I suppose there is always archive.org and wikipedia, but the more we rely on those to keep old content around the more worried I get about single points of failure. I remember when I first went online in the 90’s the vision of the internet was decentralized content, where information would never go away once it was pushed online. The reality is that the internet is now a decentralized network of central depots, with information rarely jumping between those central depots, and with old layers of content shedding off the edge of those depots and falling into the abyss.
Gatekeeping is essential to the modern web. Content is valuable and/or dangerous and therefore (b)locked away instead of freely copied. Some exceptions exist, of course, but we are very far from that 90’s vision.
The fundamental problem with the Web is that archival was never part of its job. You send HTTP queries to the server and it answers. If somebody switches that server off, all that information is gone. Even just updating the server will inevitably destroy the content. You can try to preserve it with well named URLs, mirror it and such, but it's a lossy process. There was never a mechanism to keep URLs working long term and even if an URL still works, you can't tell if the content is still the same that was there back when the Link was created.
As bad as the modern centralization is in a lot of other ways, I actually quite like the way Youtube works. Every video gets a unique id, a permanent URL and is immutable (with some exceptions). That means most videos from the very early days of Youtube are still around and accessible, while most of the Web of that time is long gone.
The Web really could need a more robust way to publish content, just throwing stuff on a server really isn't cutting it.
That said, substantivization of adjectives is common. Even adjective is itself a word that was originally an adjective, but has been substantivized into a noun. It's a well-known linguistic process:
"Conversions from adjectives to nouns and vice versa are both very common and unnotable in English"
This process has become especially common in Modern English as inflexional endings have been abandoned and speakers rely almost entirely on word order to deduce parts of speech. Converting a word from one part of speech to another simply involves putting it into the proper place in the sequence. However, substantivization of adjectives is not exclusive to English (in Latin, for example, you can slap practically any adjective into the neuter and use it as a noun, whence we receive such words as animal and dual).
...as long as someone finds it worth paying the costs of continuing to run the blockchain, and the people who run 51% of it don’t decide to wipe out or change anything, and...
Remember the slogan "information wants to be free"? I haven't heard that one in a while. Seems like once the Internet was profitable enough, everyone wanted their slice of the pie after all. Happens to every generation, I suppose.
Deep thinkers were always aware of these tensions. Cyberpunk writers in the 90s and even 80s correctly predicted a lot of what is going on right now, namely the use of technology to cover reality in gaudy advertisements (that could look back at you), as well as corporations gaining powers that used to be associated with nation states (for example, policing speech). People just hoped it would not turn out like this... I still have a feeling that it didn't have to.
I sometimes wonder how much those things were “predicted” and how much it was a case of the people with lots of money having read the same books we did and being primed toward those outcomes.
I mean, we all read Snow Crash when we were teenagers and now the ones who grew up to be billionaires are all excited to build the Metaverse. Did Snow Crash “predict” that? Or is this just another (perhaps somewhat extreme) form of fandom?
Similarly, would there be as many scientists trying to figure out how to make a “warp” drive if Star Trek hadn’t popularised the concept? And if/when they eventually succeed in making warp drives practical, will we say that Star Trek “predicted” it?
It feels like there should be a better term for “making a fictional concept compelling enough that it is forced into reality by fans of the fictional work.” But I don’t know what that term would be.
You're going to have a hard time pointing to utopian novels from the same period.
It's not that something closer to utopia was never possible. A lot of CS from the 70s was explicitly utopian. Even the Jobs "bicycle for the mind" idea was far more utopian than "Let's create a monopolistic monoculture with adtech and noise."
For some reason the cyberpunk writers chose not to imagine it or promote it.
Warp drive is similar. It seeded the idea of FTL in the popular consciousness and made it something almost everyone has heard of. Without that, it might have remained an abstract curiosity.
In Propaganda, Bernays says that instead of preaching at the public you need to dramatise the behaviours and beliefs you want to see the public adopt. The ad industry is based on this, but of course it runs through fiction and other media too.
Star Trek is one of the few attempts to dramatise a utopian future. Everything else is a wasteland of darkness.
The notion that scientists wouldn't try to make cutting edge drives without seeing Star Trek is absurd. Similarly, VR and other Metaverse-related concepts are a somewhat natural thing to try once tech advances.
Thinking this requires some heavy anti-multiple discovery beliefs.
Oo, yeah; "inspired" is precisely the word I was looking for! (I'm a little embarrassed I couldn't draw it to mind by myself!)
I'm much happier with saying (for example) "Star Trek inspired the modern warp drive" (I mean, if/when a real-world practical warp drive exists) than "Star Trek predicted the modern warp drive".
Doesn't imply that Star Trek actually invented the thing, but does assign at least a certain causal link to it; not mere prognostication of it like "predicted" implies.
We're cheering and/or allowing stepping stones moving us in those directions, without realizing the long-term implications for the web. (Controversial list incoming) Things like: HTTPS everywhere, SSL, DOH, certificate pinning, trusted-computing, DRM, CDNs, proxies have disappeared, single HTML rendering engine, subscription (SAAS) models, moving all desktop programs into the browser, extinction of caching, etc.
I have lost much of the content I created and published in the early 90s and 2000s, and mourn it in a way because my memory is failing and I wish to learn what has changed in my own life. But you're right.
The one piece that I really miss, is a detailed walkthrough of some really esoteric XSLT code. It was a long, well-demonstrated piece that I published on WebmasterWorld (I think they are still around).
It was done around twenty years ago (no less than eighteen).
It’s absolutely worthless (XSLT -shudder), but I was quite proud of it.
Despite how ugly xslt may look, and how little it's used, I think it actually solved a problem in parsing xml on streams (xslt 3.0 I think?) in a way that really isn't that simple (or used often) in any standard languages.
I have done quite a few xml parsing scripts in python/Julia/etc, but despite xpaths on dom being so much easier to write, the OOM problems are abundant.
XSLT I think had some good ideas for making xml parsing more performant while avoiding some of the major issues that crop up when processing huge amounts of data.
>I remember when I first went online in the 90’s the vision of the internet was decentralized content, where information would never go away once it was pushed online.
This is incredibly bizarre, I can't imagine where this "vision" came from. Storage was extremely limited and bandwidth was expensive back then; things were being deleted constantly. The first web forum I participated in (in 1996) only had the last week's worth of messages available, the older ones were permanently deleted. Back then if you were using Yahoo Mail or Hotmail you had to constantly delete your old emails because you'd run out of space. I would download stuff and then run out of hard drive space, so it would just be deleted and lost forever - nobody could afford more hard drive space.
Gatekeeping is essential to the modern web. Content is valuable and/or dangerous and therefore (b)locked away instead of freely copied. Some exceptions exist, of course, but we are very far from that 90’s vision.