Gemini

I started writing this hot-take on Gemini, and it got less hot, and then kind-of stalled because I felt like I wasn't really making a case. Did I have a case? It was coming across as anti-Gemini, and I'm really not anti-Gemini, but ...something was causing some me mental friction. So, I went and read the Gemini FAQ (yes, I should have started with that instead of the Wikipedia page, I know), and while that helped assuage most of my knee-jerk reaction to the project, there were still ...lingering irks.

Anyway, we'll come back to my final thoughts at the end, but this was my false start:

----- I get where the Gemini people are coming from. The Web is too complicated, too hard to implement. But... why something new in this space? Why not just create a community around using a Web subset that is simple? A very simple Web server that only responds to GET can be written as easily as a Gemini server. A very simple HTML subset browser can be written as easily as a GemText browser. The difference would be that it interoperates with the Web. Imagine that!

Moreover, why do we need a new lightweight document retrieval protocol anyhow? We can agree that the contemporary Web has moved very far from its original purpose, being that of a linked document publishing system, but how are a newly-designed, intentionally feature-restricted Gopheroid protocol and document format going to solve that problem? If Gemini provides, capability-wise, something between Gopher and Web 1.0, then what ELSE does it provide to make it a compelling alternative to the fully-functional early-Web ecosystem that lives and breathes in and amongst the contemporary Web?

I have some writing out in Gopherspace; have had since 2013. I have dabbled. I like Gopher. I, too, was attracted by the brutal simplicity of the thing, which can be a real breath of fresh air after a particularly rough day of Web content interaction (and Web development, surely). But Gopher is one of those things that I am glad I know, but which I also am glad that I don't have to use.

Its brutal simplicity can be illuminating; it gives one perspective on the way the Web has gone, what's been left behind, and it inspires one to think about how it could go from here – or what alternatives there might be. But at the end of the day, Gopher is not where I publish. That brutal simplicity, its content limitations, its anachronism – using Gopher is making an intentional step back in time to the early '90s – it doesn't quite do what I need it to do.

If I'm understanding the Gemini community's aim correctly, it's to return to a document-oriented online publishing system, where writers can write and readers can read, and where commercial and audiovisual media intrusions are gone because they simply don't have the facility to intrude into or exploit the space. It takes Gopher and the early Web as its inspiration, but is a completely new thing, and has learned from the mistakes of the past.

That sounds great!

But Gemini bothers me. And I've been racking my brain about why, because in so many ways I'm fully on-board with the project's ethos.

----- At this point, I read the FAQ. <https://gemini.circumlunar.space/docs/faq.gmi>

It makes a clear case about why committing to a Web subset doesn't meet their goals. It makes a good case for why the platform is NOT intended to be extensible. It makes, all in all, a very good case for its existence in the niche it means to serve.

And I guess what I had to realize is that Gemini's niche is not my niche. Something about Gemini calls to me, but when I get close I realize that it's not FOR me.

The protocol is tight. I like it because it fulfills its design criteria well. Those design criteria are not my criteria for a hypertext [sic] system protocol, though.

GemText is a great design, in that it fixes everything that sucks about gophermaps. It's not, however, what I want from a document format.

I do like Gemini as a sort-of contemporary implementation of a Web-ish Gopher-like system. I think it nails that. And I think that for that, it deserves the enthusiasm and evangelism that its been garnering lately.

But Gemini bothers me, and it finally occurred to me why.

In spite of the fact that Gemini is coming from a very old-school point-of-view, and is very open and communitarian, it bothers me for the same basic reasons that I'm bothered by contemporary lets-go-viral-and-die app/content-platforms like Twitter, Instagram, Vine (remember Vine?), et al: it is a system with arbitrary content/format restrictions designed to control the aesthetic of the space. To publish on Gemini is to say, yes, this narrow space between plain-text and HTML 2 is all that I need, and all that I want to see; it is to join a community of elective constraints, technologically enforced.

Gopher's limitations were not aesthetic. Gophermap was not a prescriptive design; it was not a "this is all you need" design — it was a "this will do the job" design for annotated menus. Prose was plain text. In 1991 that wasn't minimalism, it wasn't an aesthetic choice, it was practicality.

The strength of the Web is, in my opinion, that it is this chaotic and endlessly extensible shit-show: a stupefyingly banal carnival of high-tech, low-brow, commercial fluff, with pockets of intellectual weight and absolute weirdness. It is a technological tower of Babel. And you can view it through a filtered lens -- you can browse the Web from a text console, you can turn off Javascript and disable style-sheets and limit your browsing to a few sites or communities that scratch your personal itches -- and no, it doesn't all interoperate when you do these things. When you say no to Javascript: pow. Half of the contemporary Web is broken.

But, when you say yes to Gemini, how are you in any better shape?

It's a /clean/ experience. I get that.

But at the end of the day, my impression of Gemini is that it's more about fashion than utility. I value simplicity, but I also value generality and extensibility, and because Gemini doesn't scratch those itches, I guess it's not for me.