5 ms·

Tim Berners-Lee on 30 years of the web: 'We can get the web we want' (2019)

8bitsrule 6y ago
Important topic, very poorly-written article. It's as If the author wanted to disguise what Tim's ideas are.
- musicale 6y ago
  Like this bit of bizarre nonsense which ignores the existence of HTTPS and conflates communication privacy with walled gardens and user lock-in: "the very idea of accessing core swathes of Facebook’s platform from a web browser was deprioritised, in favour of the extreme privacy provided by universal end-to-end encryption"
rektide 6y ago
to me, the defining sadness is that the past decade has been the death of apis, the end of interoperability. the word web evokes a weaving together, a connectedness, and that, to me, always extended far beyond the link. of course, what everyone is worried about today is too much sharing, too much tracking. we've focused only on the down-sides of connecting, only on the practically malicious surveillance that connectedness has brought the far of & mighty trackers & ad-networks. we've focused on mis-/dis-information & conspiracy theories & a shocking rise of brazen demagoguery connectedness have enabled. to me, we have seen the evils coming out of the box, & sealed it shut, once more trapping hope inside. i still envision the web as a place for small services/preferences/aspects of personality to exist on focused, narrow sites, that these services work & aggregate together across a variety of other sites. new specifications like Web Share show promise, but are hampered by lack of adoption & implementation being limited to narrow sets of platforms. works like Foreign Fetch (RIP) highlighted the real spirit of adventure to me: the idea that a ServiceWorker could offline serve other sites, other domains. even given the new radical http-cache-partitioning security measures in place, would-have-been/could-still-be works like Foreign Fetch continue to highlight to me the real spirit of connectedness & the web. i also see a role for ambient connectivity. Project Eddystone allowed Bluetooth Low Energy (BLE) beacons to advertise short urls, that the user could see about them as they moved. primarily used by stores & spammers, this sort of immediate real world linking, to me, is something people & individuals have the best chance to connect & share their world through: a directly advertised link anyone around them can see. third, last, i really hope the semantic web comes about. i want us to begin to interact with the page as more than a giant canvas, a single entity. the data of the page ought express itself. we ought develop tools to surface & interact with data, in-place, where it is encountered.
- Shared404 6y ago
  It seems to me that the best hope is to teach the average person how computers work. Teach students how to use Linux/BSD, solder, what a web request is, how to run a server, etc. If we do these things, then people can return to setting up their own services, instead of walled gardens.
  - rektide 6y ago
    i believe in teaching & sharing & the power of people. a lot. but at the moment i think the geeks have a ton of do-for-themselves to do as well. teaching all of these things will not arise the better world. there is an inertia here that needs to be overcome by working, by doing. the ActivityPub, & the Solid people are the closest i've seen to those who are trying to build some kind of coherent basis for being & building online systems in an interconnective manner. we need a lot more dedicated & willing folks who want to believe in better worlds, more open experiences. i also think teaching the current systems is a waste of time. kubernetes is far & above better than every other way we've run systems, but there's still a lot of work to do to build happy paths for it, especially for the self-hosted crowd & their needs. some important avenues, like sharing aka multi-tenancy are basically unexplored still. to walk myself back a little, yes, a lot of the skills/knowledges you talk about are pretty fundamental. but there's nothing resembling an overarching paradigm for computing, that works together, & everything is manual stick building. trying to re-envision computing as a utility, one we can run, as clouds, that we ourselves can own, is necessary. without this, there is no base, just an endless path of staring at man pages, always knowing we'll not know >80%, and that there is no real order to it all.
  - TheOtherHobbes 6y ago
    No, the answer is to make these services commodities with security and privacy built-in organised in an open public infrastructure owned by no corporation. Most of the population is simply not interested in Linux, soldering, or what a web request is. Expecting people who can barely put up a bookshelf to manage a web server or configure Linux is nonsensically unrealistic. It's simply not going to happen. Nor should it. That's our job, not their job. What they're interested in are the life benefits that secure, reliable, effective, private commoditised technology can offer them. Providing them is also our job.

Important topic, very poorly-written article. It's as If the author wanted to disguise what Tim's ideas are.

musicale 6y ago
Like this bit of bizarre nonsense which ignores the existence of HTTPS and conflates communication privacy with walled gardens and user lock-in: "the very idea of accessing core swathes of Facebook’s platform from a web browser was deprioritised, in favour of the extreme privacy provided by universal end-to-end encryption"

to me, the defining sadness is that the past decade has been the death of apis, the end of interoperability. the word web evokes a weaving together, a connectedness, and that, to me, always extended far beyond the link. of course, what everyone is worried about today is too much sharing, too much tracking. we've focused only on the down-sides of connecting, only on the practically malicious surveillance that connectedness has brought the far of & mighty trackers & ad-networks. we've focused on mis-/dis-information & conspiracy theories & a shocking rise of brazen demagoguery connectedness have enabled. to me, we have seen the evils coming out of the box, & sealed it shut, once more trapping hope inside. i still envision the web as a place for small services/preferences/aspects of personality to exist on focused, narrow sites, that these services work & aggregate together across a variety of other sites. new specifications like Web Share show promise, but are hampered by lack of adoption & implementation being limited to narrow sets of platforms. works like Foreign Fetch (RIP) highlighted the real spirit of adventure to me: the idea that a ServiceWorker could offline serve other sites, other domains. even given the new radical http-cache-partitioning security measures in place, would-have-been/could-still-be works like Foreign Fetch continue to highlight to me the real spirit of connectedness & the web. i also see a role for ambient connectivity. Project Eddystone allowed Bluetooth Low Energy (BLE) beacons to advertise short urls, that the user could see about them as they moved. primarily used by stores & spammers, this sort of immediate real world linking, to me, is something people & individuals have the best chance to connect & share their world through: a directly advertised link anyone around them can see. third, last, i really hope the semantic web comes about. i want us to begin to interact with the page as more than a giant canvas, a single entity. the data of the page ought express itself. we ought develop tools to surface & interact with data, in-place, where it is encountered.

Shared404 6y ago
It seems to me that the best hope is to teach the average person how computers work. Teach students how to use Linux/BSD, solder, what a web request is, how to run a server, etc. If we do these things, then people can return to setting up their own services, instead of walled gardens.
- rektide 6y ago
  i believe in teaching & sharing & the power of people. a lot. but at the moment i think the geeks have a ton of do-for-themselves to do as well. teaching all of these things will not arise the better world. there is an inertia here that needs to be overcome by working, by doing. the ActivityPub, & the Solid people are the closest i've seen to those who are trying to build some kind of coherent basis for being & building online systems in an interconnective manner. we need a lot more dedicated & willing folks who want to believe in better worlds, more open experiences. i also think teaching the current systems is a waste of time. kubernetes is far & above better than every other way we've run systems, but there's still a lot of work to do to build happy paths for it, especially for the self-hosted crowd & their needs. some important avenues, like sharing aka multi-tenancy are basically unexplored still. to walk myself back a little, yes, a lot of the skills/knowledges you talk about are pretty fundamental. but there's nothing resembling an overarching paradigm for computing, that works together, & everything is manual stick building. trying to re-envision computing as a utility, one we can run, as clouds, that we ourselves can own, is necessary. without this, there is no base, just an endless path of staring at man pages, always knowing we'll not know >80%, and that there is no real order to it all.
- TheOtherHobbes 6y ago
  No, the answer is to make these services commodities with security and privacy built-in organised in an open public infrastructure owned by no corporation. Most of the population is simply not interested in Linux, soldering, or what a web request is. Expecting people who can barely put up a bookshelf to manage a web server or configure Linux is nonsensically unrealistic. It's simply not going to happen. Nor should it. That's our job, not their job. What they're interested in are the life benefits that secure, reliable, effective, private commoditised technology can offer them. Providing them is also our job.