The future of the web, which will not be

    What will the web be like in the future? They write a lot about this - starting from the new specifications for HTTP, which allow faster and more secure data transfer over the network, ending with the ideas of a “semantic” web . This is all wonderful, the web needs fast networks and organized data, but there are problems that remain underestimated and unsolvable with the technologies used. Let's collect complaints about the current implementation?

    The information on the net is mortal.

    Any service on the network may be closed, and all information accumulated by the service will be lost. Yes, some services (hello Wikipedia!) Make it possible to download a full dump of your data, and the size of this dump is not large enough, but raising a mirror and restoring a service is still a difficult task. Not only for technical reasons, but also for material ones.
    ★ The more popular the service, the more money you need for hosting.
    This is usually not a problem for the commercial development of a resource. More visitors - more money from advertising and other services. But to support the popular nonprofit international service, this could be a major blow. Donat and crowdfunding can save large well-known services, but small ones still have to hang annoying ads to at least pay for hosting.
    ★ The web is doomed to be commercial.
    It's hard to say if this is bad. Commercial service primarily follows the interest of the owner, not the community. All information on the service belongs to the owner, not the community. On the one hand, this allows you to develop the service more intensively and more directionally, on the other hand, this means that if there is no money in the information, then it will not develop. And the competitors who have more money will get an advantage.
    As we see, if the service is closed, it will be a problem to reproduce it. But what are the risks of closing the service?
    • Commercial risk (the project did not reach self-financing)
    • Commercial censorship (content is blocked by court decision on violation of property / intellectual rights / patents)
    • Commercial self-censorship (some content is not beneficial to the owner of the service)
    • Political censorship (the authorities do not like your point of view, which does not coincide with the official one)
    • Public censorship (any community does not like your attitude to any religion, acceptance / rejection of sexual minorities, attitude to weapons and means of self-defense)

    Probably, the list can be continued (write in comments, what other information threats do you see?), But obviously the risks are great. Eternal services do not exist now.

    You do not own the created content

    Content is brought into the network by millions, if not billions of people. These are articles, messages and comments in forums, videos, music and other files. Almost every user creates some kind of content - necessary or unnecessary.
    But who owns this content? Anyone who can remove it is the owner of the site. All your files in the clouds do not belong to you, the good uncle gives you the opportunity to store them there, but if the uncle’s mood or his financial capabilities change, everything may disappear. This would not seem like news, but so far “the account has been blocked on the github” or “the ubuntuone cloud has closed” or “orkut everything” is perceived as something unusual.
    What happens when Wikipedia closes? After all, the kind eyes of its owner will not always be able to bring so much money? And what will happen if Lurka closes? I will miss their content, and I imagine it will be a shame to authors who have spent years creating, editing and maintaining content.

    The network knows everything about you

    A couple of years ago, I made a small who-is-tracking-me service, which allowed me to look at the loaded history of which services transmit information about your visits on the Internet. More than 80% of sites have all possible analytics and advertising systems that can collect complete information about you. Log in to your page on the social network? Well, now it’s known what your name is supposedly. Have you left a pizza order? It is known where you live or work. They turned on Yandex-cards so that they could get home without traffic jams? Now your route is known (by the way, did you notice that the number of your daily routes is not so great, and can you predict where you are going at a certain time?).
    On the one hand, many will say “we have nothing to hide” and will be right. My life is too simple and banal, for someone such information would provide value. On the other hand, all this information in the aggregate for all users individually and in real time is just the same treasure for marketers. This information is also a treasure for agencies such as the NSA, the CIA, the FSB to identify “suspicious” citizens whose point of view differs from the generally accepted one. I’m not talking about groups committing any illegal physical actions, this must be stopped, but people should have freedom of information and freedom of thought. Now it all comes down to the Ministry of Truth with explicit or implicit censorship.

    Exit? Technically, there is

    How do I see the future of the web? Each site is a web application. HTML / CSS / JS, image statics + manifest wrapped in zip, signed with the author’s key. And a distributed document-oriented database that will work in the background in the browser and provide data to the web application. On the interface, this may look like MongoDB, so we will use its terms. Each site has its own base, identified by the guid of the site. The site collection structure is fixed and signed by the site author. One of the collections should be an ACL, with a hierarchy of rights. All content-generating users will be listed in this collection. Each document in the database must be signed by the author of the document.
    Obviously, such a database should not trust anyone from its neighbors, because all the logic for checking write permissions should be written in the database itself. A document has arrived, we verify that the id of the user who signed the document is in the ACL, and it has the necessary rights. To ensure greater user independence, each user should automatically generate an id and key when entering a new site.
    One of the difficult problems is the versioning of documents and their consistency. It is possible that two users will see different information at the same time. But for such a system, I consider normal “someday” consistency, which means that the system will try to synchronize the data and there will be a point in time when this data will coincide with two randomly taken users of the system. Obviously, this approach is unacceptable, for example, when ordering tickets, but action-systems for which synchronization is important should be connected as restful-services. You should share content that will survive a full update after a while and action services.
    For large documents, video, audio or other large documents, it is advisable not to use gridfs in a distributed database, but to use magnet links. In the business rules for the collection for user-downloaded content, you need to specify a limit on the size of the document (for example, the user's avatar). Torrent is a good transport layer for file transfer.
    Data obsolescence. Data will be saved as long as at least one user is interested in it. After the last seed disappears, the data dies. They can be resurrected if such a seed returns. Obviously, the data in the database can grow and swell without measure. A good tool in my opinion is the principle of “phoenix”. Since all the data is publicly available (including both the database data and the front-end), at some point a squeeze out of the actual data is made and users switch to the new version of the site. Those who wish can stay on the old version, but when the last of them stops using it, it will disappear. Any user can fork a site with its data. This is the second side of complete information freedom.
    The description, of course, is incomplete, many nuances still need to be taken into account, but so far I do not see insoluble problems. I will be glad to hear your opinion about the problems that you see in such an implementation.

    Technically, there is a solution, you can create a system that will solve all of the above problems. But the problem is that nobody needs such a system. For the most part, everyone is happy with everything. There is no urgent demand in society to solve the problems above, and the new system will interfere with government and business. Therefore, this is the future of the Web, which most likely will not be. I had an idea to deploy a crowdfunding company to create such a system, it would be possible to keep within 5-10 million, but does society need it?

    Also popular now: