Books of 2015, #51: “A Fire Upon the Deep”, Vernor Vinge

I’m a little bit behind. OK, a lot—I read this back at the end of September.


I don’t actually remember when I first read A Fire Upon the Deep; probably 20 years ago or so. I’ve read it three or four times since—maybe more—but it’s been a while since the last time.

It’s a better than average adventure story. It features a lot more alien aliens than most similar material, it doesn’t hinge on anyone being particularly stupid, and while there is, arguably, a deus ex machina, it’s virtually laid out as such on the first page, so I’m not sure it counts.

Still, I find its nominal prequel, A Deepness in the Sky to be superior.

These days, though, it mostly makes me remember Usenet, because the idea that any sort of interstellar network would end up looking like Usenet makes a lot of sense—bandwidth would be expensive and relatively scarce (not unlike the state of networking when Usenet was started), so messages aren’t going to be holograms and audio, but some sort of text-analogue.

As an aside, I had long forgotten that Usenet originally developed just up the road(s) as a way to link UNC Chapel Hill and Duke…until last year when I had the unexpected opportunity to meet Steve Bellovin at a conference my wife was attending in Berkeley, and his mention of his connection to UNC reminded me.

Anyway, I remember first reading Usenet using a spectacularly bad newsreader on an IBM mainframe in 1990 or so. It was an amazing revelation, really: while the web was non-existent, and the Internet was, generally, far far smaller then than it is now, Usenet was already sprawling. There were topics for everything. I suspect I hung out mostly in groups that were concerned with the Atari ST I had at the time, though I know at some point I downloaded the source to the last version of Conroy’s microEmacs, because I hacked on it some while I was still in school.

It was, in fact, around this point I stopped doing so much with Borland products—and gradually the Emacs keystrokes supplanted their WordStar equivalents in my muscle memory. These days, I have absolutely no memory of what they were.

I was on Usenet fairly often for about the next ten years; even after the Atari ST died and bought my first IBM PC, I was still quite involved—first in the ReXX and OS/2 communities, and then, before long, in the Perl and Linux communities.

I remember seeing dozens of copies of the “Green Card” spam, which is generally considered to be the first commercial spam, in 1994.

And then in 1999, I moved, got DSL, and no longer had easy access to any sort of Usenet server—I had been using the one provided by my dial-up ISP in Miami, but Verizon had no such setup. Sure, there were public NNTP servers of certain limited sorts, but the fact was it was becoming a less and less interesting place to be. So I stopped.

I’m sad to say that I think Usenet dying was always in the cards—I don’t see how it couldn’t have succumbed to the tragedy of the commons. There’s simply too many people for whom the value of the community cannot—especially when they’re not part of it—outweigh some potential personal benefit.

The problem, though, is that the alternative seems to be the commercial walled-garden model, so that the owner can fund policing it, and I don’t find that compelling—because to get any penetration, the service has to be “free”, but as they say, if you’re not paying for the product, you are the product.

While there have been several attempts to create federated systems that allow for personal, decentralized discussions and the like, none have really succeeded. And that’s sad. There is so much amazing stuff in the world, but it can be surprisingly hard to find.