Slashdot Mirror
The Rise and Fall of the Gopher Protocol (minnpost.com)
An anonymous reader writes: Tim Gihring at MinnPost talks to the creators of what was, briefly, the biggest thing in the internet, Gopher. Gopher, for those who don't know or have forgotten, was the original linked internet application, allowing you to change pages and servers easily, though a hierarchical menu system. It was quick, it was easy to use, and important for this day and age, it didn't have Flash.
The article remembers Tim Berners-Lee describing the idea of a worldwide web at a mid-March, 1992 meeting of the Internet Engineering Task Force, at a time when Gopher "was like the Web but more straightforward, and it was already working." Gopher became magnitudes more popular -- both MTV and the White House announced Gopher sites -- leading to GopherCons around the country. Just curious -- how many Slashdot readers today remember using Gopher?
The article remembers Tim Berners-Lee describing the idea of a worldwide web at a mid-March, 1992 meeting of the Internet Engineering Task Force, at a time when Gopher "was like the Web but more straightforward, and it was already working." Gopher became magnitudes more popular -- both MTV and the White House announced Gopher sites -- leading to GopherCons around the country. Just curious -- how many Slashdot readers today remember using Gopher?
I remember Gopher well. It was the early nineties and I would peruse computer networking and programming topics but I also stumbled upon so many Dungeons and Dragons resources in my Gophering. I don't know if the age of the memory is tainted somehow but it seemed like Dungeons and Dragons players were big early adopters of the technology. I am interested in what other people found on Gopher. Maybe it will help me put my own experience with it in perspective.
I remember it, at 65, actually I remember huge batch only mainframes. On a more serious note, I have a lot of time for Gopher, Lynx and all the 'simpifiers', I'd prefer everyone to have knowledge and communication at a low bandwidth rather than adverts, emojiis (whatever they are) and pictures of cats. My vision, going forward is goodbye port 80 and port 443, let's start again.
On y va, qui mal y pense!
University of Minnesota, wasn't it? I remember a story about the end of the last gopher server some years ago...
I didn't actually use Gopher that much, though I knew about it. My main memory of Gopher was around 1995 when I was a research student again. I was interested in such search tools, and I remember searching on usenet for relevant groups. I was actually expecting a different one to be more important, though now I can't even remember what that system was called. However, what I actually noticed was that something called WWW seemed to be far hotter and more active than any of the systems I had heard of before that.
The browser was the predecessor of Netscape that became Firefox, but I've also forgotten its name. What I remember was faking MathML with some version of Tex or LaTeX to create my equations as graphic objects so I could insert them into my first HTML webpages. Strange detail to remember after all these years, but the main hassle I remember overcoming was getting the background colors to be the same so that the graphic objects (equations) seemed to be part of the text.
Freedom = (Meaningful - Coerced) Choice != (Speech | Beer^2), and sad sock puppets' bad mods avail them naught.
the internet was not yet open for business. It had been built on dot-mil and dot-edu, on public funds. Programmers shared source code; if you needed something, someone gave it to you. A dot-com address was considered crass.
The internet was better then.
"First they came for the slanderers and i said nothing."
I recall having a soaring conversation with a tech friend in a Seattle back-yard party about this rumored "new thing" that was going to revolutionize the world. It was like Gopher, but had the ability to transparently serve all types of media and links were network-agnostic.
Frankly it blew my mind, and I had some difficulty wrapping my head around the concept, but most interestingly, we both found Gopher as the common-ground existing paradigm to compare against the nascent Web.
Then I threw up in a bush, but I think that was the Jim Beam.
I remember designing Gopher sites in grad school... and during the course I asked my prof if he'd mind if I did some WWW sites. That was 94 and we'd had Gopher, Archies and Veronica servers around, oh, and wais. Everyone should check out ED Krol's the Whole internet, if you can get the 1992 edition. It is a beautiful description of everything that was out there. https://en.wikipedia.org/wiki/...
https://www.tandfonline.com/doi/full/10.1080/1369118X.2013.808365
Mainly I remember it was often a pain getting to much of the information I wanted. The web was such a huge leap forward in terms of navigation - it's no wonder everyone quickly moved on from gopher.
#DeleteChrome
Token Ring is what Sauron gave to the lady of that upstart race of fairies nobody wanted to know about. I think he found it in a box of cereal.
If construction was anything like programming, an incorrectly fitted lock would bring down the entire building...
"the point of Token Ring"
Enlighten us.
Slower than TCP/IP, but 100% deterministic network behaviour and speed.
Basically it's what you want to run your Nuclear Power Plants, live-saving medical devices and bizarly expensive "failure is not an option" Space Equiment with.
We suffer more in our imagination than in reality. - Seneca
Some gopher servers still exists, I was too young when the protocol first became popular, however I wanted to learn about it so loaded up a gopher client to see what was still out there a few years ago, decided to repeat that a couple of months ago.
You keep comparing TCP/IP and Token Ring when they are different layers in the stack. Maybe you mean Ethernet rather than TCP/IP?
The point of Token Ring was to have plugs and sockets where it was impossible to put the plug in the wrong way around - or the right way around.
And indeed, you could simply plug two plugs together.
I used to do quite a few joint pitches with IBM sales-folk back then, and it was amusing to watch the show as they addressed the question of which networking hardware to go for. The plot was always the same. At early meetings they would say, "IBM sells both Ethernet and Token Ring and we recommend whichever is most appropriate for each customer. We'll need to learn more about your particular requirements before we can say which one is more suitable for you." Then, several meetings later when lots of things had been discussed, but nothing really relevant to the networking hardware the message would become, "Now we've had a chance to assess your particular requirements, we can say that for your particular case Token Ring would be better." It was always Token Ring, and never any explanation as to why.
The real point of Token Ring was that IBM owned it and they didn't own Ethernet. It set out to solve a problem which didn't exist if you designed your network properly in the first place (overloaded Ethernet provides poor service to everyone) and introduced far more of its own. Like so many IBM technologies, it was a mess. Don't get me started on APPC.
I got my first Internet porn from Gopher. I had no access to Usenet then, but one University offered access to Usenet through Gopher and that included alt.* hierarchy and specifically alt.sex.* hierarchy.
Basically it's what you want to run your Nuclear Power Plants, live-saving medical devices and bizarly expensive "failure is not an option" Space Equiment with.
The design intention was to produce something more reliable, but the implementation failed miserably. Those clunky connectors where you could never be quite sure that they'd mated correctly. Lost tokens resulting in periods of no connectivity. The fun game of going into your comms cupboard and unplugging each lead in turn from the MAU in order to plug in that magic reset gadget, which might then restore connectivity to your LAN. Token Ring was a train-wreck with an awful lot of money pushing it to big business.
It's still around. There's a small but passionate community surrounding gopher right now. A good portion of them are doing it to move away from the Web since it's become so commercialized and the technology is becoming so large and unwieldy that security is a real concern. Some make a gopher hole to mirror their website, or vice-versa. OverbiteFF is an extension you can use in Firefox to access gopher, or you can use a gopher-to-http tunnel or use lynx (not links or elinks). Lynx will even automatically use UTF-8 so you're not constrained to ASCII when you browse gopherspace.
I've considered creating an anonymous BBS or forum for gopherspace. The input links in gopher are largely under utilized; a piece of software that used those to accept input and handled linking in a smart way could get a nice, trimmed-down forum that still had much of the features you'd come to expect from community software. The best part is it's pure text and its limitations prevent a lot of the bullshit that's been tacked onto the Web.
That said, the community is super small and may remain that way due to its relative lack of maturity in server software. As far as I know, there are no packages/zip files/whatever that you can extract to a gopher-controlled directory and get an extra feature tacked onto your gopherhole. Until we get some fun projects like that, gopher will remain small. imo the best types of projects are those that abstract the server entirely and guide the user to manipulate the file-system, which falls in line with much of the content that gets served: often text files that you have a script generate a gopher index for as needed.
The cool part is you aren't constrained to a language at all. Serving Python over the web, for example, can be a hassle. Hooking a language up for gopher just needs the ability to process stdin (if needed) and returned either plain-text or valid gopher indexes to stdout. You could probably even write a gopher script in Brainfuck if you cared enough :)
Slower than TCP/IP, but 100% deterministic network behaviour and speed.
Basically it's what you want to run your Nuclear Power Plants, live-saving medical devices and bizarly expensive "failure is not an option" Space Equiment with.
Slower than Ethernet (you can run TCP/IP over token ring) but 100% etc etc. The original advantage was that you could do it with no additional network hardware. The problem is that it never worked that well in that mode, so we got MAUs to implement a star-wired network, and CAUs ("cows") eventually came out with more advanced networking features. The problem is, by the time token ring was actually usable at a whopping 16 Mbps, ethernet was up to 100Mbps for less per NIC than 16 Mbps token cards, and you could afford to drop some packets on your network whether you were using TCP or not. Since nobody was using token rings in the classic fashion any more and just star-wiring all nodes through
CAUs (if they could afford it) for reliability, the industry discovered that token ring no longer had any benefits and it died rapidly.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
less free-form.
www was not intrinsically better than gopher. It won out because there was more free porn accessible with it.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
You realize that TCP/IP is a layer 3/4 protocol while token is a layer 1/2 protocol as such they have nothing to do with each other really. The first networks I installed were TCP/IP over token.
Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
Token Ring ... "Slower than TCP/IP,"
There's no basis for comparison between the two. Token Ring is a link layer technology (ISO Layer 2), and compares to Ethernet, not IP. IP will run on both.
"National Security is the chief cause of national insecurity." - Celine's First Law
Back in college, Gopher was my first introduction to the Internet. I remember excitedly clicking from link to link, amazed at the information at my fingertips. Then, I got to a link titled "Middle East" and suddenly got worried that I would get in trouble for incurring long distance charges for my college. I closed it down and left.
The next time I went to the computer lab, I had a better understanding how networking worked (and why there wouldn't be long distance charges no matter what link I clicked on) and explored Gopher further.
My sci-fi novel, Ghost Thief, is now available from Amazon.com.
ARCNET
Token ring was a ripoff of ARCNET which was the first token based network protocol. Widely adopted in the 80s.
IBM needed to have their own proprietary network to sell to corporate types so they "invented" Token Ring.
I don't read your sig. Why are you reading mine?
Lynx and Gopher sucked mostly [..] Telnet on the other hand was pretty cool and could do a lot, but was massively underused.
I'm not sure that comparing Gopher and Telnet in that way is even meaningful. Perhaps I misunderstood the point you're trying to make, but the fact that you say Telnet "could do a lot" suggests you don't realise you're comparing apples with oranges.
Telnet itself was little more than a text-based terminal facility for accessing remote systems; that's not a criticism, since this is what it was meant to do. Of course, you can provide pretty much any (text-based) facility you like over that connection- which I guess is why one might say you can do "a lot" with it- but telnet itself is still just a remote access facility for all that.
Technically and functionally, it's not the same thing (nor intended to be the same) as Gopher, or the World Wide Web.
If you added so much to it that it became anything plausibly akin to Flash, then I'm not sure it would be Telnet as we recognise it any more.
(#) And Lynx was just a browser that happened to support both Gopher and the Web; it wasn't a protocol in itself. Unless you're referring to the unrelated modem protocol, in which case the comparison makes even less sense.
"Slashdot - News and Chat Sites Deviant". (Click "homepage" link above for details).