Talk:Gopher (protocol)

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Computing / Software (Rated C-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
Taskforce icon
This article is supported by WikiProject Software.
WikiProject Internet (Rated C-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Internet, a collaborative effort to improve the coverage of the internet on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.

gopher protocol[edit]

Would it be possible to add in the browsers support section that Konqueror ( ) supports gopher:// if kio_gopher ( ) is installed and that it shows the textual information mozilla does not?

Just write something along the lines of "Konqueror has full Gopher support with the kio_gopher plugin installed" (if it is a plugin) and add it. And create an account so you can take credit! magetoo 02:54, 28 Apr 2005 (UTC)

This may be stolen from another site, or that site may have stolen this.[edit]

Worldvillage encyclopedia has the same Gopher article word for word, including the suggested addition.

That site has copied this one - it says "from wikipedia" :) --CatS 15:33, 21 August 2005 (UTC)
And they're completely within their rights to do so - see Wikipedia:Mirrors and forks and Wikipedia:Copyrights. - IMSoP 18:04, 21 August 2005 (UTC)


Gopher "was released in 1991", but "when the World Wide Web was first introduced in 1991, Gopher was well-established and quite popular". Maybe someone who was there could add a bit more details here; did Gopher usage just explode during the first few months of 1991, or what happened? 14:04, 22 September 2005 (UTC)

I added some more detail to the that should help explain a bit about why gopher became popular quickly. In a nutshell, CIOs at universities all decided they needed to have some sort of campus wide information system. Gopher was a quick easy way to do something, and had the added benefit that you leveraged the work at all the other universities. If one university put up a searchable collection of recipes (which the University of Minensota did), every other site could link to it an magically have recipes as part of their CWIS. Add that to a way of organizing online phonebooks (CSO nameservers were the weapon of choice for this in the early 1990s) and there wasa recipe for very rapid growth. Another Cat 21:51, 29 December 2005 (UTC)

Same question here. --Walter 00:25, 11 November 2005 (UTC)
The Web was first announced in 1991 by Tim Berners-Lee but it wasn't truly introduced until 1993, with the unveiling of Mosaic. Gopher was first announced in 1991 too, probably a little earlier than the web was, and free Gopher server and client software was released by the UMN shortly after. Gopher was usable technology years before the Web was, even though the two were both announced the same year.

What is it?[edit]

This article completely fails to explain what Gopher is... It breifly glosses over the definition and then talks about its history. I want to know how it works, what exactly it does etc... Maybe a bit more on this?

A Gopher consists of a series of hierarchical menus. A University's Gopher site, for instance, might have a main menu with 5 to 15 or so subdivisions, looking like this:

ISU-- the Imaginary State University

1. Academic Departments 2. Athletics 3. History of the Institution 4. Libraries 5. Student Life 6. Visiting Campus

When you select (say) 2. Athletics you bring up another menu which might read as follows:


1. Baseball 2. Basketball-Mens 3. Basketball-Women's 4. Football 5. Lacrosse

And then there might be further links down to specific aspects of the lacrosse team.

Note that the choice of menu items and titles is arbitrary. There could just as easily have been four items on that menu, with men's and women's basketball linked together as one item. Once you get to a text document which is out there in Gopherspace, anybody else can link to that document from their OWN set of hierarchical menus, arranged in some completely different way, with their own titles. The ISU lacrosse gopher page might be one item in somebody else's lacrosse gopher. If there were a whole bundle of ISU lacrosse pages, the whole bundle could be linked to other gophers as a unit.

Thus, to actually get to the text you wanted, you needed to click on several links, and you had to read through a menu each time before you chose another link. That worked more rapidly than people might think today, and in the heyday of gopher, there was a lot of time spent choosing and organizing links.


"released in 1991" "By the late 1990s, Gopher had all but disappeared." Could someone imporve the timeline here?

A quote:

"The Internet Gopher (the official name) has nothing to do with that pouchy rodent, except the name. Gopher is an Internet protocol, a distributed document search and retrieval system developed by the University of Minnesota. The official definition explains this protocol quite clearly: "The Internet Gopher uses a simple client/server protocol that can be used to publish and search for information held on a distributed network of hosts. Gopher clients have a seamless view of the information in the Gopher world even though the information is distributed over many different hosts. Clients can navigate through a hierarchy of directories and documents menus! or ask an index server to return a list of all documents that contain one or more words. Because the index server does full-text searches, every word in every document is a keyword."

Citation: Go for all the information around the world with the Internet's Gopher Milewski, Darek. InfoWorld. San Mateo: Jul 12, 1993.Vol.15, Iss. 28; pg. 41, 1 pgs

Article split[edit]

The article should be split into two. Gopher protocol should be about the protocol itself and Internet Gopher should be more general. Benn Newman 19:55, 29 December 2005 (UTC)

The article needs to be cleaned up. If someone feels that the best way to clean it is to split it first, then I'd say "go for it!" I miss the days of gopher, but I don't really have the knowledge to do it myself. 18:20, 31 December 2005 (UTC) Whoops, forgot to sign in before. Matt Deres 18:22, 31 December 2005 (UTC)

  • I oppose the split, as long as the section on the protocol itself is as short as it is. The article needs improvement, but I don't see how the split would improve anything right now. Kusma (討論) 10:52, 16 January 2006 (UTC)
I don't know if it's worth splitting completely, but I have had a go of at least collecting together the information that seemed to belong together - everything historical is now under "History", and so on. It all still needs work, but this goes some way to eliminating the worst arbitrarynesses I think... - IMSoP 15:58, 16 January 2006 (UTC)
Kusma: Part of the whole point of Gopher is that it is simple. Of course it is short. :) Benn Newman 22:07, 16 January 2006 (UTC)
  • The page looks a lot better, IMSoP. I came off as sounding like I'm in favour of a split above, but I'd vote oppose at this point. Keep it together until the article is just too long. Matt Deres 17:33, 18 January 2006 (UTC)

I've just taken a piece of earlier discussion which I posted several months ago, before signing up as a user, edited it, and put it in as a new section titled "What is an Internet Gopher?" That discussion sat here for a couple of months and nobody added to it or commented on it. I suspect that the provision of a secion like this obviates any need for a split, so I vote oppose on this.Chris MullinAs you could probably tell, I'm just starting here, had not yet read through Help:Editing :-(CGMullin 17:28, 23 February 2006 (UTC)

Summarisation and reword[edit]

I've summarised the example and reworded some parts to make it more understandable for non-technical background readers. (Hands up, those of you who know what CR LF is... without referring to the ASCII table. :P) — | Talk 18:58, 23 February 2006 (UTC)

Some example still needed, IMHO[edit]

.. so maybe mine WAS too verbose! I admit I'm verbose. But if somebody who has vaguely heard about Gopher wants to know what it was-- the present version does not show them that. See the comment somebody (not me) made last year above under What is it? I quote the revision:

"A Gopher system consists of a series of hierarchical menus. The choice of menu items and titles is set by the administrator of the server.

Similar to a file on a Web server, a file on a Gopher server can be linked to as a menu item from any other Gopher server."

If I did not already know what a Gopher is, this explanation would be opaque. The only way I know to clearly show how this differs from hypertext is to give an example. A real Gopher typically had several dozen hierarchcally-arranged topics and subtopics, on four or five levels.

Especially as time passes, the chief interest in an article on Gopher is likely to be historical. The article now, in conjuction with References and footnotes, explains how to set up a Gopher, but not what a Gopher user experience was like. I voted against a split because this is always going to be a short wiki, and readers are apt to be interested in both the technology and the interface.

I'm quite willing to hear suggestions on how to shorten my earlier draft!CGMullin 21:17, 23 February 2006 (UTC)

FWIW, I think I'm a fairly appropriate person to be contributing this wiki section, since from the beginning of the University of Montana's Gopher, up til the end several years later, I was one of the people on the local "Gopher-Tamers" committee who actually organized and selected the links. CGMullin 21:22, 23 February 2006 (UTC)

OK, I understand your point... lots of people nowadays have never seen a Gopher menu. :(
However, I think a real example is better than a fictional one, so I will be adding one later this afternoon (UTC time), since I'm stuck with real-life work now. — | Talk 01:52, 24 February 2006 (UTC)
What is this historical-aspect of Gopher you speak of? ;) Benn Newman 23:07, 23 February 2006 (UTC)

I have made some screenshots to replace the example. I believe this will be more fascinating to those who have never used a dedicated Gopher client - a picture speaks a thousand words. And we're really "showing them" Gopher in operation. ;-) — | Talk 17:55, 24 February 2006 (UTC)

For a Gopher `paper' I am writing, I took some screen shots of various Macintosh clients -- I could add them (but probably not for a few days). Benn Newman 22:08, 24 February 2006 (UTC)

I'd suggest that since Gopher was most prominent in the days of 80x25 ASCII screens, if only one more illustration is needed, it would be MUCH more appropriate to include an ACSII screen rather than both a Windows client and a MAC client.]] In the heyday of Gopher there was, as yet, no graphical web browser. Gopher was so successful precisely because it was an alternative to hypertext, and if you look at it on either a Mac or a Windows machine, you miss that. Kimchi, can you (or anyone?) still replicate a monochrome screen, 80x25? Amber for preference!

Alternatively, one more Windows screen, AND an ACSSI* screen, with a lable to show that this is how we did things in the distant past. With the same text in both, or all three interfaces, of course. Probably a second-level menu is only needed for one example, so we are talking about either one or two more illustrations. CGMullin 06:55, 25 February 2006 (UTC) CGMullin 06:55, 25 February 2006 (UTC)

Gopher *is* hypertext. I could upload a screen shot of the UMN UNIX gopher client from a terminal emulator. We could also point people to [1] or [2]. Benn Newman 16:34, 26 February 2006 (UTC)

I'd sincerely hope that Benn (or somebody) does upload the screen shot. If I felt more capable of my abilities I'd try that one myself. That seems more useful than another link, since (IMHO!) only one additional ASCII example is needed, but I think that example *is* needed. I'd have no problem with adding the links also, of course. CGMullin 23:40, 27 February 2006 (UTC)

I added a screenshot of the UMN curses client. I wasn't sure exactly what colour amber is supposed to be so I used green instead. :) Benn Newman 03:44, 28 February 2006 (UTC)

Green is fine, Benn! I guess you were not around in the days when monochrome monitors had changed from black to green to amber (a sort of brownish-yellow), on the basis that amber produced even less eyestrain than green. The large-scale adoption of Windows and the increasing affordabilty of VGA monitors killed off the amber monitor. But it WOULD be appropriate for a Gopher menu from one of the original Gopher clients, because this, like Gopher itself, was what most people were using around 1992. But I repeat-- green is fine. Probably most other people who got to the web after Mosaic arrive alos have no clue what an amber monitor was. CGMullin 15:29, 28 February 2006 (UTC)

I am a fairly recent convert to Gopher, about a year and a half or so. After discovering rements of older technologies -- I am using the word technologies somewhat losely -- I am amazed at why people think the current way of doing things is any better. Benn Newman 22:12, 28 February 2006 (UTC)

The current way of doing things is NOT better if you are trying to organize something like a website for one University-- except that it's harder to meaningfully integrate all the pretty pictures into all the Gopher menus. Gopher was especially great for people with visual problems. Much easier for them to navigate than even a typical compliant webpage. And the hierarch imposed by people who were actually desinging the Gopher made it much less likely that people would get lost in their searches, if the design work was well done. OTOH, Gopher did not permit a convenient search over multiple pages. And even if you put thumbnail images in all over the menu pages, it does not LOOK as sexy as a good webpage. That was what really got people going to the WWW after Mosaic appeared-- the looks. The fact that you could finally *see* something on the Internet besides endless 80x25 ASCII text. By the time people started using Mosaic, in 1994 and 1995 for most folks, everybody was used to VGA graphics in games, and even used to being able to show different fonts in Windows applications. But most people had never had a chance to see any graphics on the Internet prior to Mosaic. CGMullin 22:49, 28 February 2006 (UTC)

Is Gopher really hypertext?[edit]

Commenting on Benn's statement of 16:34, 26 February 2006 -- Well, I guess so, but it's ALSO a menuing system, rather than the embedded links which are almost universally the only form in which hypertext is encountered today. That's why I'm pushing for examples that include screen shots-- because Gopher just does not work the way that today's typical software does. Many people looking at this article who want to understand Gopher won't be clear on that unless they see an example. Those people are less likely to actually leave the site to visit a real Gopher as their initial step toward learning more than they are to click on an image. CGMullin 23:40, 27 February 2006 (UTC)

I note, BTW, that currently implies that Gopher is NOT hypertext-- a fairly through rewrite of a paragraph or two would be needed to fix that I think. Since I'm new here, I will give others working here a heads-up on that, and see how you fix it CGMullin 15:33, 28 February 2006 (UTC)

Yeah, I noticed that before. I just rethought it in my head to correct it. :) Benn Newman 22:20, 28 February 2006 (UTC)

I've corrected History of the Internet to show that Gopher really is hypertext. CGMullin 00:22, 22 March 2006 (UTC)

Item type characters[edit]

With the 'complete list' of item type characters, new ones (or ones that were never used) just come out of the blue. I am reverting it back to the list of most used. Benn Newman 13:22, 15 August 2006 (UTC)

any proxy ?[edit]


I think it might be useful to put a list of proxy who still handle gopher, in the external links, what do you think ? I know does, but only for users of this ISP (free.) —Preceding unsigned comment added by (talkcontribs) 2006-11-10T14:21:31

I am not sure exactly what you mean. Feel free to add any links that are allowed under the external links policy. --Benn Newman 22:31, 10 November 2006 (UTC)

Insofar as information management is concerned, the shift in user base from Gopher to the web can be seen simply as a preference progression from text-based to graphical interfaces.

NO NO NO! The download speeds were so slow at the time that GUI's that were not entirely client-sided were unbearably slow. As a computer graphics student in the mid 90's, I use to watch as illustrations and pixelated photos downloaded VERY painfully slowly, one pixel line at a time!! Gopher could just have easily caught up with graphics, but I guess no-one wanted to pay the fees. Do not underestimate the power of free. Cuvtixo 18:32, 19 August 2007 (UTC)

Remaining sites being run by individual enthusiasts[edit]

I have removed the line "With the vast popularity of the World Wide Web, Gopher is all but disused at present, with remaining sites being run by individual enthusiasts" from the introduction as there are still quite a few gopher servers around that do not fit this description, they are mainly maintained by educational institutions. --Thefrood (talk) 03:26, 12 September 2008 (UTC)

- there are still a couple left that are run by universities and the like but a lot are run by enthusiasts. Towel401 (talk) 13:17, 12 September 2008 (UTC)

-- ...but that is not what the line said and quick browse through the known server lists at Floodgap shows the number of servers linked to education organisations to be quite a bit more than just a couple and then there are servers linked to non educational organisations such as the Network Time Protocol project. --Thefrood (talk) 19:20, 12 September 2008 (UTC)

Gopher and Chrome[edit]

Have not tested it, but would like to see the browser support table updated with details from Chrome's support. (talk) 21:02, 11 March 2009 (UTC)

What should updated if there is nothing to update? Webkit doesn't support gopher natively and google seems to be uninterested to implement it in. Chrome does not support gopher (at the moment). mabdul 06:53, 11 October 2010 (UTC)

Broken connection between images[edit]

In the original 2006 image, the 'fun and games' link was visible, such that the 2nd image (what can be found after following that link) makes sense. However, in the 2008 image, 'fun and games' is no longer visible, so the connection no longer makes sense Fatphil (talk) 13:06, 30 May 2011 (UTC)

Table of browsers[edit]

Browsers that do not support Gopher should be removed. Including them in a table of browsers makes sense for an article about browsers. They don't belong in an article about Gopher. When time permits I will remove them if they are still shown. (talk) 21:39, 20 July 2011 (UTC)


I first removed Konqueror but then reverted myself. This is why; The website of the Kio-plugin - - The download section; all links are broken. It is not available anymore at KDE. The developer websites of it; latest release is of 2008. Download links are broken. It can still be found and downloaded here; . But I could not install it on Ubuntu 10.04 because dependencies that could not be resolved. But just now I noticed it is still in my repository of Ubuntu - and it installs and works. So it may be a dead project that plugin but is not yet extinct. At least in Ubuntu 10.04 LTS, what is an older, version of Ubuntu. --Walter (talk) 21:00, 25 September 2011 (UTC)

lftp and gopher[edit]

Cannot find any mentions of gopher protocol support in lftp, are there any?

$ LC_ALL=C lftp gopher://
lftp: gopher - not supported protocol

--Ygrex (talk) 02:13, 12 August 2012 (UTC)

Using Gopher[edit]

I will try to answer requests above to share what it was like to use Gopher, and to better show Gopher's place in the rush to the World Wide Web and search engines we know today.

There are complaints ("If I did not already know what a Gopher is, this explanation would be opaque" and "This article completely fails to explain what Gopher is...") and, both as a computing and as an internet article, it has received a "C" grade. The TALK contributors who made those criticisms also wrote good text to improve the article, which inspired what I wrote. My main reference is the well-known article which taught me Gopher, widely reprinted at the time (my copy came from the NIH, Bethesda computer center) and fortunately still available on the Web.

INSERTION & HEADINGS: Wikipedia articles are never done. To make additional editing as easy as possible, I will delete nothing, only add the text which is also given below. The opening sections after the article's opening paragraph are now ORIGINS and STAGNATION. Egad! How about ORIGINS, FULL GLORY and STAGNATION? OK, I'll put what I wrote below in as ORIGINS, A NEW USER EXPERIENCE, STAGNATION. (New experience? The level of unification of several services under Goppher-as-a-client was bigger and better than anything before, even if there's never anything really new under the sun.)

"GOPHER EXPERIENCE" and "TECHNICAL DETAILS" SECTIONS: The just-inserted "A New User Experience" deals with the GOPHER EXPERIENCE. The short GOPHER EXPERIENCE section fails to capture the experience as widely TALKed about, but it does have good technical information. I will put it into the following TECHNICAL DETAILS section under a lower-level paragraph heading.

Please save this section and the illustration. It would be just great to introduce the **linkage technology** and **protocol characteristics** that supported the menu construction so well for sys admins setting up gopher servers. The GOPHER ITEM TYPES now listed so prominently in this article as it presently stands (Dec2012) are significant because they never had to be seen by users. Look at the what-was-it-like intro again (A NEW USER EXPERIENCE) and see if you can find here at the technical level what drove Gopher to be what it was. Is TCP to a "well-known" (specified) port number on a particular machine limiting in some way compared to what HTTP added? DNS at the IP level? There's more to Gopher/WWW differences than just adding HTML. (If I knew, I'd say. I pass the torch to you.) This is pivotal history, this is not "C-level" and, if you can link the different fate of Gopher and the WWW to technical details of protocol and coding, then this is more than "mid-importance" information for nostalgic old guys in rocking chairs.



Contemporary accounts offer a glimpse of what using Gopher was like when the program burst onto the scene, unified several resources, and created "Gopher space".

Exploring the Power of the Internet Gopher, by Lynn Ward; UIUCnet, Dec. 1992 - Jan. 1993, Vol. 6, No. 1

On a screen with 25 lines of 80 green characters each (no graphics then, long since remedied) you got a series of hierarchical menus. The top level menu might be all the departments and publications (e.g., the campus newspaper) of your university, which had invested in running a Gopher server as a way of delivering documents electronically, at least to the local community.

There were no search engines as we know them today. Available information was presented in a series of nested menus, intended to resemble a hierarchical file system of folders, something familiar to computer users who saw any search for information as trying to find that particular file (document) which held a particular answer.

If you wanted recent news for the women's volleyball team, you would go to the "Daily Crimson Newspaper" menu item and choose the "Word Search of Latest Month" item. When asked to enter a search string, you would enter the word "volleyball." All articles in the local Daily Crimson newspaper from the last month that contained the word "volleyball" would be listed as a separate menu. You could select which one to get first. In part for lack of bandwidth, the system's presentation was intended to appear as listings of files, without their content. There were no paragraphs from inside the "hits", each with the word "volleyball" in boldface type.

And if you don't know what the newspaper is called or even if it is available? Searching a top level menu called "Keyword Search of [all] Gopher Menus" with a keyword "daily" would get you into the Daily Crimson publication you wanted for your "volleyball" search.

Note that this core Gopher functionality searches one publication in one location only (the campus newspaper at a local university). Nevertheless, the system was robust and its power soon grew. The sys admins setting up any local Gopher server had freedom to change menu hierarchies and names independently of the underlying file systems. What enabled Gopher to give so many who worked with it a foretaste of the World Wide Web yet to come was the ability to add links to other Gopher servers around the world. Now users could hop from one server to another in "Gopher space" (the first "cloud") without thinking about a single underlying network address.

True, to actually get the text you wanted, you had to click on several links, and read through a menu each time before you choose another link. Yet that worked more rapidly than people might think today, and in the heyday of Gopher, much time was spent choosing and organizing links in layouts that could be grasped at a glance. Gopher became the text-handling, document-delivery system it was intended to be. A user could display a text document on her screen, save it to a file, print it out, or even e-mail a copy to another person on the Internet. Gopher became the dominant client for other information services: Wide Area Information Servers (WAIS), FTP, and Archie, a database of the files held by most of the major anonymous (public) FTP sites on the Internet. Initially, Gopher could search only one WAIS database at a time, and WAIS's "relevance feedback" tool (find content-similar documents) was not available.

The World Wide Web added graphics to text, but lost the menus. It took a while for increased power in communication (bandwidth a thousand times broader; 9600 baud MODEM to fiber optics), storage (arrays of disk drives each a thousand times larger; under 1 GB to over 1 TB) and computation (cheap servers; PC CPUs a thousand times faster) to permit us to regularly crawl the Web and catalog it for search engines. While Gopher's menu system seems quaint, we should remember the time between the decline in Gopher usage and the arrival of search engine ascendency. Back then, early World Wide Web users looked eagerly for lists of links ("my favorite links" pages), and users were anxious to bookmark good links that they might never find again. Without structured menus, users had taken a step backwards, but, with less structure, something with more generality and much greater power emerged by the dawn of the 21st century: the World Wide Web.
--end NEW TEXT
Jerry-VA (talk) 17:47, 6 December 2012 (UTC)

This reads like an ad brochure or high school essay. No offense, but the style doesn't fit Wikipedia IMHO. -- (talk) 15:25, 19 February 2013 (UTC)

More Browser Information Needed[edit]

I don't think this article explains how to access a gopher server. I think it says some OSs and or some browsers no longer provide the client and or browser user with access to gopher servers.

If you wanted to access gopher servers, what did you do? What could you do now?

I agree with the other commenters here that this article needs more work. — Preceding unsigned comment added by (talk) 18:45, 19 February 2013 (UTC)


The usage of Gopher (edit|talk|history|protect|delete|links|watch|logs|views) is under discussion, see talk:gopher (animal) -- (talk) 05:55, 17 January 2014 (UTC)