Regarding "No one ever takes a photograph of something they want to forget" - I take photos of things all the time of things precisely so I can forget them - I take pictures of gates closed on farmland, electrical items switched off etc. I never look at these pictures again - but the simple act of taking the picture means I don't fret about them afterwards....
Many of the pictures I take have a lifetime, sometimes measured in hours or minutes. A restaurant I passed by before dinner that I'd like to check out later in the evening; a sign in a language I want to translate; a visual demonstration of where I am so people know where to find me in a busy train station; any number of things whose worth is almost immediately used up and subsequently deleted.
When you try to recall some past moments, do you recall the moment as you saw it with your eyes or do you recall the photo of it?
I have noticed this lately that when I try to recall (e.g. a moment or time with my family) all I get back is one of the photos of it. I don't usually take photos, when I do I try not to look at them for the exact same reason as they seem to replace the actual snapshot of the moment I saw with my own damn eyes.
But I recall a lot less snapshots of what I saw with my own eyes compared to what I can recall in terms of photos.
We take photos of our kids to track and timestamp anaphylaxis or other medical reasons. Incredibly important at the time, and then when dynamic wallpapers start mixing them in to the roster you desperately want to forget them.
I'm the same. I offload stuff to my phone's camera. Not really "offload" because I barely load it into my brain, just photograph it and don't worry about it. It has saved my mass quite a few times.
This seems to mostly be about Xanadu, which I've heard of in passing before but never really looked into. Some of the features sound nice (version control, content-addressing), some are confusing or undesirable (some mechanism to avoid sharing copyright-protected content?, backlinking, micropayments?). The screenshots of what I presume are Xanadu browsing in action look more like schizophrenic corkboarding than a useful document format, with the emphasis placed more on tracing the provenance of summarized or quoted text than on presenting something readable. Perhaps this style comes more naturally to one more familiar with working in academia and reading/writing papers in that sphere. Being a document distribution system, I don't think Xanadu competes with or is even in the same category as The Web, which as we all know, is an application distribution platform and runtime first and foremost.
Ah, good old Ted Nelson. His "Computer Lib/Dream Machines" captivated and inspired me as a counterculture kid in the '70s, and played a big part in me becoming a graphics nerd today. (https://en.wikipedia.org/wiki/Computer_Lib/Dream_Machines)
Worth noting that this appears to be pretty old. The code follows 90s conventions, it mentions things being nascent that are now quite established (like Wikipedia), and other little yellow flags.
Also, geegaws?
If I had to guess, I'd say this is twenty years old.
- it's not personal, yes we can save much contents, but it's not that manageable nor can be evolved much easier personally, we need a textual web, where anyone can grab text, store it locally naturally, change it as they wish. The "app" part should be just APIs and UIs;
- it's not immediate for anyone, yes we can buy a domain name, it's pretty cheap in most cases, we can use a ready made static website generator, but we do miss the most important part: IPv6 with a static global per host, so we can made OUR OWN WEB with our own friends;
- it's isolated, we have the modern web and WebVMs improperly named browsers for legacy reason, then we have desktops that act more like WebVM bootloaders than classic connected desktops, than some local apps, that tend to be limited and limiting. We need INTERNET much more than the web.
We need RSS for anything, with modern feedreader and a spread culture that content aggregation is a personal thing. We need easy to import data, pure natural language text and numerical data, so for instance if we go to our Amazon profile we can see a feed of all our purchase we can import in our reader, piping from it to our favorite financial software that also import other feeds from our banks/cards etc to make automation easy for things we can automate. We need data ownership, local iron in a classic desktop model.
May I suggest that this quite interesting content will confuse readers that do no come with some context
IMHO, It would be much more approachable if it started with the section "What is wrong with the World Wide Web?" ?
Also, you are casting a very wide net, is this section really necessary, and I quote:
"""
Java - versatile, cross-platform, but difficult to learn. It hasn't lived up to all the hype that originally surrounded it, but is still important for many things.
"""
1. a DOM system with commit history and git blame per node.
2. link preview like on Wikipedia
I have thought about writing a DOM alternative in rust. The biggest change is that it would return static data structures in a format of a users choice (yaml, json, xml, csv) and not JavaScript objects. Another thing I would do is eliminate most node types so there are only: document, element, element property, and text except text nodes would not return unless each given area of text contains more than 0 non white space characters and the list of white space characters would be expansive similar to the space switch in JavaScript regular expressions.
I remember listening to a pitch about Xanadu around 2010 or 2011.
One of the major, major issues that the presenter (don't remember the name) had is that features like two-way linking only work if the entire system is centralized. Yet, to "replace" the world-wide-web, the system has to be decentralized, or otherwise encumbered with a lot of the problems of blockchain. (IE, blockchain protocols require significantly more politics to update than HTTP.)
I very quickly stopped taking the presenter seriously; and I think other people had a similar sentiment.
Seems misguided. The real problem with the internet is that large corporations are allowed to have a presence on it. Google for example, should be completely banned from the internet.
A good question. Perhaps a rule like this: the top 10 tech companies should be put to a worldwide referendum on whether they should be allowed to exist. If "no" is the majority, they should be dismantled into smaller companies or turned into public utilities.
Not the op, but there are several that jump to mind:
- A Google-branded network separate from the internet
- Browsing the cache
- Since this hypothetical reality includes them being restricted from the internet, maybe they don’t need to link at all and they just give the information instead of the link to the page it’s on
Or a more modest attempt: how about keeping the assets but breaking up the large corporations? With Android on iPhone, Chrome blocking Google's tracking, Amazon.com hosted on Azure, MS 365 and Google docs competing on merits rather than sheltering under respective coorporate umbrellas....
We'd still have much of the same crap, but the ecosystem would be much more dynamic and interesting.
Breaking up how? Splitting MS into Office and Windows creates two (functional) monopolies. Split it into Windows, Excel, Word, Powerpoint creates 4 monopolies.
Imo the only split that makes sense is to split it down the middle: TWO companies which BOTH have the rights to Excel. Half the engineers in each.
That is an interesting approach: Excel "Community Edition" and Excel "Old Microsoft Edition", and have them compete. Also, perhaps: after 20 years, software must be open-sourced.
> Also, perhaps: after 20 years, software must be open-sourced.
Wow! Wouldn't that be great?! Like as in an international constitutional kind of ground rule saying any software must ultimately become source-available after X years and becomes public domain after Y years.
(Still plenty of wiggle room for monopolistic closed source, but at least it wouldn't be that way forever.)
I was saying two competing closed source companies. Markets work best when they emulate nature: speciation, selection, competition. Successful companies should be split with all IP shared by the two daughter corporations.
I'm old enough to remember when Google was good. I don't think this is a scenario you have thought through.
Now, the problem with social networks is that they all are going for maximum market share. Which is the same as going for toxic and divisive communities. The collapse of Usenet in the Eternal September was just the opening salvo. It has happened to Facebook, it happened to Twitter, it has already happened to Threads (that place is worse than twitter now by far!), and it happened to many other less known things in between.
Any IRC channel above a certain size becomes toxic or useless or both, any Discord. Any group of people in the same room all screaming.
The solution isn't to ban big corporations. The solution is to build smaller "cells" with limited "immigration".
Well, if large corporations are banned, then large social networks could not exist easily since no one would have the resources to run them. So, I still stand by destroying large tech corporations.
Yea, but Google really ushered in an era of making the internet MUCH more useful for decades. Then they destroyed it all, but for a while it was great.
When was this written? Is it still being maintained? The page history at the Wayback Machine [1] suggests that it has been mostly static for well more than a decade.
I've noticed a lot lately, that sites, projects, even big projects, don't have an easy to find "TL;DR", an "elevator pitch", right at the top of the page.
Most of the time I scroll, it looks like something that might be interesting, but it takes too much effort to figure out, so move on. There is so much content on the web now, these efforts need to put little bit more into the marketing if they want attention.
I'm sure scrolling down maybe it is discernable, but not really obvious.
Or maybe this is just so high concept that a short pitch can't explain it?
This is really hard to follow.. It gives off some TempleOS vibes.. I agree in principle the WWW has some significant shortcomings and there could be some better way to think about it. It will require some outside the box thinking but this is really hard to follow cognitively.
> What Xanadu has that others don't:
> When responding to any argument (for or against something), you will be able to comment on each and every relevant sentence. The document you are commenting on will automatically link to your analysis. Whether readers of the document see your comments or not is up to them. They will have filtering tools to include or exclude whoever they want. That is their decision to make. Thus every possible viewpoint can be heard and can cross-comment on each other as much as desired.
This could be an extension of Twitters behavior, to link your analysis directly into other peoples document.. And we all know what great social collaboration happens on Twitter ..
Ted Nelson's critiques are good. Tech/Culture needs critics. Ted also tried being a culture-tech builder, and Xanadu just didn't work out. Despite several attempts, lots of resources spent...
Its ideas probably influenced Twitter. Because they were good ideas. Could've been independently discovered in the Platonic space of ideal tech-powered media too...
Yes, great social collaboration happens on Twitter. A bunch of inane squabbling and criminal threats too.
> This could be an extension of Twitters behavior, to link your analysis directly into other peoples document.. And we all know what great social collaboration happens on Twitter ..
Surely you're not going to leave us hanging like this, are you? What will happen with a Project Xanadu?
Ted Nelson has a vibe kind of like that. I still like some of his rants and speeches, and he has been quite important in the history of computers, but some of it is plain insufferable. Though not as explicitly offensive and deranged as Terry Davis was sometimes.
Where are you seeing signs of a web replacer? Native apps? Or big social media platforms built on the web, that replace the need (for mose users) to 'touch' the base web layers directly?
It would be sad if those bury the web and bury our rich, free open platform paradigm.
If a good decentralised open thing replaced it instead, cool.
It's like tragedy of the commons. Open platforms are always susceptible to EEE (embrace extend extinguish) by design (would like to be proven wrong). And there will always be businesses (people) who when platform is popular enough would like to take over that for profit.
A commons is indeed vulnerable to collapse though bad management. So it won't take care of itself, but needs deliberate effort and coordination to maintain. If we want to keep the open web, can't we put the necessary work in?
One challenge for web-enjoyers is it's not obvious in a consensus-making way where 'weak points' that need help are, and what's an acceptable loss (say: video on the web. We can post raw video on a webhost. It's expensive, so no one does. And most video content is crap, so who cares - let the platforms have that?)
I disagree that open platforms are always susceptible to EEE. In my mind, their openness (or more precisely, the true openness of their protocol(s)) prevents that.
Sure, there's a lot of DRM crap on the WWW now, but there's also still plenty of plaintext. The mainstream going with the EEE crowd is a given (being money people, they have the motive, means, and opportunity to streamline onboarding in a way the openness true believers don't), but the WWW platform remains fundamentally open despite everything thanks to the openness of the protocols it's built on.
A classic example of EEE closer to a protocol takeover is how Google handled XMPP. Yet people still use XMPP. It's just not huge. But open != huge or mainstream. Usually quite the opposite (see above about money people herding the mainstream to their platforms).
A rather more blatant takeover is the more recent hostile takeover of Freenode IRC. In that case the community revolted against the takeover and relocated (some to another IRC server and IIRC some to Matrix)
I have noticed this lately that when I try to recall (e.g. a moment or time with my family) all I get back is one of the photos of it. I don't usually take photos, when I do I try not to look at them for the exact same reason as they seem to replace the actual snapshot of the moment I saw with my own damn eyes.
But I recall a lot less snapshots of what I saw with my own eyes compared to what I can recall in terms of photos.
"Everything is deeply intertwingled."
Also, geegaws?
If I had to guess, I'd say this is twenty years old.
- it's not personal, yes we can save much contents, but it's not that manageable nor can be evolved much easier personally, we need a textual web, where anyone can grab text, store it locally naturally, change it as they wish. The "app" part should be just APIs and UIs;
- it's not immediate for anyone, yes we can buy a domain name, it's pretty cheap in most cases, we can use a ready made static website generator, but we do miss the most important part: IPv6 with a static global per host, so we can made OUR OWN WEB with our own friends;
- it's isolated, we have the modern web and WebVMs improperly named browsers for legacy reason, then we have desktops that act more like WebVM bootloaders than classic connected desktops, than some local apps, that tend to be limited and limiting. We need INTERNET much more than the web.
We need RSS for anything, with modern feedreader and a spread culture that content aggregation is a personal thing. We need easy to import data, pure natural language text and numerical data, so for instance if we go to our Amazon profile we can see a feed of all our purchase we can import in our reader, piping from it to our favorite financial software that also import other feeds from our banks/cards etc to make automation easy for things we can automate. We need data ownership, local iron in a classic desktop model.
Also, you are casting a very wide net, is this section really necessary, and I quote:
""" Java - versatile, cross-platform, but difficult to learn. It hasn't lived up to all the hype that originally surrounded it, but is still important for many things. """
less is more!
1. a DOM system with commit history and git blame per node.
2. link preview like on Wikipedia
I have thought about writing a DOM alternative in rust. The biggest change is that it would return static data structures in a format of a users choice (yaml, json, xml, csv) and not JavaScript objects. Another thing I would do is eliminate most node types so there are only: document, element, element property, and text except text nodes would not return unless each given area of text contains more than 0 non white space characters and the list of white space characters would be expansive similar to the space switch in JavaScript regular expressions.
One of the major, major issues that the presenter (don't remember the name) had is that features like two-way linking only work if the entire system is centralized. Yet, to "replace" the world-wide-web, the system has to be decentralized, or otherwise encumbered with a lot of the problems of blockchain. (IE, blockchain protocols require significantly more politics to update than HTTP.)
I very quickly stopped taking the presenter seriously; and I think other people had a similar sentiment.
- A Google-branded network separate from the internet
- Browsing the cache
- Since this hypothetical reality includes them being restricted from the internet, maybe they don’t need to link at all and they just give the information instead of the link to the page it’s on
We'd still have much of the same crap, but the ecosystem would be much more dynamic and interesting.
Imo the only split that makes sense is to split it down the middle: TWO companies which BOTH have the rights to Excel. Half the engineers in each.
Wow! Wouldn't that be great?! Like as in an international constitutional kind of ground rule saying any software must ultimately become source-available after X years and becomes public domain after Y years.
(Still plenty of wiggle room for monopolistic closed source, but at least it wouldn't be that way forever.)
Now, the problem with social networks is that they all are going for maximum market share. Which is the same as going for toxic and divisive communities. The collapse of Usenet in the Eternal September was just the opening salvo. It has happened to Facebook, it happened to Twitter, it has already happened to Threads (that place is worse than twitter now by far!), and it happened to many other less known things in between.
Any IRC channel above a certain size becomes toxic or useless or both, any Discord. Any group of people in the same room all screaming.
The solution isn't to ban big corporations. The solution is to build smaller "cells" with limited "immigration".
[1] https://web.archive.org/web/changes/http://hyperworlds.org
Most of the time I scroll, it looks like something that might be interesting, but it takes too much effort to figure out, so move on. There is so much content on the web now, these efforts need to put little bit more into the marketing if they want attention.
I'm sure scrolling down maybe it is discernable, but not really obvious.
Or maybe this is just so high concept that a short pitch can't explain it?
So what is this about?
I think it's more than 20 years old.
> What Xanadu has that others don't:
> When responding to any argument (for or against something), you will be able to comment on each and every relevant sentence. The document you are commenting on will automatically link to your analysis. Whether readers of the document see your comments or not is up to them. They will have filtering tools to include or exclude whoever they want. That is their decision to make. Thus every possible viewpoint can be heard and can cross-comment on each other as much as desired.
This could be an extension of Twitters behavior, to link your analysis directly into other peoples document.. And we all know what great social collaboration happens on Twitter ..
Its ideas probably influenced Twitter. Because they were good ideas. Could've been independently discovered in the Platonic space of ideal tech-powered media too...
Yes, great social collaboration happens on Twitter. A bunch of inane squabbling and criminal threats too.
Surely you're not going to leave us hanging like this, are you? What will happen with a Project Xanadu?
https://www.youtube.com/@TheTedNelson
Did it fulfil original goals beyond expectations? Yes.
Is it going to be replace with something else? Looks like it.
Will that be Xanadu? No.
It would be sad if those bury the web and bury our rich, free open platform paradigm.
If a good decentralised open thing replaced it instead, cool.
One challenge for web-enjoyers is it's not obvious in a consensus-making way where 'weak points' that need help are, and what's an acceptable loss (say: video on the web. We can post raw video on a webhost. It's expensive, so no one does. And most video content is crap, so who cares - let the platforms have that?)
Sure, there's a lot of DRM crap on the WWW now, but there's also still plenty of plaintext. The mainstream going with the EEE crowd is a given (being money people, they have the motive, means, and opportunity to streamline onboarding in a way the openness true believers don't), but the WWW platform remains fundamentally open despite everything thanks to the openness of the protocols it's built on.
A classic example of EEE closer to a protocol takeover is how Google handled XMPP. Yet people still use XMPP. It's just not huge. But open != huge or mainstream. Usually quite the opposite (see above about money people herding the mainstream to their platforms).
A rather more blatant takeover is the more recent hostile takeover of Freenode IRC. In that case the community revolted against the takeover and relocated (some to another IRC server and IIRC some to Matrix)