back to article Here comes the blob: Asia's top 'net boffin thinks 'shapeless services' could replace the Internet

What will the internet look like in the year 2071? Geoff Huston, chief scientist the regional internet registry the Asia Pacific Network Information Centre (APNIC), thinks there may not be an internet – or at least not as we know it today. Huston's thinking is outlined in a presentation he made to a recent IBM Research event …

  1. claimed

    I predict

    Email will still be around. Perhaps some service layers will fracture, which sounds reasonable, but addressing globally is damn useful for lots of things.

    1. Anonymous Coward
      Anonymous Coward

      Re: I predict

      Global addressing, plus commonality at the bottom of the stack. There's an awful lot of aggro and thought going into getting 0s and 1s moving about between places, the shapeless services idea is one slice of the internet but not all...

      Reminds me of microservice kernels that were going to take over the world...

    2. Charles 9

      Re: I predict

      Perhaps as consumer-level computing capability continues to improve, more and more stuff accessed via server will be kept at the user's end. E-mail is one possibility: being addressed directly to you instead of to a server you access: with portability and privacy implications.

      1. Doctor Syntax Silver badge

        Re: I predict

        I've always been in the habit of storing emails locally. It's very convenient. It means I've gone through several service providers, migrated from ISP addresses to own domain and switched the MSP for the domain. It still does work on a basis of a server - fairly unavoidable, I'd have thought, with the store and forward nature of email unless I were to run my own server. It means that I have a local archive of more or less anything I need going back years (and a lot of stuff I no longer need but don't have to waste time reviewing to discard).

        My sister-in-law, however, has difficulty grasping the notion that the icon for her email on the desktop is just a link to her MSP, opened by the same application as her Google icon with saved credentials and that by going back to the login screen and entering her husband's ID they could use it to access his email instead of using his mobile. People like that would have problems making the switch.

        1. Martin Gregorie

          Re: I predict

          Same here, except that my email store is a PostgreSQL database.

          There is a good reason for that: I also run SpamAssassin locally. It queries the email database, and whitelists email from any address that I've previously sent mail to. This lets me safely use a set of fairly aggressive antispam rules.

          The final safeguard is that all spam gets quarantined for a week. Each morning I get a daily report of any spam that was quarantined during the previous day so I can easily retrieve anything that was incorrectly treated as spam: this is quite rare and is almost always messages from a new service I've just signed up to offline or over the phone.

          1. Cliffwilliams44 Bronze badge

            Re: I predict

            I don't think you can get away from the store and forward model of email as not all end user devices are going to be "online" 24/7. So there has to be some service that receives and stores the email to await pickup by the client. Then what about browser based email? So many people only use the browser interface to interact with email, or just their phone. I certainly cannot have my gmail account completely downloaded to my phone.

            1. Charles 9

              Re: I predict

              "I don't think you can get away from the store and forward model of email as not all end user devices are going to be "online" 24/7. So there has to be some service that receives and stores the email to await pickup by the client."

              Why not the sender? Cut out the middlemen and it becomes easier to know when or if the recipient is in a position to receive.The sender can just store the message itself and keep trying until it succeeds or until enough time passes that it can come back, "No can do."

              "Then what about browser based email? So many people only use the browser interface to interact with email, or just their phone. I certainly cannot have my gmail account completely downloaded to my phone."

              The e-mail can be a computer owned by the end-user instead. Your e-mail, your business. It doesn't take much to house the average person's e-mail or run a small web-based portal for the purpose. This is what I'm saying. End-user computing power is reaching the point that increasing amounts of stuff that used to be done on servers can be moved to the end-user, much like how peer-to-peer protocols had a boom a while back.

              1. claimed

                Re: I predict

                Pretty sure they thought of that in the 80's... so if the sender is the one who has to monitor all the destinations, and you send to a mailing list of several thousand emails, who are all now offline, your network connectivity is going to go to pot while you wait for your contacts to pop up online for 10 seconds to check their email.

                That is not a scalable or desirable design, which is why email works the way it does.

                1. Charles 9

                  Re: I predict

                  "Pretty sure they thought of that in the 80's..."

                  Pretty sure at the time the idea was that the e-mail was hosted on a corporate server with clients connecting via terminals, thin clients, and so on. It's still done that way on corporate e-mail systems for legal reasons, but in this day and age does the end-user really need a third-party holding (and snooping on) their business?

                  "...and you send to a mailing list of several thousand emails, who are all now offline, your network connectivity is going to go to pot while you wait for your contacts to pop up online for 10 seconds to check their email."

                  Think more about how P2P software like BitTorrent works. The software sits there on the computer all the time while you're doing something else. An end-user e-mail system can do the same thing: monitoring the connections as needed, doing the polling; it's not really all that different and really no sweat for a modern (<10 years old) system. Remote access can be done by phone and so on to the end-user's "home base". The end-user's computer essentially becomes the e-mail server: a personal e-mail server with a few extra details for ease of use and security, but is it really all that different from a BitTorrent endpoint? If BitTorrent can do it with gigabytes of data (AND potentially thousands of peers, so there goes your mailing list argument), why not e-mail?

                  1. Doctor Syntax Silver badge

                    Re: I predict

                    "does the end-user really need a third-party holding (and snooping on) their business?"

                    The snooping should have been dealt with long ago. The existing system should have incorporated a PKI years ago so that encryption could be built into the clients.

                    The equivalent of a torrent would be that you send out the email to lots of other people and the when the intended recipient goes online they pick it up from whoever is online at the same time.

                  2. claimed

                    Re: I predict

                    The main argument is that BitTorrent relies on people wanting to share an identical file, so peers going offline doesn't mean you lose your data. It also uses small chunks of files so peer bandwidth and latency don't slow you down. If I email someone in another timezone, I would have to leave my computer on until they turn theirs on, or we'd all have to agree to share encrypted blobs of other peoples emails, and then sync all the email in the world so we can peer it. So all computers on all the time? It's a crap, crap suggestion.

                    You might enjoy applications that hog resources, but designing in Network and CPU wastage seems sub optimal. Just think of all those SSL auth cycles and the extra electricity your suggestion requires. PGP hasn't caught on, so we're still talking plaintext email or plaintext over SSL.

                    I hope you're not a delegate at COP26!

    3. Blackjack Silver badge

      Re: I predict

      IRC is still around so yeah why not?

      1. sreynolds

        Re: I predict

        Yeah I reckon that USENET will never die. I mean store and forward. Talk about fragmentation of the namespace.

    4. Jamie Jones Silver badge

      Re: I predict

      .... and still people will be CCing sensative distribution lists instead of BCCing!

    5. Justthefacts Silver badge

      Re: I predict

      I predict email definitely *won’t* be.

      A couple of reasons. Firstly, young people don’t use it, only older people like me, plus business users.

      Now let’s look at that corporate use. You already see Teams taking over during pandemic as a wider concept for corporate internal (and client) communications. I’m not saying that will win in the end, but some *proprietary platform* might

      And secondly for corporate working on projects, we all know that email threads are just a lousy way of having (traceable) discussions with half a dozen or more people involved. There’s the social issues of people feeling compelled to comment who don’t know what they are talking about hijacking top-of-thread. And there being only two levels of attention: copy-me-in (why do I have 600 emails a day that I can’t read) vs i-couldn’t-have-known. But there’s more.

      I think there will be a corporate-collaboration-tool app which will replace it. The problem has always been that the companies which could write it, are staffed by the worst email offenders who don’t know what a non-pathological working culture looks like

      1. The Man Who Fell To Earth Silver badge

        Re: I predict

        The only "young people" who don't use email are the unemployed ones. Email has so many advantages over chat in a work place that it's hard to see it going away anytime soon.

        1. logicalextreme

          Re: I predict

          Yeah, I think with a lot of these "young people don't use X" there's an implicit assumption that said cohort of young people will continue to not use X when they're at the age of the people who do currently use X, which often simply isn't true. This may even apply to e.g. Facebook, not sure.

          As for email specifically — it doesn't matter how young you are, the overwhelming majority of retail websites aren't going to let you check out without an email address. I can't see an entire generation refusing to register accounts and using BugMeNot. No, email will survive like the cockroach that it is and people like us will continue having to write horrible code to make it do things it wasn't designed for like HTML, or authenticating the sender.

          1. jason_derp

            Re: I predict

            "X" is usually a service. And often, young people DO use X, just in a different way.

          2. Justthefacts Silver badge

            Re: I predict

            email address as proof of identity?

            I honestly don’t see that lasting fifty years. We will have other, better, forms of online identity before that. Government centralised? Token-based?

            I feel I’m in a weird place here - I can’t tell you what will replace it, because otherwise I would be The Zuck and doing it. But I do know our current solution is a) Crap b) Only 30yrs old for the majority of the population. So it doesn’t feel like it will last twice as long as it has already. The fact that the protocol has been around over fifty years isn’t the point.

            1. Yes Me Silver badge

              Re: I predict

              Identity management systems, in my experience, rely on an email verification step, or an SMS verification step, or both. There's a reason for that: those two mechanisms are hard to roboticise. Identity providers who can't insist on a face to face contact with photo ID use email and/or SMS as the next best thing. Some identity systems also rely on postal mail delivery of a one-time password.

              I haven't seen one using fax, but I expect they exist.

            2. Doctor Syntax Silver badge

              Re: I predict

              "otherwise I would be The Zuck and doing it"

              The Zuck is doing it. Haven't you come across sites where a Facebook ID is offered as an alternative to an email address.

              1. Justthefacts Silver badge

                Re: I predict

                Of course I have. It’s why the FB outage “took out half the internet”, so far as most people are concerned.

                Various password wallets are another stab at solving the problem; both proprietary apps, and within Chrome itself.

                And Microsoft has been making attempts to have SSO for 365 take over this world.

                Plus one you may not be aware of: Shopify are essentially trying the same SSO gig for all the ecom websites they run. If they partnered up with Squarespace & Woocommerce, think what that would be like?

                None of those have won the game yet.

                But I think the fact that there have been *five* somewhat successful attempts to storm the castle, in just ten years, suggests that the gates won’t hold for fifty years.

                It’s like claiming that because so many websites use CAPTCHA, that will still be true in fifty years. Obviously not, but no I don’t know what will replace it.

          3. Justthefacts Silver badge

            Re: I predict

            Well, I *can* tell you about Facebook demographics.

            Yes, there are still loads of younger *users* (which is the metric FB would like you to know about).

            But as an advertiser, I can tell you that over 75% of all FB impressions are to the over-55 demographic, and over 50% of all impressions are over-65s.

        2. Justthefacts Silver badge

          Re: I predict

          I agree, *for the workplace*, chat won’t replace email.

          The thing that will replace email hasn’t been invented yet. It will probably be something like wiki with notifications, chat rooms, integrated document management systems. Like I said, hasn’t yet been invented.

          But with the statement about “unemployed ones”, you are already implicitly accepting that email will become corporate only. And I do mean corporate. People here seem to be so blind that *corporate* is a relatively small fraction of all workplaces. The fraction that has IT departments, true….but do you think a tree surgeon really needs email? Or will the “Contact Us” function on their website do? How do you contact your GP? Will it be email….or an evolved NHS App?

          1. Doctor Syntax Silver badge

            Re: I predict

            "The thing that will replace email hasn’t been invented"

            Doesn't this tell you anything?

            1. Justthefacts Silver badge

              Re: I predict

              The energy storage technology that will make renewables an *economic and reliable* source of electricity generation, hasn’t yet been invented.

              Doesn’t that tell you anything?

      2. Cliffwilliams44 Bronze badge

        Re: I predict

        Corporate emails and corporate electronic communications in general CANNOT be only stored at the users endpoint. They must be maintained in a service that can retain them for compliance purposes.

    6. Anonymous Coward

      Re: I predict

      > Email will still be around ...

      ... and we still won't have seen the year of the Linux desktop!

  2. xyz Silver badge

    50 years from now???

    All hail Emperor Zuck! Mind you, the big clog in the "network" will be slow human thought, so that lot will have to be deprecated and replaced by virtual human constructs that provide his Glorious Empire will data whilst avoiding all that eat, sleep and shit downtime.

    Or we'll be eating rats or something.

    1. R Soul Silver badge

      Re: 50 years from now???

      I for one salute our google/apple/facebook/twitter/cloudflare overlords.

  3. Anonymous Coward
    Anonymous Coward


    Both 1921 and 1971 were years just before or across big technologies breakthroughs. In 1921 radio technology and wider diffusion of electric power and internal combustion engines would have changed society. Two large wars and big political changes shadowed somehow that period breakthroughs, but it was also a real revolutionary ones - maybe we should call it the "electromagnetic revolution". That also paved the way to the "digital revolution".

    The idea of transmitting not only voice but video and data was already there - even telegraph lines were used to deliver quickly stock exchanges data, and all the military technology to compress (through encoding) and encrypt data was already in use to transmit what were actually "documents". Just there was a person in the loop with his/her eyes and a typewrite because storage was on paper.

    TV came quickly enough when radio technology became developed enough) - the real problem was the technology to capture, store and display data easily - only analog audio data technology was already there and cheap enough - which required the miniaturization of electronic components in the 1970s.

    After all, once you have an handheld electronic calculator, why it shouldn't turn into a computer given the basic workings are exactly the same? After all this is what happened when combustion engines became small and powerful enough you didn't something large as a locomotive or boat to make them useful. Many existing ideas had to wait for them to become viable.

  4. Sandstone

    Back in the Day

    It was called Compuserve.

  5. Anonymous Coward
    Thumb Down

    I don't see it

    To me, the barrier to the blob is the proliferation of single purpose, data slurping, possibly paywalled, apps.

    The Washington Post app has a negative incentive to play nice with the New York Times app and vice versa.

    No game publisher wants to play nice with other publishers and even within a single publisher there are developer rivalries.

    Beyond that, in IoT, Ring doesn't want to allow mix and match with Nest and vice versa. The list goes on.

    Basically the companies that write for the internet like the status quo.

    1. Anonymous Coward
      Anonymous Coward

      Re: I don't see it

      I don't access WaPo or NYT through an app. An app wouldn't let me block the ads and bypass the paywalls.

  6. Irony Deficient

    He then repeated the exercise with 1971 as his starting point, …

    … and concluded that predicting computers would become a consumer product would have been hard to do at a time when the dominant personal electronics product was a pocket calculator.

    1971 was a few years too early for the pocket calculator to be the dominant personal electronics product. In 1971, it was still the transistor radio.

    1. ecofeco Silver badge

      Re: He then repeated the exercise with 1971 as his starting point, …

      This. I've also noticed a lot of comments are off by at least a decade in their past timelines.

      I predict people will still not actually use the Internet to do research except for watching YouTube/TikTok celebutaunts who also did not do their research because somehow having (almost) all the information in the world at your fingertips is hard.

  7. localzuk Silver badge

    Seems to be mixing up the internet with services running atop it

    It may just be pedantry, but surely this is misusing the concept of the internet? The whole point is that the underlying systems use universally accepted protocols, and have redundancy built into it.

    What the Huston seems to be describing is a shifting of user level services and applications to servers nearer the client. Surely the underlying technology serving them would remain to be whatever form of internet there may be?

    1. Anonymous Coward
      Anonymous Coward

      Re: Seems to be mixing up the internet with services running atop it

      The real issue is carrier are terrorized by the investment they need to be able to cope with the increasing traffic - without being able to increase the end user connection price. So they're looking for something that could avoid such investments.

    2. Blane Bramble

      Re: Seems to be mixing up the internet with services running atop it

      I think it's another case of someone thinking the World Wide Web and The Internet are the same thing.

      1. ecofeco Silver badge

        Re: Seems to be mixing up the internet with services running atop it

        But isn't the Information Highway actually just tubes that make a web and thus the Interwebhighwaytubes?!

        Don't be a sheep, man! :)

  8. sreynolds

    Probably some bias here

    Was he talking from the limited viewpoint of the typical Australian for whom there are really on 5 centres and anything overseas takes at least 100ms to reach?

  9. AdaLoveseal

    "Common referential mechanisms..."

    ... surely don't mean IPv6, right? No way the Internet won't still be running on IPv4 50 years from now.

    1. ecofeco Silver badge

      Re: "Common referential mechanisms..."

      At this point, we can only hope. IPV4 is way past its shelf life, yet here we are.

  10. Brewster's Angle Grinder Silver badge

    That's okay for the earrth...

    ...but what about the rest of the solar system? The moon? The asteroid miners? Maybe the mars colonies (Muskville and Bezosbase)?

    We seem to be going through a space revolution. Maybe it will fizzle out---I'm still not quite sure why you'd want to be in space---but if not, they'll want to be part of the network.

    1. Cliffwilliams44 Bronze badge

      Re: That's okay for the earrth...

      The whole problem with space is the inevitable "elephant in the room", TIME!

      If your in Muskville and you want to look something up on Wikipedia, unless Muskville (or Mars in general) has a cached copy of Wikipedia your results will arrive at the speed of 1989 dial up!

      Currently it takes 5 to 20 minutes for communications from Mars to reach Earth. (Depending on the position of the planets) Now, will that improve in 50 years? I don't see how unless we get find a way to make radio waves move quite a bit faster. Even at the speed of light, it would still take 3.03 seconds (closest) to 3.11 seconds (furthest) to reach Mars. We all know 3 seconds on the internet is an eternity!

      So, for the residents of Muskville to have an internet experience we have now in the future they would need the largest caching server ever built! Large enough to cache a significant part of the internet!

      (I am sure the residents of Bezosbase will have a server caching the entirety of!)

      1. Charles 9

        Re: That's okay for the earrth...

        So TL;DR, unless we break all the laws of physics as we know them and find a way to pass information faster than the speed of light, interplanetary communication is a non-starter.

        1. Crypto Monad Silver badge

          Re: That's okay for the earrth...

          Interplanetary communication will use UUCP.

  11. ecofeco Silver badge

    Hand waving for fun and profit!

    See title.

    Good work if you can get it.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like