So users have a problem. Companies that solved it the right way became enormous, because protection from all that crap is something people want, and they want it to be easy to use. In the early days it was anti-spam (Gmail). More recently, Snapchat helped teens avoid creating a permanent record of things.
Sure, you can go outside the walls at any time, but how many people really want to? The way to win users is to provide better protection than before, including defenses against new privacy threats, not to somehow convince people they don't need to care about it.
Sometimes providing better protection results in more freedom (of a sort). For example, a well-maintained app store means users don't have to worry about malware and viruses, so they can download whatever games they want without worrying that they'll infect their machine. This wasn't true of Windows in the old days.
Ironically, the web started out pretty safe (or so we thought) and degraded as more adtech came along. These days you can hardly click on a link without getting ads in your face. There's opportunity for someone.
It's almost like some kind of weird digital feudalism. You're free to come and go, and if you want, you can live in the wild and take your chances. But if you want the safety of the walls and armed defenders, that comes with a lot of strings attached. And if you ever want to move out of those walls, you leave your plot and your house, with everything valuable on them, behind.
But there is always going to be some kind of a speedbump when sharing between places with different policies. When you have borders between things that are genuinely different (even if they're gratuitous differences), you have border issues. Bigger social networks mean you reach more people without going over a border, which encourages growth.
Borders are everywhere. In the enterprise space, you can look at the growth of Amazon and Google Cloud. In devtools, each programming language tends to be its own domain. (Which is why we call them "foreign" functions.)
I'll grant you the first part, but sharing is actually a lot harder than it used to be, these days. Try sharing a Facebook post or comment with someone who doesn't have a Facebook account...
> In devtools, each programming language tends to be its own domain. (Which is why we call them "foreign" functions.)
True, but we also develop FFI as a way to call those foreign functions, and meta-frameworks to enable cross-language interop (WinRT being the most recent prominent example of such). Worst case, you can use standard formats and protocols to at least exchange data. Whereas "walled gardens" tend to be deliberately hard to interop with, or even to extract your existing data from.
Why isn't Snapchat more popular than Instagram? I assumed that it was the case that more people have stage fright than don't. Is this a faulty assumption, or are people just acting in defiance of this fear?
Full disclosure: I bought a bunch of snapchat stock at the bottom because I fully invest in the self-centeredness of humankind. Worked great w/ FB.
The UX diverges from common design principles of both major mobile OS, making it feel awkward. The app slaughters batteries in Android phones, occasionally you can actually watch the % drop.
That's not to say it isn't popular, wildly so in younger users, Instagram just has more ad value because of its maturity (reportable metrics) and unbridled desire to be an ad platform.
It was there before Eternal September. And yes, it's still there, and far richer than it's ever been.
> The problem is that it got too hostile for most users due to spam, viruses, malware, phishing, trolls, doxing, and so on.
It's arguably those mainstream users who are problematic. In large part, they're the trolls. And the ones who are trolled, of course. And they're the bloody flesh in the water that's attracted all the sharks.
> It will get worse, as attackers adopt machine learning techniques and automate social engineering at scale.
No doubt.
> Sure, you can go outside the walls at any time, but how many people really want to?
Anyone with any sense will. The rest can have their AOL/CompuServe experience. And good riddance.
That sounds a bit ignorant. Those mainstream users are the vast majority, they make a lot of internet business models worthwhile and made the internet the success that it is. Of course they also attract sharks as you rightfully say, but we should try to improve their experiences instead of denying them any sense or intelligence and putting them in walled gardens.
I think you need walled gardens for mainstream users. The reason is that the weakest link in any security system is people. Especially normal, pleasant, trusting, non-confrontational people. People like my grandparents, who just install whatever a guy who claims to be tech support tells them to install.
Should people like my grandparents have a good experience on the internet? Absolutely. Can they have it without taking away all their freedom to make mistakes? I don't see how. Any freedom we give them will be exploited by the sharks. Education is a good thing which can make them more shark-resistant, but not everyone will take it upon themselves to be educated.
Many idealists considered that a ridiculous idea.
> Incredibly so. But I have realized that the large majority of Internet users today have no point of reference to that time because they were either not alive, where very small children, or weren't paying any attention to what was going on. But for a brief time, humanity was being uplifted instead of being loaded onto digital cattle cars. There truly is a threat here--technocracy is turning America into a gulag and it needs to stop. No, I am not kidding. It's not just the Internet, it's all the companies collecting massive data "to better serve us" that are creating the . chains.
> I'm just glad I experienced that early period. Sunsite at UNC was the greatest. There were many true humanitarians back then. Today, we have the growing evil of Google and the Never-Was-Anything-But-Evil Facebook!
Yeah, it's been very sad to see possibility crushed, and people pervasively surveilled and manipulated.
I would guess that today's distributed web (IPFS, etc.) has more content, and users, than the "wild web" of 1992 or whatever era you fondly recollect.
Yes, centralized players created tools that produced exponentially more content (and control). But it's lazy to blame that peripheral phenomenon for a perceived "loss".
The wild web still exists, and has the potential to grow (cf. Urbit). What's stopping it?
There's a lot of promise behind the "distributed web" or "P2P web" prototypes, but they haven't been effectively commoditized, and their value proposition over the top-level "spyweb" isn't strong enough for normal users who don't understand the inner workings. In fact, they tend to offer less end-user benefit, since a P2P web is necessarily slower as peers negotiate connections, deal with vaporizing peers, etc.
Consumers care only about getting their desires fulfilled as quickly and easily as possible. If you want a new web to dominate, you must make something that gives the consumers and/or the companies upstream from them an incentive to exert the effort to change.
Personally, I believe the web is already sufficiently "decentralized" in the necessary sense. We just have legal restrictions that prevent true competition with the tech conglomerates, as they're allowed to effectively claim exclusive domain over all user-generated content.
There's no reason I shouldn't be able to go to NewGoog or NewFace and multiplex in the available data streams hosted by the web nodes at google.com and facebook.com, except that this violates the CFAA and the Copyright Act, and Google and Facebook can and will sue you to make you stop. This is the primary constraint that has led to the recentralization of the web.
What do you expect when you make it a crime to send any type of undesired communication to a server? The CFAA doesn't just prevent harmful or disruptive access, but all unauthorized access. In practice, this is almost always used to stymie small innovators who are putting the data from a data source like Facebook to better use.
This is fundamentally anti-competitive, and again, it's no wonder that the web is coalescing into the hands of a few central controllers. It has never been more apparent that the user is the product being sold and peddled.
We need to reorient our perspective of the internet with users, not services, as the central entity, and give them full control over the data that they generate. Then, the user will be free again, without the need for complex P2P solutions.
There are also things like OpenBazaar and a whole host of other applications running on IPFS that are quite useful. Sure, it may not be mainstream, but nothing starts out being mainstream.
What's stopping authoritarian entity from poisoning/redirecting ipfs.io to static page?
(Which is IMHO one of the weaknesses of IPFS, that they haven't made that more prominent/easier, through browser plugins or a dedicated IPFS-browser. Compare to e.g. the publicity boost Dat got from Beaker Browser)
https://chrome.google.com/webstore/detail/ipfs-station/kckhg...
It'll redirect all IPFS URLs to your local daemon.
It has not yet brought about the collapse of civilization, or a descent into despotism.
^ It has also taken place in the US, but to a different extent.
That's quite a short time scale.
Perhaps, but the "reigning in" referenced in the title is now being done by commercial interests at the request of governments in exchange for staying in those government's good graces.
They do the same thing with many (large) companies, the State Department, all military/intelligence branches, etc. It's how you get access, and they seem more willing to play ball and churn out these puff pieces over doing real journalism than anyone. The Times is the most prominent example of this pattern, but it's becoming more and pervasive.
edit: I mistakenly put Wash Times. Yehowa.
Which might- if freedom is allowed to fluctuate- may result in competition. Look at chinas weechat
This is like a wild meadow turning to a manicured lawn. The near-monoculture of the web will have a much harder time withstanding legal assault by state actors than a distributed web would have.