The Fathers of the Internet Urge Today’s Software Engineers to Reinvent the Web

re silo effect

Schneier calls this the Feudal Model of Security or Convenience with nice write-up here:

https://www.schneier.com/blog/archives/2012/12/feudal_sec.ht...

We can also look at it as a form of lock-in. In any case, recent discussions on Elsevier and other scientific publishers shed light on where it might take us. Many academics gripe about not knowing the state-of-the-art or even prior work in their field since they can't afford access to the silos its stored in. Many, despite being customers of Elsevier et al, rushed to download all kinds of stuff from Sci-Hub when it appeared. Now lets imagine that effect applied to most knowledge or content to see how bad it could be for progress of both knowledge and society. Let's, if not paywalled, think of how restricted search and selective promoting can create similar effects by preventing people from connecting dots or even experiencing new things. Then, we see that the siloing could have tremendous, negative impact on people in many ways. Better to switch to something similar to old web where all kinds of content appeared, was easily accessible, and easy to build on.

re trading privacy for free stuff is a myth

It's actually a reality given users dumped their freedom, privacy, and paid offerings in mass for ad-supported, web content/services. The demand side of this was so strong and so many experimental alternatives failed that providers were largely pushed in the direction of ad-support just to survive. It also came with significant, financial rewards. Good write-up here:

http://www.theatlantic.com/technology/archive/2014/08/advert...

So, he needs to quit pretending people are ignoring some solution that works in favor of ad-supported, free-for-users content in a target market that almost exclusively goes with ad-supported offerings. The rational choice is to do what works in a market or with given demand. If they want privacy, they can pay for it or take steps to get it. It's why I have a paid, MyKolab account w/ GPG keyring. Many others used Fastmail or Lavabit for years. Yet, vast majority uses surveillance platforms (eg Gmail, Yahoo, Microsoft) that sell them out to advertisers but also reliably handle the email on the side. I can't remember the market share but I put money that it massively contradicts those arguing against ad model in terms of what people actually do versus what they say.

re sites blinking on and off. Big problem. Needs to be eliminated in next architecture or at least Wayback Machine-style thing with greater integration/convienience. Think snapshots or rollbacks at the browser level.

re sketchy privacy controls. User's fault. They didn't care in practice. They do business with scumbags whose whole model is selling them out and who have a string of abuses. Most won't pay even $2 for private messaging app or $5/mo for private email. Yet, they gripe about privacy issues. I say stick with self-selection plus reboot a simpler, effective model for evaluation of product/service privacy or security along lines of Common Criteria. Security experts, esp experienced in realities of fielded programs, would contribute to it from many different countries to reduce risk of subversion or simply unworkable ideas. Baseline of features & assurance activities critical to privacy and security of product or service plus independent review they're implemented & trusted distribution. Nothing more unless company volunteers as differentiator.

re Vint Cerf. Good ideas across the board with products/services actively attempting to deliver all of them except copyright. That one isn't legal yet, though. The pub-subscribe is a decent idea given there's many robust implementations, even high-assurance schemes, for that sort of thing. Even military is deploying something like that now with at least one high-security demonstrator (below). Commercial/FOSS sector has things like ZeroMQ, which has other benefits. Much field experience out there in doing it right. The older & more field-proven something is, the more likely it will work right the next time. Tried and true beats novel and new.

http://www.dtic.mil/get-tr-doc/pdf?AD=ADA425566

Note: DTIC is another source of wisdom in terms of old papers with great ideads or implementations in them if you know how to find them. Can't help there but keep the DTIC link to anything you find that doesn't have a steady link elsewhere. DTIC link usually stays available longer than average website. CiteseerX and obviously archive.org as well.

re Lee. His idea on URL has been implemented many times over. Just doesn't get acceptance due to bootstrapping problem where all the web browsers have to support the alternative but they're not adding something with little demand most of the time. Dot-archive is nice and could integrate with archive.org. Might even do it with a small fee that simultaneously supports archive.org (or its replacement) plus gives clean link in return similar to subdomains or shortcut links. "Surface the data" is idea behind Semantic Web. It was largely a failure. Market went with API's instead. They're probably better but mixing two might create interesting hybrids.

re Kahle. Decentralized clouds like Amazon definitely worth imitating. Google applied principle to RDBMS's nicely with F1 RDBMS. Awesome stuff. JavaScript will be necessary evil due to market share, ASM.JS, and so on. However, still room for another Flash to happen across a significant chunk of market if done well enough. Don't think of blockchains as about every goal we've listed has been solved in isolation & sometimes decentralized without them. Its inefficent alternative. Now, Merkle or hash trees will likely be useful at some point. Keybase.io & others working on public key angle. That "Wordpress" and "Wordpress alternative" are typed into Google many times a week make his last point solid. Even Freenet and I2P support forms of blogging.

re Doctorow. Vulnerability research being legal is a must. "Computer obeys owners" is a good principle but lay owners vs technical attackers make that a weakness. Feudal model gives up control for safety with good results on Apple, etc. So, maybe an override the user can activate locally or maybe physically. I'm still a fan of jumpers or physical switches for write-protect of critical storage. :)

So, that's my take on these statements.

If we're talking about the ideal world that may be true, but the real world is far from perfect. If you ask 100 people do you want to live a meaningful life, probably 99 of them will say yes. But do they? Probably 1 out of those 100 will live a meaningful life. If you ask people do you want to be told what to do, or do you want to do what you want, most of them will say they want to do what they want. But in reality most people just want to be told what to do because making decisions and being responsible for it is not an easy thing.

Coming back to Facebook feed, there's a reason why people keep using Facebook even though many people hate it. Sure it's not ideal but there will never be an ideal world. I think the reason why most "decentralization" advocates never succeed is exactly because they're being too ideal (read naive) about this.

That said I think the pendulum will swing back someday in the future surely, it just won't be by these guys. It will be from some random technology which didn't even aspire to "disrupt" the web.