web to Gopher.
These outstanding services allow computers with access to HTTP-HTML, but
without direct access to Gopher, the ability of accessing indirectly all
known Gopher documents in the World, plus access to Veronica-2 search engine
for finding topics of interest in any Gopher document world-wide.
Besides a Web to Gopher gateway, Floodgap also offers other services for
Gopher Protocol, such as Veronica-2 Gopher search engine, Bucktooth Gopher
server, or Overbite Project for enabling Gopher Protocol in various Internet
user agents that have it disabled by default. Not only Gopher, but also some
other protocols are attended by Floodgap, such as Hyper Telnet. All those
services are maintained by Mister Cameron Kaiser, who always provides useful
advice to persons requesting information on Gopher, Hyper Telnet, or other
important though not generally well known Internet resources.
Floodgap
As it has been explained above, the Meulie Web to Gopher gateway allows two
non-search methods for reaching a Gopher document using a Web client, either
by entering a Gopher Uniform Resource Locator into the Web interface of
Meulie, or by using a combined Web Uniform Resource Locator, where the Gopher
address appears as a subdomain or a directory inside Meulie. Meulie also
permits a query into Veronica-2 search engine, to all known Gopher documents
world-wide. There are about a hundred active Gopher servers as of 2016, which
contain hundreds of thousands of items (pages showing a menu list, pages with
a body of text, images, sounds, software, or other resources).
Meulie
Search engines for FTP and Gopher
Archie (Archiver) was a general search engine for File Transfer Protocol.
Created by Peter Deutsch, released in 1990. Last known public server of
Archie was at a university in Poland, whose Archie service closed in 2013.
Veronica was a general search engine for Gopher Protocol. Created by Steven
Foster and Fred Barrie, University of Nevada at Reno, released on 17th
November 1992. Last known public server of the original Veronica closed in
the early XXI century.
Jughead was a limited search engine for a single server of Gopher Protocol.
Created by Rhett Jonzy Jones, University of Utah, released in 1993. Last
known public server of Jughead closed in the early XXI century.
In the vein of good humour that is often characteristic of Gopher, Veronica
means 'Very Easy Rodent-Oriented Netwide Index to Computerized Archives',
while Jughead means 'Jonzy's Universal Gopher Hierarchy Excavation And
Display' (HEY, DON'T LOOK AT ME LIKE THAT, I DIDN'T INVENT THOSE NAMES !).
Out of a most fortunate 'coincidence', Archie, Veronica and Jughead are also
fictional characters of a North American series of stories in drawings (comic
strips in vignettes), printed on paper. It seems that some Gopher enthusiasts
are fond of the Ninth Art. Peter Deutsch, who developed Archie, insisted that
its name was a short form of 'Archiver', and had nothing to do with the comic
strips. He was supposedly disgusted when Veronica and Jughead appeared. Or
else, he did not appreciate certain manifestations of the Ninth Art.
Archie and Veronica offered keyword search of, respectively, most FTP and
most Gopher documents in the World. A Veronica search produced a menu of
Gopher items, each of which was a direct pointer to a Gopher data source.
Because Veronica was accessed through a Gopher client, it was easy to use,
and it gave access to all types of data supported by the Gopher Protocol. To
search with Veronica, it could be selected from the 'Other Gophers' menu at
the Gopher server of the University of Minnesota, or at a few other Veronica
servers. Alternatively, a Gopher client could be pointed to, for instance:
Name=veronica Type=1 Port=70 Path=1/veronica Host=gopher.scs.unr.edu
gopher://gopher.scs.unr.edu/1/veronica
The above Gopher address of the original Veronica is not operational anymore.
All that functionality is still available in Veronica-2, a general search
engine for Gopher Protocol created by Mister Cameron Kaiser, of Floodgap.
Veronica-2 is operational as of 2016, accessible from Floodgap or from Meulie,
hyper linked above. As of 2016, there is no known search engine for FTP.
Gopher, Veronica, Jughead by electronic mail
There is in the World Wide Web a document (not updated since 2001) that
explains how to access Gopher documents, and how to perform a search for them
using Veronica or Jughead, exclusively from a client of electronic mail (SMTP
and POP3 Protocols), without need of a Telnet, Gopher, or Web client.
This service (which might be called a 'Mail to Gopher gateway') is necessary
for those who lack Telnet, Gopher or Web software, or for those who work from
a country where access to those protocols is difficult or impossible (such as
Cuba or North Korea), but who can use electronic mail.
The technique consists in sending an electronic message to the Mail to Gopher
gateway, specifying the word 'HELP' in the Subject line. The gateway will
reply with a list of commands. It is also possible to send the electronic
message including a Gopher Uniform Resource Identifier in the Subject line,
in which case the gateway will retrieve a copy of that Gopher document (if
being able to reach the indicated Gopher server), and will reply with the
desired Gopher page, or with a message informing that the Gopher server is at
the moment unreachable (or that it may not exist anymore).
Archie (for FTP), Jughead and the original Veronica (for Gopher) were wiped
out of existence many years ago, but Veronica-2 can still be searched via
electronic mail, using the Mail to Gopher gateway. The afore mentioned Web
document explains the correct procedure in detail, but after due thought, CSS
Dixieland has chosen not to include all that information in this page. That
is because ONLY TWO of those gateways were known to exist in 2001, and there
is a high risk of the information becoming obsolete in a short time. Or it is
already obsolete, because it depends on the availability of those gateways.
CSS Dixieland expresses apologies to those persons who might find such a
gateway service useful. The electronic mail addresses are given below, in a
manner that is hopefully understandable to humans, but not understandable to
programmes that automatically collect mail addresses (often for purposes of
spamming). The two Mail to Gopher gateways are located in Japan, they are:
gopher AT ncc DOT go DOT jp
gopher AT dna DOT affrc DOT go DOT jp
The string "AT" must be substituted by the character "@" and the string "DOT"
by the character "." without surrounding spaces and in single horizontal line.
Sorry for the inconvenience, but there is a bunch of niggers from Niggeria
whose favourite sport is sending lots of spam. They use computer programmes
for automatic harvesting of addresses, and the above substitution is one of
the methods for thwarting such programmes and their nasty monkey users.
As it has been said above, the word 'HELP' in the Subject line will make the
gateway reply with a list of commands. For search, the host of Veronica-2 is:
gopher.floodgap.com
Unfortunately, Gopher mail methods for reading Usenet news groups (NNTP),
were already non-existent as of the year 2001. Usenet can be accessed via a
dedicated client or via Web, but not via electronic mail anymore, except for
a few Usenet news groups that maintain posting lists for their subscribers.
Free-DOS operating system
The following hyper link points to Free-DOS, the most advanced DOS operating
system ever created, whose distribution includes two user agents (clients)
with full Gopher support: Gopherus and DOS Lynx, and another with partial
Gopher support: Arachne. Their hyper links are listed in the section
'Software, Net' inside the Web document of Free-DOS, beside many other
resources of interest to DOS enthusiasts. DOS Lynx is text-only, Arachne is
graphic. Gopherus was first released in 2013, being the work of Mister
Mateusz Viste, a long-time member of the Free-DOS project who also created
the Motsognir Gopher server. Intelligent and dedicated Gentlemen such as
Mister Cameron Kaiser of Floodgap and Bucktooth, Mister Mateusz Viste of
Gopherus and Motsognir, or the Gentlemen of Meulie, Super Dimension Fortress,
Free-DOS, or BSD, keep Gopher an alive and efficient Internet protocol until
today. So be it for many years to come.
Free-DOS
Gopherus is the best user agent for accessing Gopher documents from DOS. Its
colours can be fully customised, if using polychrome screen and video card.
They can even resemble the classic phosphor cathodic ray tube screens of the
1980's. Gopherus follows the official Gopher Standards RFC 1436 (The Internet
Gopher Protocol) and RFC 4266 (The Gopher URI Scheme). Gopherus needs a
packet driver loaded in Random Access Memory. There are ports of Gopherus for
DOS, BSD, Linux, and Microsoft Windows Operating Systems.
As for Gopher server software, three stand out in 2016 as perhaps the most
relevant for those Gopher Masters who operate their own computers AND WHO ARE
LUCKY TO HAVE INTERNET CONNECTION IN THEM: Bucktooth, Py Gopherd, Motsognir.
Bucktooth is the work of Mister Cameron Kaiser, mentioned above in relation
to Floodgap gateway and to Hyper Telnet Protocol, besides other outstanding
works. Motsognir is the work of Mister Mateusz Viste, also mentioned above in
reference to Free-DOS operating system and to Gopherus, Gopher client. In a
communication sent to CSS Dixieland in December 2015, Mister Viste informed
of important modifications introduced into Gopherus, making it able to work
in Unics systems (such as BSD or Minix) directly from the command line,
without X-11 or any other graphic interface. Gopherus already worked directly
from the command line in DOS systems (such as Free-DOS), without the Graphic
Enviroment Manager (GEM, original from DR-DOS) or any other graphic interface.
Mister Viste also pointed out that Motsognir can be used as a Gopher server
even in an isolated computer, without network connection of any sort, thus it
can be handy for creation and maintenance of Gopher documents, which may be
uploaded to a remote Gopher host by a second, shared computer with Internet.
The problem then is 'only' to find a workable Gopher hosting, because Super
Dimension Fortress predominantly uses Telnet or R-login.
Motsognir is a server of Gopher documents stored in on-line or off-line
hosts. It can be used in an isolated computer, local network, wide network,
UUCP, or Internet, but the host needs Unics compatible operating system (Bell
Seventh Edition - Official Standard IEEE POSIX 1003), it cannot be used in
DOS. Motsognir has no external dependencies, it allows CGI or PHP for dynamic
content, redefinition of Gopher item types, secure mode for serving only
resources with 'world readable' permission, runnable as not Super-user after
opening the listening port (often port 70, but it can be changed), it
is trappable in a chroot jail, supports ASCII or UTF-8 characters, Internet
Protocol versions 4 or 6, Gopher maps, CAPS.TXT with extra text, virtual
hosting for different content in different domains (not a standard in Gopher
Protocol), and has many other configuration options. Highly recommended !!!
Motsognir. Gopher hyper link
Motsognir. World Wide Web hyper link
The image of the dwarf featured in Motsognir is based on the original work of
Lorenz Froelich (1820-1908).
Gopherus Gopher client and Motsognir Gopher server have been copyrighted by
Mister Mateusz Viste and are legally protected by GNU General Public Licence
version 3 of 29th June 2007, which can be read or copied from the Uniform
Resource Locator of the Free Software Foundation:
Free Software Foundation
The Free Software Foundation also announces as its address in the World Wide
Web the Uniform Resource Locator http://www.gnu.org/
Open Source is another outstanding institution in the Open Source Movement.
Its Uniform Resource Locator is:
Open Source
Combining Gopherus and Motsognir
Gopherus is a Gopher client and Motsognir is a Gopher server, therefore they
can be combined for complementary work. Mister Mateusz Viste, author of the
two softwares, has provided valuable information to CSS Dixieland on how this
combination can be accomplished. They must be installed in a computer running
an operating system compliant with Bell Seventh Edition - Official Standard
IEEE POSIX 1003, such as Minix, BSD, or another Unics system. Motsognir has
no external dependencies, and Gopherus can be executed using X-11 Window
system for graphic interface, or using the Curses library for command line.
With Curses and the two softwares installed, access to a Gopher document
located inside the local computer can be done by entering the command line:
After the Gopher Protocol prefix (or another communications protocol prefix),
that number '127.0.0.1' is an address of the test loop-back interface, one
of the about 16 million possible addresses for a local host (127.0.0.0/8).
The packet of data thus sent is not routed to a network, but remains inside
the local computer, which may be totally isolated, without network connection
of any sort. By using that technique, Motsognir will serve the Gopher document
to Gopherus, while in reality all of them are located only inside the local
computer (Motsognir, Gopherus, and the Gopher document). This is necessary,
because the Gopher Protocol does not resolve relative paths inside a local
computer (unlike for instance the World Wide Web, which supports relative
paths). Gopher needs an absolute path address, a Uniform Resource Identifier.
In the case above, that address is in the local host (the local computer).
The topmost current authority for the Gopher Protocol, Mister Mateusz Viste,
warns that such a concept of 'test loop-back interface' in a local host, must
not be confused with another concept known as a 'Martian address'. He explains
that a Martian address is a concept associated with Internet Protocol security
in some operating systems. In such systems, whenever the kernel happen to
receive an Internet Protocol packet on one of its interfaces, it performs a
simple 'reverse-lookup' verification by asking to itself the question:
"Through which interface should I route a packet, if the destination were X ?"
...where 'X' is the source of the packet that has been received. If the result
of such an internal look-up happen to point to the very same interface that
the interface through which the packet has come, then all is fine. Otherwise,
the packet is marked with a source address that is 'beyond this World' (hence
the nickname 'Martian'). The packet is then tagged as a 'Martian packet'. It
might be potentially dangerous, therefore it might be dropped by the receiving
computer, depending on the security configuration of the operating system, or
on the paranoia of the system administrator. In this regard, the address of a
local host is not a Martian address, since it is closely attached to a test
loop-back interface that is well-defined (typically lo0), which holds the
127/8 address plan:
Everything can be done by command line, with the Curses library installed in
the computer. Thus, funny graphic interfaces are not needed at all. The good
old command line can perfectly be used, but it is necessary to install the
Curses library. This library is publicly available for free, or it may be
already included by default in the distribution of Minix, BSD, or another
Unics system. The combination of Motsognir and Gopherus can be used inside a
single computer, or local area network, or wide area network, or UUCP, or
Internet itself, but all the participating computers must possess both
softwares and have them correctly configured. Instructions are given in the
documentations of Gopherus and of Motsognir, which in part are presented in
Adobe Portable Document Format (some converter to ASCII plain text may be
necessary). For asking any questions AFTER HAVING READ ALL THE DOCUMENTATION,
Mister Viste is reachable from the Motsognir hyper links given above (via
Gopher or via Web), or by electronic post. He is a very helpful Gopher expert
who will gladly explain all possible solutions. He has been of enormous
assistance to CSS Dixieland, by the high quality of his technical reports.
Gopher logs and Web logs
Gopher logs, also known as 'glogs' or 'phlogs', and Web logs, also called
'blogs', are simplified documents of easy creation by non-technical authors.
Gopher logs are transmitted by the Gopher Protocol, and Web logs by the Hyper
Text Transfer Protocol, these ones then rendered by the Hyper Text Mark-up
Language or another standard language of the World Wide Web. The idea of logs
began in December 1997, with the name 'log' reminding of an entry written
into a navigation diary. They are a resource often used by those who wish to
publish information in Gopher or in the World Wide Web, but who for whatever
reason do not want to bother about learning the technical details necessary
for programming their own documents. Gopher or Web logging organisations
offer to them a certain variety of ready made templates and of authoring
tools, and from those templates and tools they choose one of their liking,
they define some parameters such as colour or text size if permitted by the
tools and by the protocol to be used, and begin entering their own text.
A few loggers publish really interesting information that is a pleasure to
read, but by far the vast majority of 'gloggers-phloggers' or of 'bloggers'
write only garbage without any interest at all, except maybe for themselves.
They just show family pictures, personal hobbies, holidays, daily life, and
other idioticies that are typical of stupid average people. They are just the
usual gang of idiots, in the words of Mister Cameron Kaiser. Nothing special,
nothing lasting much beyond their empty lives. A few Gopher or Web loggers
offer something better, but they are exceptions. In general we may say that a
person who be serious about publishing in Gopher, in the World Wide Web, or
in another Internet protocol, will also be serious about learning in detail
the technicalities needed for it. The same with, for instance, someone being
serious about Photography, who will enthusiastically study and experiment
with photographic technique instead of limiting himself to ridiculous 'point
and shoot' fool-proof amateur cameras.
Therefore, CSS Dixieland does not recommend publishing in logs. Instead of
logs, those of our readers who have something important to say to the World,
are encouraged to take the trouble of learning how to say it in a professional
manner. The message will show much more respectability as a well composed and
well constructed personal Gopher or HTML document than as a Gopher or Web log,
and the satisfaction felt by its author-programmer will also be greater than
if using a mass produced template. Notwithstanding that advice, those readers
who may at present feel insecure about their technical preparation for the
daunting task of creating their own Gopher or HTML documents, will do well in
joining some community of enthusiasts who will help them in their quest.
One of those communities is Super Dimension Fortress, where members can
create their own Gopher log, or their own full Gopher document if they feel
prepared for it, under the tutoring and guidance of more experienced authors
who can often be contacted via electronic post, Telnet, Gopher, or World Wide
Web. It is even possible to include a book of visitors in the Gopher document,
using a script of Common Gateway Interface for dynamic content, or to link to
images, sounds, or other resources. Super Dimension Fortress is hyper linked
via Gopher or via Web in the section of Resources available for Gopher,
sub-section of Free host servers of Gopher documents, above in this page:
Resources available for Gopher (Free host servers of Gopher documents)
And with all that plethora of information, CSS Dixieland finishes this page
devoted to the unforgettable Gopher Protocol. We have to recognise, sadly,
that Gopher is as of 2016 'playing second fiddle to the Web' (in the words of
Mister Cameron Kaiser). Yet, Gopher is an efficient protocol and does not
deserve to be thrown into the attic of old memories. For one thing, because
Gopher is fortunately free of the repugnant commercialism into which a vast
part of the Web has fallen, with so many 'features' of dubious usefulness,
supposedly for 'enhancing user experience' by means of uncountable pop-up
advertisements, either legitimate or surreptitiously introduced by means of
'ad-ware' malicious codes, into the computers of many ignorant users who do
not know how to protect themselves. Marketroid influence is rife in the Web.
For another thing, because Gopher is a simple and uniform standard, unlikely
to change versions (Gopher Plus or Hyper Gopher never had a wide acceptance),
which means that it is comparatively easy from the technical point of view
to create and maintain a Gopher document. The World Wide Web, on the other
hand, boasts of a huge collection of standards of all sorts: HTML up to
version 4.01 (based on SGML), HTML version 5 (based more on XML), Cascade
Style Sheets or other style sheet languages, Java Applets, Java Script,
I-HTML or other proprietary languages, exaggerated dynamic content, frames,
exclusive proprietary extensions in many Web user agents (such as Netscape
Navigator, Microsoft Internet Explorer, or even X-Chaos Arachne), ActiveX
Controls, Microsoft Active Server Pages, Java of Sun Systems, many other
implementations for rendering visual effects... it all goes on ad infinitum.
The World Wide Web Consortium exerts tremendous efforts for preventing the
appearance of a jungle of incompatible proprietary 'standards', or of a new
edition of the 'Browsers War' (which happened mainly between Netscape and
Microsoft in the late 1990's). Certainly, in the hands of a knowledgeable Web
Master, those technical resources can be used intelligently. Unfortunately,
most of those who own a Web document are not so knowledgeable. A few big
corporations, or academic or government institutions, have their own experts,
technicians and artists who create and maintain a presentable and efficient
Web document. Medium corporations or private institutions tend to hire the
services of professionals for Web creation, or for occasional maintenance.
Small corporations, or individuals, too often use Web creation software of
the type 'What You See Is What You Get', and rarely think of maintenance.
The result of using that 'WYSIWYG' software is, invariably, spaghetti code,
which makes maintenance difficult and modification almost impossible. That is
the consequence of having turned into mass consumption a number of complex
techniques that only a relatively small number of experts are capable and
willing to understand. For the rank and file of stupid consumers, it all
sounds Greek. Correct construction of Web documents is too cryptic a task for
them. They prefer watching television, it is much easier than perusing a
manual of technical specifications and learning all its 'obscure' details.
Gopher, fortunately, is not so 'cryptic' as for example HTML 5 is. Again,
this is not to detract HTML 5 at all, because a careful study of the HTML 5
specification of 2014 shows a remarkable work done by the World Wide Web
Consortium and by the collaborators who prepared the HTML 5 draft. This is
only to say that Gopher is the most simple and uniform hyper text protocol,
free of commercialism. And that in all probability, so it will remain.
Robot or human visitors to CSS Dixieland are recorded in raw access log. This
is a passive register purely for statistical purposes, no cookies are stored
in the client computer.
Go to top of this page
Go to page with index, history, exchange policy, contact CSS Dixieland:
Start
Hosted by Neocities:
https://konqueror.org/
Walkyrie who takes our dead heroes to Walhalla in Asgard.
Wagner Frost Illustration
History: past and present of Gopher
Resources available for Gopher
Gopher logs and Web logs
{CSA, CSN} !confederate!military!naval!cssdixieland
The Arpanet and NSFnet that gradually became Internet had little traffic till
the early 1990's, when compared to bulletin boards or proprietary networks.
The mass of consumer idiots became aware of Internet only about 1992 or later.
65 536 characters, millions of combinations, covering almost every language past or present
http://www.unicode.org/
"The Internet Gopher was seen for a long time as a preferable information
system, avoiding the complexities of HTML, but rumours of the technology
being licensable by the University of Minnesota provoked a general
re-evaluation".
Those words are valued their weight in gold.
One of the few free hosts for Gopher documents (for free during one year or two)
http://sdfeu.org/
-Second, by using a combined Web Uniform Resource Locator, where the Gopher
address appears as a subdomain or a directory inside the gateway.
-Third, by activating a Gopher hyper link already listed in the gateway (as
it is done in a portal).
-Fourth, by performing a search in Veronica-2, the search engine for Gopher,
which can be done from the interface of the gateway.
Gateway from World Wide Web to Gopher, Veronica search engine, Hyper Telnet
http://gopher.floodgap.com/gopher/
Gateway from World Wide Web to Gopher, Veronica search engine
http://gopherproxy.meulie.net/
Operating system that includes three Gopher clients (Gopherus, DOS Lynx,
Arachne)
http://www.freedos.org/
Very configurable server of Gopher documents, for Unics systems
gopher://gopher.viste.fr/1/projects/motsognir/
Very configurable server of Gopher documents, for Unics systems
http://motsognir.sourceforge.net/
Promotion and legal defence of free and open source software and other kinds
of works
http://www.fsf.org/
Promotion and legal defence of free and open source software, OSI approved
http://www.opensource.org/
gopher://127.0.0.1
# ifconfig lo
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
https://www.neocities.org/