💾 Archived View for dfdn.info › dfdn › goingdark.gmi captured on 2024-12-17 at 10:35:47. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2024-09-29)
-=-=-=-=-=-=-
The title of this article is a play on words because "the end of the Internet" can mean two things. I mean both. I'm looking for the undiscovered places at the edges of the Internet, and I'm looking and waiting with anticipation for what may replace the Internet--a next Internet, or next Internets. Perhaps, if it occurs, the next Internet will be what many now call "the darknet". Perhaps it will be something completely different.
Back in the early 1990's, "Internet 1.0", as some have called it, was populated with websites created and maintained by hobbyists, people who loved computers and wanted to share their knowledge and enthusiasm with others. This was before the web had dynamic content, so web pages were composed mostly of text and crude, static, ASCII-character, monochrome, or 8-color graphics. You may have seen examples in computer games from the 1970's. The few companies that had websites had mostly single-page placeholders. The Internet wasn't a place where people bought and sold things. You could surf the Internet anonymously because no one was watching what you were doing or where you were going. Governments were not concerned yet about the Internet; they didn't even have websites. Websites existed mostly around information and file sharing. Although email existed, most of the general population were not on the Internet, so few had email addresses. No one ever asked for your email address. Instead, we used telephones and snail mail. We talked on our telephones. We didn't text on them. Back in those early days when the Internet suffered from a lack of content, some, including myself, were hopeful that it would grow into something great, and we could hardly wait.
I know I am not the first to point this out, but the Internet has become a very different place since its birth back in the early 1990's. Maybe I should say since its childhood in the early 1990's because the ARPANET has been around since the 1960's. In some ways, the Internet has improved, in others regressed. Although it has some great content now, it is over-commercialized to such a degree that you have to actively seek out any left-over hint of the old Internet. Internet users are tracked everywhere they go, both by companies trying to sell them things and by governments paranoid about what they may see, say, or do. The Internet today is well on its way to becoming an Orwellian nightmare, a fenced-in corral for Internet cattle. This is a very different future than the one I hoped for back in the early nineties. This future-now-present makes me nervous.
As a result of my growing nervousness and disillusionment with the Internet, I have been looking around for something better. I have been sampling alternative networks, including what some call "the darknet". I put "the darknet" in quotes because having seen some of it, I realize that label doesn't describe it very well. It is not even one thing. There are many so-called dark networks, with more appearing all the time. Unfortunately, I don't have a better label. I would like to call it the "next Internet", but something completely different and unforeseen could come along to fulfill that role. Perhaps "the altnet" is a better name. I don't know. I have tried a handful of "dark networks" so far--including ZeroNet, the Invisible Internet Project (I2P), the Interplanetary File System (IPFS), the distributed social network Diaspora, and others--some of which I have written about on cheapskatesguide.org. I have also tried some alternative email providers, including protonmail and tutanota. And, I have permanently switched from Google to duckduckgo for search because I now understand that Google's search results are geared toward further corralling the Internet cattle.
Although my search for the end of the Internet has led me to become increasingly knowledgeable about the dark web and darknet, I am by no means an expert. As I said, many darknets exist. And, I still haven't learned more than a small fraction of what can be known about the huge, diverse network space that is out there. I do, however, know enough to recognize that the view of the darknet we get from governments and the news media is myopic. In my opinion, what they tell us about the darknet is very similar to the somewhat self-serving warnings they have been giving us for twenty years about avoiding "shady" websites (i.e. anything other than Facebook, Amazon, Google, and websites run by giant corporations). For example, this article on the FBI's website refers to "going dark" as a problem that "...is eroding law enforcement's ability to quickly obtain valuable information that may be used to [identify] and save victims, reveal evidence to convict perpetrators, or exonerate the innocent." The FBI's article goes on to say, "With the widespread horizontal distribution of social media, terrorists can spot, assess, recruit, and radicalize vulnerable individuals of all ages in the United States either to travel or to conduct a homeland attack." While some of that may be true in rare instances--and I stress the word "rare"--a bigger picture exits here. The bigger picture is the one I'm concerned with.
A hint of the bigger picture can be found in the following comment that I found on a forum on the I2P network:
"I used to worry a lot about what would happen to Tor back when I was really active here. Services would go down and not come back, people running legit sites would get threats from [law enforcement] to share user info, Vigilante assholes would go and take down sites cause they didn't like the lack of censorship, and to top it all off rumours spread like wildfire throughout the popular media. Rumors that not only hurt the reputation of the [dark web] but of the people using it. You know the ones. The red rooms, the hitmen, human slavery, etc. All bullshit but it had a way of working it's way into the sensationalist media news cycle. But the more I've thought about it and the more I've had a chance to interact with the communities that use Tor the more I've realised that it isn't going away. Not for along time anyways. Services come and go but the fact of the matter is that there will always be something new to replace the old. I've seen it a million times now. A service goes down and is mourned by the community. Shortly after someone makes a replacement service and that new service takes the place of the old one. I know that tomorrow when I wake up Tor will still be here. That the community will still stand behind it build on it and improve it. And if it isn't then someone will have built a replacement. We'll still be here. Despite government opposition or regulation or the public's fear we'll still be here."
This is a picture of a wild-west kind of "community" that, despite seemingly existential threats, isn't going away. The reason this community exists at all is that many are beginning to wake up and realize that their ability to speak freely and anonymously on the Internet is being infringed upon. They don't like that, so they are creating a more secure, more anonymous Internet where they can be more free. In almost all cases, this is not about breaking any laws--except those against free speech. It's about taking back free speech rights curtailed by governments and corporations. The question in my mind is, "What is this community becoming?"
Despite the many warnings about the negative side of the "dark web", networks like IPFS, ZeroNet, I2P, Retroshare, Scuttlebutt, the TOR network, and others are much more well-lit than you have been led to believe. The dark web is nothing more than a loose conglomeration of websites hosted on various encrypted networks that require special software to access. This software is freely available on the regular Internet and can be downloaded and used by anyone who is interested. For the most part, the dark web is frequented and maintained by your neighbours, your work colleagues, people you go to school and to church with, and people you see at Walmart. The majority of darknet users welcome newcomers. Unfortunately, some are also hostile to newcomers, simply because they want to keep their networks from becoming nothing more than copies of the regular Internet. From what I can tell from months of perusing the dark networks that I've visited so far, denizens of the dark web seem to be somewhat younger than average, somewhat more intelligent, more forward looking, more free-thinking, more socially conscious, much more freedom-loving, and maybe a bit more paranoid. Perhaps, that is why governments around the world seem to be so determined to block and control them. To say more than that about the individuals who frequent the dark web would be, I think, to risk making an artificial distinction between them and the rest of society that is unjustified.
Unlike Internet 1.0, where most sites were run and visited by hobbyists who loved computers for their own sakes, today's so-called dark websites are mostly created and populated by people who have specific goals in mind. Those who love computers are still there. They are just not the majority. Many darknet devotees are activists who don't like what the modern Internet has evolved into and want a place where they can speak their minds unhindered by corporate censors kowtowing to advertisers and various freedom-averse governments around the world. Others simply want to be able to download copyrighted music, movies, TV shows, and books for free. And some, fortunately a very small minority, want to buy drugs and spread child pornography. But, for the most part, the darknet is just like the regular Internet--in the sense that it has a light side, and it has a dark side.
The new trend in the darknet world is the appearance of distributed, or peer-to-peer, networks. These are networks that transfer data between and sometimes store data directly on the computers of users, called peers, rather than relying on central servers. This type of network is very difficult for governments to block or control. And that is the main reason distributed networks are appearing, especially in countries like Russia and China.
As extensions of the darknet, the new distributed networks are similar in many ways to the early Internet. Nearly all of their websites are run by individuals. You'll find almost no commercial content. You will find blogs, chat rooms, forums, email services, gaming sites, file sharing sites, leaked government documents, lists of links to popular sites, and documentation on how to access and use their networks. Unlike the early Internet, however, many websites on distributed networks run javascript, so their content is often more modern-looking and interactive. Because of this, you'll also find rudimentary search engines and Youtube-like sites. However, because content is hosted by hobbyists with small budgets, were you to try to download very large files, like HD movie files, you would often find them to be unavailable. In fact, as was pointed out in the I2P user's comment above, many websites on distributed networks exist for some period of time and then disappear when their owners lose interest. So, having access to up-to-date lists of active websites can be important.
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
I see a new trend developing among geeks. Maybe I am just deluding myself, or perhaps it has always been this way. Maybe it's just wishful thinking--along the same lines that has caused me for decades to wish that the average engineer would suddenly begin to care about using his work day productively. But lately, I have been noticing more individuals who run their own websites writing articles about the value of text-only, static, and non-commercial websites. These hobbyist website designers seem to care deeply about the efficient transmittal of information. They seem highly frustrated by the bloated Internet on which the average web page takes 10.3 seconds to download and transmits next to no useful information to the user. An example, though you may want to avoid it if foul language offends you, is this seven-year-old website. Another article highlights some of the practical benefits of text-only websites. And, this article discuses the broader topic of the value of small, independent websites and the reasons they differ from commercial websites. Then, there is the question of information density, or the amount of information that should be shown on a web page.
The Streamlined Internet is Already Here--You Just have to Find It
Having a more streamlined Internet is about more than just text-only web pages because the Internet is about more than just web pages. Gopherspace and usenets are still around. Project Gemini is being touted as something to fill the gap between gopherspace and the Internet. RSS feeds seem to be once again gaining popularity among geeks. And then there is the Tilde Town server and its overflow onto similar servers. While the average Internet user may see Tilde Town as nothing more useful than an experiment to see how many clowns can be stuffed into a clown car, Tilde town and its imitators have drawn thousands of users to connect to their servers using the text-only technology of SSH. This is a command line world that the average Internet user under 30 likely cannot even imagine. But for thousands, these few servers not only provide free avenues for their creativity and comradery, they also boldly proclaim the truth that some much can be accomplished with so little.
Text-only news websites are also coming back. The reason for this is that some news agencies have finally recognized that the Internet may be less accessible in areas hit by natural disasters. Lower bandwidth connections may be available as backups at these critical times, but they require more efficient news sites to provide continued access to the news. These sites are also more accessible to people in emerging markets.
I should also mention distributed and dark networks like ZeroNet, Tor, I2P, and IPFS. These are perhaps some of the last strongholds of geeks on the Internet, and they are all about text. At least, that is the way it appears to me as I peruse the sites of these networks. And, by the way, these relatively new networks are populated mostly by geeks in the same way that computer hobbyists were once the source of most of the activity on the early Internet. Perhaps the text-only format of most sites on darknet and distributed networks is one reason average Internet users tend to shun these networks like vampires hissing at the sight of garlic.
Text is Beautiful
If you are not one who inherently feels the allure of text, it may not be possible for me to help you understand why text making a comeback is a good thing. You may view my attempt to explain as something akin to Greg Kinnear's character in the movie "You've Got Mail" waxing poetic about his type writer. And, maybe it is in some ways. I'll try to explain anyway.
Think of the difference between reading a book and watching a movie. Each has it's own strengths and weaknesses. A book uses words to convey the inner thoughts of a character in a way that a movie just cannot. Books increase your vocabulary. They also give you a window into the heads of enough people that you develop insights into people that you cannot to the same degree by watching movies. Books are a slow enough form of communication that you have plenty of time to think about the message the author is trying to transmit. If you are initially confused, you can go back and re-read sentences or paragraphs as many times as it takes to understand the full meaning. With a movie, if you sneeze, you may miss something important.
Perhaps the feeling of reading pure text on a computer screen is something that you just have to be of a certain age to fully appreciate. I can still vividly remember my college and graduate school in the days of VT100 terminals with green or amber text scrolling by at 300 baud against a black background, just a little faster than I was comfortable reading. I remember keyboards that seemed to have a full half-inch of key travel. I remember working on assignments in computer rooms during the day, surrounded by other students, having a shared experience, often of frustration, always of learning. At night--especially late at night, or on holidays--I was frequently the only one in the room. I would be debugging some code or running a simulation and after deep brain fry be ready for a few minutes of exploration. That was how I discovered email in the mid 1980's before that was a even a word that most people had in their vocabularies. Back then, email was commonly referred to as electronic mail. Later, it became e-mail, and then finally email. I also have a vague memory of usenet, though I didn't have time to delve into it much. I wish I had had more time in those days to smell the mainframe flowers. I remember better the text-based computer games of the early 1980's. Those were the days when text was king on microcomputers, as they were called then, simply because the computers that individuals could afford to buy didn't have the power to present the flood of eye-candy that we are force-fed today.
The allure of text is about more than just nostalgia. It is primarily about the transmission of information. Even in this Youtube-and-podcast-centered generation, nearly six hundred years after the invention of the printing press, text is still what we use when we are primarily interested in conveying in-depth information. Other media focus primarily on entertainment with information transmittal sometimes as a secondary goal. Information that must be digested slowly is still best conveyed with text.
The beauty of text is most apparent when we look at it through the lens of useful information conveyed. Text is the most information-dense medium on the Internet. We say a picture is worth a thousand words. How much real information is conveyed in this ten minute and 16 second, 1080p cat video on Youtube? I do not deny that entertainment has value and is therefore useful in some way. But, one of the things that irks me when I am looking for information about a computer-related topic is to discover that the only information available on the Internet is in the form of a Youtube video. This is the worst medium for conveying technical information that I can think of. However, I understand that the author of the video may have chosen that format because it was likely to deliver the largest audience of any format to which he had easy access.
Let's look at just how efficient text is in comparison to other media. Leaving aside arguments that can be made about what constitutes useful information, no one can deny that this cat video takes up 150.6 MB of storage space or that it takes about 100 seconds to download across a slower 12 Mb/s Internet connection. The same amount of storage space can hold about 5.2 hours of 64 Kb/s MP3 audio data or about 25,000,000 words of ASCII text. That is the equivalant of around 40,000 pages of text in a paper book. It is the equivalent of about 11,000, fifteen-hundred-word static web pages when the inefficiencies in hand-coded HTML and meta tags are folded in. It is more like 1,100 fifteen-hundred-word web pages coded by Wordpress software. Eleven thousand, 1500-word webpages read aloud at 160 words per minute would take about 1700 hours to read. Reading silently at 300 words per minute would take about 920 hours. So, I would argue that for each human-brain-absorable unit of 1080p video informaton stored on a hard drive, something like 30 units of useful audio information or at least 9,900 units of useful text information can be stored in the same space. That is nearly ten thousand times as much useful text information as 1080p video information.
But, why should anyone care about the efficiency of data storage and transmission? Hard drives are cheap. We now have gigabit per second Internet connections--well, at least a few of us do. Why in the year 2020 should we acknowledge the beauty of text as the most efficient means of conveying information? I can think of several good reasons. But first, let me just state that not caring how much space is taken up on a hard drive illustrates the mode of thinking behind the phenomenon of data expanding to fill the space available. This phenomenon has plagued humanity since we first learned to store information and has cost truly staggering amounts of money. I will now address the role of efficient data storage and transmission in free speech.
Text and Free Speech
Perhaps the most important reason to acknowledge the beauty of text in efficiently conveying information is that we have nearly lost the battle for control of the Internet to commercial companies whose only motive is profit at the expense of all else. I should also point out that companies are regulated by governments whose major concern is usually remaining in power. It appears to me that geeks increasingly reject the eye-candy that predatory websites of large companies use to lure childish "normies" into their cars. That sounds like I'm being overly judgmental of average Internet users. Maybe I am. I just feel that most of what is wrong with the Internet these days is the fault of the majority of its users. They have made themselves the cattle of the Internet by allowing themselves to be corralled into Facebook's and Google's corals. The cravings that normies have for inane eye-candy has given giant companies like Facebook the power to dominate the Internet in ways that are beneficial to Facebook and damaging to the rest of humanity.
The big question is to what ends are these companies using their power? And, how will they use it in the future? Facebook's primary goal is to keep users on its website for as long as possible. Most likely, Facebook could not care less about transmitting useful information, except as one means of keeping users on its website. Facebook and Google combined with other major companies now have the power to effectively bar a large portion of Internet users from accessing small websites, or any website that has not paid for eyes on pages through the purchase of key words in search results. I call this obscuration "the commercial smog of the Internet".
I have seen for myself that rather than using the key words that I specify in the meta tags of my articles, Google chooses key words on which to base its search results that are in my opinion largely irrelevant. The result is that Google does not lead readers to my website who are looking for the information that I have to impart. In my opinion, the vast majority of readers that Google sends to my website are led there by keywords that have little or nothing to do with the information that I have to convey. The result is that readers who happen to be led by Google to my articles most likely leave immediately. Google can then use this to justify an even lower ranking of my articles in its search results.
Google and Facebook effectively choose which websites many users visit. This means that they control the information users see. This is the real problem. Pervading the Internet that large corporations have established under the supervision of governments is the idea that users should be "protected" from untrustworthy websites containing unreliable or false information. But having set themselves up as the arbiters of truth, large corporations have put themselves in a position to filter out legitimate opposing views. They now have the power to erect censorship barriers to prevent individual voices from exposing fallacies in any narratives they may choose to present to the public.
In practical terms, this means that much of what regular people like you and I say on the Internet can be hidden from most Internet users. One example can be illustrated by an article I posted several weeks ago on one social media website. The article explained the reasons I no longer vote in political elections. It received close to 2000 page views before it was pulled less than half an hour after I posted it. No reason was given. My guess is that my article did not fit well with the world view that large corporations and governments are cultivating in their workers and tax payers.
Large companies have also very effectively barred all individual Internet users except the most determined and knowledgeable, I would guess perhaps 0.01% of users, from running their own email servers. So, we are for the most part forced to use company-provided email services. As a result, we are subject to their terms and conditions, which some have used to justify reading our private emails. Luckily, there are still alternatives like Protonmail, but one has to wonder how long they will remain available.
At least part of the problem of restricted free speech on the Internet could be solved with a more pervasive use of text that gives individuals the power to speak from their own websites using limited bandwidth and computing power. But before this can be effective, the average Internet user must be educated sufficiently that he becomes willing to go on an eye-candy diet. Only then, will small websites be able to meaningfully compete as a group with giant corporations for the eyes of average Internet users.
While I am on the topic of free speech, let me speak briefly about the mainstream Internet alternatives. Decentralized networks like ZeroNet, IPFS, I2P, and others that are run solely on the computers of their users are subject to many of the same limitations as small websites. They have limited storage and limited bandwidth because their users have limited storage and limited bandwidth. It may be difficult to see if you do not have direct experience using them, but believe me (or don't), it is true. They are mainly useful for transmitting text. They do transmit audio, pictures, and videos, but they do not do it well. I have seen first-hand how users from 8chan flooded into ZeroNet immediately after 8chan was shut down, only to leave a few weeks later when they could not get what they were looking for, a largely visual experience. The fact that decentralized networks have limited ability to transmit the eye-candy that the average Internet user wants means they will probably continue for some time to have only a small voice in the increasingly one-way conversation that is the Internet.
You are no doubt aware that this article begins with a picture. The reason I include pictures in most of my articles is in the unlikely event that an average Internet user stumbles upon one, he will be less likely to leave immediately. I try very hard to limit pictures to 50 KB per article. I do this for two reasons. The first is that in order to keep page loading times low, I have to reduce the bandwidth required. The second is that on rare occasions when a large number of readers want to read an article I have just posted on social media, I have to keep the amount of data associated with the article small enough to make that possible. If not for the fact that I am transmitting this information to you largely through the use of text, I would not be able to provide it for free. Being forced to take money from advertisers to pay for the functioning of this website would mean that I would likely be at least partially hampered from telling the truth as I see it.
ISP's do not Recognise Our Right to Run Websites from Home
Unfortunately, limited bandwidth and computing power are not the only problems that plague operators of small websites. In the year 2020, most Internet service providers do not recognize residential customers running their own websites as making legitimate use of their Internet connections. Have you ever wondered why? Is it simply because too few residential customers run their own websites? Perhaps. It cannot be due to the reason ISP's usually give, which is increased traffic. The average blog only receives something like 3,000 page views a month, or about 1.5 GB of data transmitted a month in the case of a Wordpress blog. That is the equivalent of only a couple of hours of Netflix watching. In the last six or seven years, I cannot recall hearing any ISP claim that watching Netflix is not a legitimate use of residential customers' Internet connections. Thank goodness that applications like Skype and interactive gaming are seen as legitimate needs, or residential customers would have virtually no upload capability at all!
Text Conserves Natural Resources
Another reason to acknowledge the beauty of text is that we have something like 4.6 billion people accessing the Internet. All that access consumes vast resources. By some estimates, the Internet uses 10% of the world's electricity. In the US alone, about 500 million tons of coal are burned per year just to generate electricity. If we can increase efficiency significantly by using text-only web pages, or even text-mostly web pages, wherever possible, we can save an enormous amount of resources. And if that seems too far outside the realm of your personal self interest to concern you, think how much faster web pages would load if we could only somehow convince marketers to stop using so much eye-candy and tracking software to lure buyers into making purchases.
Final Words
To wrap this up, let me just say that if text really is making a comeback among geeks, there are good reasons. Text has the potential to free us from the Internet information gatekeepers, and in so doing make us more free to speak our minds and actually be heard on the Internet. Text can also help us conserve more of our limited natural resources. When given a choice between the dispersal of eye-candy and the dissemination of useful information, the thoughtful practice of choosing text can be a continual reminder that just because we can fill our websites with useless junk does not mean we should. Alas, if only we can somehow educate average Internet users to stop craving eye-candy over substance.
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Welcome to the latest instalment of "Going Dark: Looking for the End of the Internet". This continues my on-going exploration of 1) lesser-known places on the Internet, 2) computer networks that can be reached via the Internet with additional software (e.g. darknets), and 3) networks that may someday evolve into the next Internet or Internets. This week's topic is the Gemini Project. Although the "network" or "space" defined by the Gemini Internet protocol may meet the definition of a darknet, I have not seen anyone refer to it as such. Perhaps that is because its primary goal is not to hide the identity of the user but to provide a useful network for accessing predominantly text-based content. Gemini's purpose is to present text more efficiently and aesthetically than the regular (or "open") Internet.
"What is the point of text-only webpages?" you may ask, especially if you are under 30. Gemini will probably not appeal to those who use the Internet primarily for entertainment, rather than as a source of information. But many, including myself, have lamented the demise of the 1990's Internet. We want an Internet with webpages that do not take an average 10 seconds or more to download--despite having very little user-readable content, let alone content we may actually want to read. We yearn to return to the days when we could actually find non-commercial websites with an Internet search engine. Remember the days before about 2007 when a Google search could yield millions of search results, and Google would let you access as many as you wanted? Now, we get only a few pages of results that Google thinks are worthwhile. Though I have no proof, I suspect these may be mostly websites that have paid Google for the privilege of appearing in its search results. Go ahead and call me pessimistic. Perhaps I am.
No problem can be solved until someone has first pointed out that it exists. One encouraging result of all this lamentation over the end of the 1990's Internet is that some capable programmers have begun writing software to solve the problems of the present-day Internet. As a result, the world has received several "darknet" peer-to-peer networks like I2P, Freenet, IPFS, ZeroNet, and others. More are emerging all the time. In June of 2019, we received another possible solution to the bloated, corporatised Internet, the Gemini Project.
How is Gemini an Improvement over the Networks We already have?
Gemini shares some similarities with, but is in other ways better than, our old warm-and-fuzzy early 1990's friend, the Gopher network, which is currently just barely hanging on by the skin of its long yellow rodentine front teeth. Like Gopher, Gemini is all about text-based pages, but Gemini provides more functionality than Gopher. The value of Gemini is that it works more like an earlier version of the HTTP-based World Wide Web, before Web 2.0. The Gemini protocol is analogous to the HTTP and HTTPS protocols but is not just a stripped down version of either. Unlike Gopher, Gemini supports TLS. The Gemini protocol does not allow JavaScript to be downloaded and run automatically on a user's computer when he visits a Gemini page, nor does it allow a number of other practices that cause the current World Wide Web to be overly complicated and slow. Gemini does allow manual downloading of files of any type, so one can download images, pdf's, software, etc., if one chooses. More information about the Gemini protocol can be found on the FAQ page of the Circumlunar Space web page.
The simplicity of the Gemini protocol brings other important advantages over the web. One is that developers can much more easily create browsers. This means a giant company like Google should not be able to dominate the Gemini browser market. Thus, a single company is unlikely to be able use its browser's dominance to unilaterally redefine the Gemini network the way Google is now doing with the web. For users, Gemini promises super fast browsing, none of the security problems associated with Javascripting, no pop-up advertisements, and none of the ubiquitous tracking found on the web. Gemini does not support the annoying animated advertisements that are designed to distract you from the information you are hoping to obtain from the websites you visit. By not allowing images, videos, and JavaScript to be automatically downloaded, the Gemini protocol vastly reduces the amount of data associated with the typical webpage. This can be especially valuable to users who happen to be on metered or slow Internet connections. Gemini is also designed to be difficult to extend in the future so that it will continue to be fast, simple, private, advertisement free, and text-based.
Internet users who are very privacy focused should appreciate the Gemini network. My understanding is that the privacy advantage of Geminispace compared to the web has much to do to the absence of JavaScript. The absence of JavaScript means cookies and tracking software cannot be downloaded to users' computers. In addition, the Gemini Protocol does not provide a Gemini server with a visitor's browser information via a "user agent", nor does Gemini know the last site that the user visited, via a "referrer header". For those who want even more privacy, Gemini servers and clients can be run over the Tor network. Notes on using the Tor network with Gemini can be found at:
gemini://tilde.team:1965/~tomasino/journal/20200706-gemini-on-tor.gmi Notes on using the Tor network with Gemini
I will explain how to access Gemini pages later in this article.
gemini://gemlog.blue:1965/users/acdw/1595607025.gmi Here is an informative Gemini blog page that explains why the author of the blog likes and uses both Gemini and Gopher.
The Gemini protocol has some shortcomings. Gemini has no support for the resumption of the downloading of large files after an interruption. This may mean for example that no one will ever create a Gemini site for downloading HD movies. According to the FAQ page on the Circumlunar Space website, "The TLS requirement of Gemini limits it to more modern machines." My understanding is that this means one will never be able to surf Geminispace with say, a Commodore 64, but he should be able to do so with a computer powered by a Pentium III. This makes Gemini slightly less appealing to me because I would prefer a network that can be accessed even by very old computers. While I may personally never have the need, I can envision the potential for the widespread adoption of very cheap Internet appliances among school children in less-developed countries. Thanks to their overlapping goals, my guess is that Gemini will grow at the expense of Gopher. This may inadvertently extinguish the only readily-available computer network in the world for extremely low-powered Internet appliances or computers from the 1980's and 1990's--except, perhaps, for a few dial-up BBS's.
Although the Gemini network is usually very fast, at times, it seems to slow significantly. My guess is that this may be due to a temporary slowdown in the world-wide computer network that supports the Internet, upon which Gemini depends for the transmission of data. If we only had a separate, world-wide mesh network... But, that is a topic for another article. Gemini's occasional sluggishness may also be due to localized, insufficient computer resources being dedicated to it at the moment, but I am only speculating.
The Key to the Gemini Door
Geminispace is accessed over a regular Internet connection. Other than that, all one needs to begin surfing in Geminispace are a Gemini browser and the URL of a Gemini page. Since Gemini is a protocol, like HTTP or HTTPS, Gemini URL's begin not with "http://", "https://", or "ftp://", but with "gemini://". If you prefer not to go to the trouble of downloading and installing a Gemini browser, you can also get an idea of what Geminispace looks like by using this web proxy: https://portal.mozz.us/gemini/gemini.circumlunar.space/ .
A list of Links to source code for eight Gemini browsers can be found here on the web. Unfortunately, unless you have some in-depth knowledge, you may not be able to get any of them to run without some help. So, I will explain how to get the AV-98 browser to work in Linux. But first, let me say something to developers in general. Hey guys and gals, how about bundling some documentation with your code? Does everything you create have to come with its own built-in I.Q. test, so no one with an I.Q. below 140 can figure out how to use it? And while you are at it, how about providing documentation in some standard file format, like maybe ASCII text or a version that is compatible with Microsoft Word? And, would it kill you to provide an x86 binary?
I looked at four of the most promising browsers in the above list: AV-98, Castor, Geminaut, and Kristall. Geminaut looked especially promising because it is supposed to run in Windows. Unfortunately, it came with zero documentation, and I could not get it to run in Windows Vista or Windows 7. I did not try it in Windows 10, because I refuse to use Windows 10. I also tried installing Kristall on my Linux machine, but it is dependent on four or five packages, including QT-5. When I tried to install QT-5, I was directed to the QT website that insisted that I create an account before I would be allowed to download their software (which may or may not have installed successfully on my computer). I draw the line at being forced to create an online account for a piece of software that I only intend to use once. So, I aborted my Kristall installation right there.
If you are familiar with Linux, you are aware of the problem. Much of the Linux software out there has dependencies. And, those dependencies have other dependencies, and so on. What is supposed to be a simple installation sometimes turns into a marathon session of identifying, locating, and installing packages from several different software repositories until you finally run into a package that you absolutely cannot find a compatible version of anywhere on the Internet. If this process goes on too long, you also run the risk of destabilizing your installed Linux distribution. Not fun! Perhaps the solution is to find some Linux distribution that is designed for developers that already contains every piece of development software known to man. I just don't want to go that far. I prefer to run light-ish Linux distributions, not those with everything including the kitchen sink slowing down my computer and requiring 32 GB of RAM to run. I am sure many readers disagree with this philosophy. If you do, feel free to leave a comment telling me why you think my head is full of mush.
I really wanted to try the Castor browser, but it has as many dependencies as Kristall. So, rather than putting myself through that, I chose to install the AV-98 browser that has no dependencies other than Python. "Python" is a Python computer language interpreter that resides on the majority of Linux computers, so this should present no problems. The "README.md" document that comes with AV-98 says two optional "dependencies" can be installed for support of multi-colored text and increased security, but you do not have to install them to get AV-98 to work. The base AV-98 browser is a very simple single-colored-text terminal emulator, so it does not deliver the full Gemini experience that can be obtained with a more modern-looking browser. But AV-98 will put words on a screen, which should satisfy those who are merely curious about Gemini's content.
To install the AV-98 browser, begin by downloading it here.
To download the compressed AV-98 file, find the download icon on the far right side of the page that says "Download Repository" when you hover your pointer over it. When I tried downloading and extracting the "TAR.GZ" version, my computer displayed an error message saying, "This does not look like a tar archive". So, download the "ZIP" version of AV-98. Then, move it to where you want it on your Linux computer's hard drive and extract its contents. Now, go into the AV-98 directory that you just extracted. The setup.py script appears to do nothing. If Python is on your computer, double clicking on the "av98.py" script should open a terminal window with an AV-98 prompt. If this does not happen, try typing "python --version" at the Linux command line. If you have Python installed on your computer, a message like this should appear: "Python 2.7.18". Unfortunately, not all Python software is compatible with all versions of Python. So, if you have Python, but AV-98 does not run, that may be your problem.
Once you have an open Linux command window with an AV-98 prompt (i.e. an AV-98 terminal window), everything gets much easier. Typing "help" (without the quotes) at the AV-98 prompt produces a list of 31 commands. Typing "help" followed by a command tells you what the command does. Typing "go gemini://gemini.circumlunar.space" takes you to the Gemini page for Circumlunar Space. If you get this error message instead, "ERROR: DNS error!", you miss-spelled the URL. Each additional page that can be reached from the Gemini page that you are currently viewing has a number in front of it. Typing the number at the AV-98 prompt takes you to that page. Typing "back" takes you back to the page you just left, just like the "back" button in any web browser. Typing "url" gives the URL of the Gemini page you are currently on. The AV-98 command line also provides direct access to the Gemini Universal Search engine (GUS). Just type "gus" followed by the key words you want to search for. One thing that distinguishes GUS from search engines on the web is that when GUS is used to search for the word "blog", it actually yields search results that include personal blogs! Type "exit" to close AV-98 when you have finished. Easy, right?
In space no one can hear you scream. Is the same true in Geminispace?
Well, Geminispace has some air. Since it is still very new, Geminispace is not teaming with content like the web, but it does have content. On the Gemini page of Circumlunar Space, you will find, among other things, links to two search engines, two Gemini-to-web proxy services, downloadable Gemini software, and about 50 Gemini servers. Another list of links to over 200 Gemini severs (many of which are inactive) can be found at gemini://gus.guru/known-hosts. (Update 3-8-2022: A new search engine called "Totally Legit" Gemini Search (TLGS) now exists at gemini://tlgs.one .)
Content that is created entirely or almost entirely by hobbyists tends to disappear. This is just as true in Geminispace as it is in Gopherspace or on the dark networks. This is simply a fact of life, so just enjoy reading whatever content you find. Hopefully, Gemini will become more popular, and more people will create content in the future. Maybe even YOU will create content.
One thing I love about Geminispace is that anyone can put a Gemini server on line and host content. You do not need anyone's permission. You do not need to pay a fee to ICANN, the agency that manages domain names on the web. You do not even need to identify yourself. I have not yet found any detailed documentation that explains why any of this is true. My guess is that it may have something to do with private TLS/SSL certificates, but I do not know how the Gemini DNS system works. Perhaps, since no commercialisation yet exists in Geminispace, no one has had an incentive to cause any problems, so the system is still being designed?
To get an idea of how much content existed on Gemini in the autumn of 2020, when I wrote this article, I went to GUS and performed a search. I typed in the name of our old friend Bill Gates and found 401 search results. Then I went to the web. Somehow, with JavaScript disabled in Firefox 75, I managed to get Google to tell me the number of search results that it produced. I was not able to reproduce that feat the next day. Google said it had 78,600,000 results for "Bill Gates", but I would guess it will probably not let us see more than 20 or 30 pages of results. I typed "model rockets" (with the quotes) into Google, and it produced 759,000 search results. It would only let me see 14 pages of them. Where are the other 758,700 search results, Google? When I typed "model rockets" into GUS (without quotes, because GUS does not understand quotes), I was told GUS had found 510 search results. I think this is not because model rockets are more popular on Gemini than Bill Gates (as they should be), but simply because what is actually being searched for is anything with the word "bill" or the word "gates" versus anything with the word "model" or the word "rockets".
Personal blogs (or gemlogs) contain some of the most interesting material on Gemini--at least, I think so. In the Fall of autumn, perhaps only a few hundred existed. Many more have been created since. Included among the links below are some of the best blogs and collections of links to blogs that I have found:
gemini://rawtext.club:1965/~sloum/spacewalk.gmi
gemini://zaibatsu.circumlunar.space/~solderpunk/phlog/
gemini://alexschroeder.ch:1965/do/more
gemini://tanelorn.city/~bouncepaw/gemlog/
gemini://gemlog.blue:1965/users/acdw/
gemini://tilde.team:1965/~tomasino/
gemini://supernovas.space
gemini://gemini.conman.org/boston/
gemini://park-city.club:1965/~invis/phlog/
gemini://gemini.circumlunar.space/servers/
gemini://flounder.online
gemini://gemlog.blue/users/birchkoruk
gemini://gemlog.blue/users/twh/
gemini://gemlog.blue/users/left_adjoint/
gemini://gemini.circumlunar.space/servers/
gemini://tilde.pink/
gemini://gemini.cyberbot.space/gemlog/
gemini://gemlog.blue/users/NetCandide/
gemini://medusae.space/
gemini://envs.net/
gemini://envs.net/~negatethis/
gemini://envs.net/~seirdy/
gemini://envs.net/~mukappa/
gemini://jayeless.flounder.online/gemlog
gemini://republic.circumlunar.space/users/korruptor/blog/index.gmi
gemini://simplynews.metalune.xyz/theguardian.com
gemini://midnight.pub/
gemini://kwiecien.us
gemini://samsai.eu/gemlog.gemini
gemini://calcuode.com/gmisub-aggregate.gmi
gemini://gemini.circumlunar.space/capcom/
gemini://gemini.circumlunar.space/~solderpunk/gemlog/the-standard-salvaged-computing-platform.gmi
gemini://inconsistentuniverse.space
gemini://gemini.cyberbot.space/smolzine/
gemini://gemini.rlamacraft.uk
gopher://gopher.black:70/1/phlog
gemini://spikydinosaur.com/
gemini://bbs.geminispace.org>/li>
gemini://station.martinrue.com
gemini://tlgs.one/
gemini://carcosa.net/journal
gemini://beyondneolithic.life/
gemini://ywcafe.cities.yesterweb.org/
gemini://warmedal.se/~bjorn/posts/
gemini://rawtext.club/~ploum
gemini://gem.saayaa.space
gemini://gemini.thegonz.net/cdg
gemini://michaelnordmeyer.com/gemlog/
gemini://laniakea.rodoste.de/journal/index.gmi
gemini://rawtext.club/~winter/gemlog/
gemini://willowashmaple.xy
gemini://gemi.dev
This concludes the latest instalment of "Going Dark: Looking for the End of the Internet". I think one of the most enjoyable and interesting activities in the jungle of the Internet is discovering little-known parts or off-shoots. One of my goals for the Cheapskate's Guide is to motivate readers to do their own exploring. Hopefully, you now have enough knowledge about Geminispace to motivate you to visit it. In the years to come, I look forward to seeing Geminispace grow and flourish with many more people producing interesting content.
I just realised that I have not mentioned free speech even once in this article. Perhaps I will just have to write more about Gemini in the future.