Hacker News new | past | comments | ask | show | jobs | submit login
Tim Berners-Lee says tech giants may have to be split up (reuters.com)
372 points by adventured 4 months ago | hide | past | web | favorite | 363 comments



> “If you put a drop of love into Twitter it seems to decay but if you put in a drop of hatred you feel it actually propagates much more strongly. And you wonder: ‘Well is that because of the way that Twitter as a medium has been built?’”

How exactly is breaking up the tech giants going to solve this problem?


Suppose tech companies were no longer legally allowed to use the network effect (i.e. they had to provide open-access to their public data via API).

Then this would enable alternative front-ends on twitter.

This would then enable me to filter messages by positivity, or whichever criteria I choose (e.g. non-Russian-bots, non-shills).

This is turn would then change the whole ecosystem. It would remove the perverse incentives of outrage-attention machine.


I think that people have an (almost) unalienable power, very strong, and pretty underused.

It's the power of walking away.

If you find an environment toxic, leave it, and ask your friends to leave it, too. Yes, you have to research alternatives first, and make compromises, at least temporarily.

Lately this somehow worked for Facebook. I bet they are going to see that on their bottom line.

So, if you dislike the outrage machine, leave it. Consciously ignore it. Do not retweet, do not link to tweets (instead, quote worthy tweets, they are short).

If enough people did that, and migrated elsewhere, that would be noted. Speak to companies in the language they understand, that is, the language of money. Their money comes from your attention spent on their property. Vote with your dollar — the dollar not spent by online advertisers for your eyeballs that are not there.


This is the part that I don't understand. Facebook, in the grand scheme of things, is still an incredibly new development for people. Most users (at least those who abide by the age limit) have spent most of their lives not being on Facebook. What grand utility does Facebook provide that makes it so difficult to stop using it? I cancelled my Facebook account a few years ago and the only time I ever even thought about it was when I'd try to sign in to services (like Spotify) that were linked to my account at one time and it got reactivated through that. I'd have to login and cancel the account again.

What is it about Facebook that has suddenly and inexplicably turned it into a necessity for people when they lived for so long without it? It's not like a cell phone or other technology that has massive utility. Most people don't even communicate via Facebook (from what I hear). They just post divisive nonsense.


Facebook (along with Google, Amazon, EBay, AirBnB, Stripe, etc.) get their strength from the long tail. I had deleted it off my phone and logged out on desktop, but then I heard (via email from my sister, who heard it on Facebook) that my aunt was in a coma and my cousin was posting updates on Facebook. That was enough to bring me back. And my cousin posts updates on FB because it's by far the easiest way to distribute news to everyone who cares without worrying if you're forgetting someone.

I'd assume a lot of casual FB usage is similar. There's 90% outrage posts, political stuff, memes, ads, chain letters, people sharing glamor shots of their vacations - and 10% pics of the grandkids, reconnecting with long-lost friends, networking into chance opportunities, and birth/wedding/death announcements that you wouldn't otherwise see. Missing out on the good 10% is a sufficiently large incentive that people put up with the bad 90%.


I'd argue that communication via email can work just as well as FB. It's a little harder to "opt in" to updates, but with spam filters and other email features, it seems like it can be a lot less noise to have to filter through. But this is coming from someone who doesn't use FB.


It works on a small scale with a group of people who regularly communicate with each other. That's how I organize most of my social events - e-mail a bunch of friends and say "Hey, wanna get together on Saturday?"

It fails when groups are larger or more loosely attached. In my cousin's case - my dad was one of 10 brothers and sisters (many with their own spouses), I have 16 cousins on that side (again with spouses), 3 half-cousins, 10 cousins-once-removed, 1 cousin-twice-removed, and there's a tendency for at least one person to feel offended if they don't get the news when everybody else gets the news. It's somewhat understandable that my cousin would want a broadcast medium rather than trying to remember all that.

Or as another example - a friend of mine died recently, and I found out through FB. I hadn't been in touch with her for several years, since before she got married, I'd never met her husband, and he certainly didn't have my e-mail. Still, I appreciated knowing, and passed on that info to other mutual friends, who also appreciated knowing. That's the long-tail; in my parents' generation, they might've found out at some reunion 30 years in the future, long after the funeral has passed and people are done sharing memories & photos.


The problem that Facebook solves is finding a way to contact someone when all you know is their name. If you don't have a phone number, address, or email but you know First and Last (or even First - if you have related friends), then Facebook still works to get in touch. Whereas phone books for personal numbers aren't really a thing. Facebook is that phone book.


I mean it technically can be equivalent but the experience is significantly worse.

One huge problem is that you can't add people's emails with just their name. Another problem is that there's no concept of accepting a friend request.

Sure you can get around it but why bother?


One thing that I have been discovering recently: Discussion groups. Back in the day it was Usenet, of course, and there are still websites that host niche forums. But it seems I am now running into weird new interests where the major discussion is happening on a facebook group. I go months without logging into facebook, but this seems to be what's drawing me back in.


> What grand utility does Facebook provide that makes it so difficult to stop using it?

For example, all your friends might be on it, and organise events via it. Then, if you don't use FB, you don't find out about events you might have wanted to go to.

This is the power of network effects, whiuch causes monopolies. The monopolies would be broken up if social networks had to use open protocols to allow interoperability.

As Irina Bolychevsky and James Moulding put it ( https://newsocialist.org.uk/do-we-really-need-a-statebook/ ):

> Can you imagine using WhatsApp to chat to your friends on Reddit or share photos from Flickr to Facebook and still see likes and comments? That’s the power of open protocols.


Can’t a Google calendar invite provide the same functionality?


It could, but if you rely on being sent a calendar invite you'll miss the event, because people aren't sending calendar invites, because everyone is on Facebook, because people aren't sending calendar invites, because everyone is on Facebook. This is the network effect, as the parent post was saying.


To most of my family in the Philippines, from children to elders, they only started using the internet heavily when Facebook was available (cell phones and decent data coverage were big enablers of that though).

Facebook provides clear utility to a society that has had no decent online method of replicating their group/communal way of sharing experiences before. Do they _need_ to share their experiences or communicate through Facebook? No, but it’s an obvious extension of what they’ve already been doing.

FWIW the issues I see with Facebook here in the west—fake news and divisive nonsense specifically—are exacerbated there, so definitely a double-edged sword.


It's not so much utility as it is addiction. A lot of the popular platforms are implemented to take advantage random reinforcement / variable reward schedules, social approval, ego validation, etc.


Walking away sounds easy on paper, like for gaming addicts. The articles about Facebook from the past come to mind where management was fully aware of the addiction factor, and in fact did everything to encourage people to stay on Facebook longer, including showing them "relevant" content and encouraging likes.

Maybe restrictions, similar to gambling could be put into effect? For instance, clearly stating the business intent of the site, requiring ID, limiting to certain hours,.. its sounds weird but it reflects how underregulated the internet is compared to other areas.


If such addiction is something that can be diagnosed via a reasonably rigorous procedure, then yes, we'll probably have to do something similar to other addictive things: age limits, clear warnings, maybe even independent ongoing testing of the effects. Compare to selling alcohol.

Most social sites already impose age limitations.


> Compare to selling alcohol.

Is the potential risk of addiction related to why alcohol sale is regulated? Most rationales I've seen for preventing sale to minors relate to health effects and immediate intoxication effects (car crashes etc.) instead.

Don't get me wrong, I'd be thrilled if addictive potential was considered primary among reasons to consider regulating something. I'm just not sure it is.


If you don't like smoking, you should quit. Sounds simple, but in reality it's very hard for a lot of people.

Facebook et al. are habit-forming, and have been designed to be so.


Or we could support strong regulation and not have to passively accept being pushed into a false dichotomy!


And whom do you suppose would be writing those regulations? Of course the answer to this question is the very same reason that the tech giants will not be split up. We have a mountain of rules and regulations and anti-competitive behavior, price fixing, and all other sorts of stuff. And then we have Time Warner/Comcast openly agreeing to not compete, charging arbitrarily high prices, and the FTC apparently deciding there's no such thing as a merger that might lead to reduced competition. Then you have the head of Comcast golfing, partying, and most importantly fundraising like a fiend for the former president.


One of our two political parties is campaigning on a platform to reign in this type of campaign finance influence:

"Democrats would use their first month in the House majority to advance sweeping changes to future campaign and ethics laws, requiring the disclosure of shadowy political donors, outlawing the gerrymandering of congressional districts and restoring key enforcement provisions to the Voting Rights Act, top Democratic leaders said on Tuesday."

https://www.nytimes.com/2018/10/31/us/politics/democrats-mid...


Everybody says this before elections. Trump was going to 'drain the swamp' and Obama was going to take down k-street influence. Then they get into office, tip their hat, and turn their back until the next election when suddenly they're incredible again, or at least they promise they will be. Obama ended up creating a special executive chair to let a Monsanto VP take control of the FDA, tried to jam the corporate written TPP through in the most undemocratic and secretive way possible, chose not to even try to hold the banks accountable for collapsing the economy, and so much more. Trump's apparently decided his go-to man on foreign policy and national security is none other than John 'Yellowcake' Bolton.

That old joke really is true. How do you know when a politician is lying? Their lips are moving.


Does cynicism help solve the problem?

There are also bunch of promises that Obama kept (and Trump). Focusing on some that he didn’t keep doesn’t change that.

https://www.politifact.com/truth-o-meter/promises/obameter/


It’s much more fashionable and edgy to point at any failure and then judge their entire administration on solely that point. Don’t forget to dramatically overestimate the presidents ability to effect change in the government as a whole, and instead hold them accountable for congressional obstruction.


Indeed it does! We live in a democracy and we have the ability to elect literally nearly anybody we'd like for any office. And fewer Americans than ever strongly identify as either republican or democrat. Yet, somehow, republicans and democrats make up nearly 100% of all office holders at all levels of American government. Isn't this interesting?

The reason is tribalism, and more specifically politicians becoming extremely adept at driving and exploiting tribalism. Many of the issues held as cornerstones by both parties are ones that are relatively unimportant. The reason they are held as cornerstones is because they do an extremely good job of dividing people. The reason for the desire to divide is because it helps both establishment parties to maintain a grip on power.

How? In the most recent presidential election what percent of people do you think voted for Hillary thinking 'Yes, this person truly represents what I value most and will make a great president.' By contrast what percent voted for her because the alternative was simply unacceptable? And similarly for those that voted for Trump. By focusing all of their energy on dividing people it makes people ignore the failings of their preferred side and instead focus on the awfulness of their less preferred side. This, in turn, does a fantastic job of getting people to vote against their own interest. And, in turn, this also encourages both sides of the political spectrum to play up to their own villainy (from the other side's perspective).

By continuing to vote for the 'least awful' choice instead of the choice people actually want, this system will be perpetuated indefinitely.

---

Interestingly enough, I think the efforts made towards sharply dividing people started to happen around the early 90s. And something happened then that was probably not just a coincidence. Ross Perot, at one point, looked set to win the US presidential election. He was polling ahead of both Bush and Clinton. He ultimately ended up taking 18% of the vote and, at the minimum, working as a spoiler. Ross Perot was a complete outsider. And he got those tens of millions of votes running a straight forward platform that he advertised on public broadcasting and infomercials. All these fancy political campaigns, teams, and political strategery was was nearly upstaged by this [1] guy with some printed out graphs and plain speaking.

And so it looks like works on min-maxing elections went into overdrive since then. It just turns out that getting people ragingly mad at each other over inane issues turns out to be one of the best ways to keep getting elected. And so yes, I do think pointing out how both parties in the US are complete trash does help solve the problem. What we need is people to stop being so scared and tribal. Instead vote for whoever you think will do the best job, instead of trying to decide who's "electable" and will do the least awful job. Trump, if nothing else, should show that the notion of "electability" is a complete lie and tool to maintain power. Anybody is "electable", anybody is "presidential". Vote for who you want instead of kicking in that cognitive dissonance, pretending the future won't be exactly like the past, and voting for the exact same idiots over and over.

[1] - https://www.youtube.com/watch?v=mPIVI0CbCmg


>I think that people have an (almost) unalienable power, very strong, and pretty underused.

>It's the power of walking away.

Sure, although it seems the power of collective social pressures is far stronger. So while that might work for some people, it won't for most.


>the power of walking away.

Very true. But it would be easier to walk away if there were an alternative to walk to. Such as alexandercrohde's proposed site which is essentially a reality/emotion filtered twitter/facebook stream.


The alternative is the real world, where you look into the eyes of the faces around you and put your energy into those relationships more than others. (Not being flippant. I was on FB 9 years, but realizing that I was prioritizing the far over the near was what got me to say goodbye).


I think the comment you're replying to is trying to strengthen that ability. Now you don't have to convince some critical mass of your friends to switch over; you just switch over to the new platform, which everyone is connected to via some open source API/protocol, or the company running the data service is not running a front end to present the data. You continue to get the data from your friends and family while using a front end that you feel safer in.

It would require regulating the companies who hold personal data and divorcing them from the entities that tailor feeds and provide a user experience for the data. I don't think it's that ridiculous of an idea, really.


My point is exactly not having to impose regulation.

The power to move away is already in the people's hands. You can do it at any moment, without a governmental mandate.

The point is to use this power. But first people have to recognize their power, and learn to use it a bit.


When AT&T was a monopoly and you had to rent your phone from them at an outrageous price, you could also walk away. But having to either walk/drive around or send slow mail to communicate was not very efficient. Perfectly doable, mind you, but very inconvenient. Sometimes the cost of walking away is just too steep, so abusing a dominant position becomes very easy and tempting.

I don't think Twitter or Facebook are anywhere close to that threshold. Maybe that's just because I barely use them. But I'm sure some people were also not using phone in the seventies, so I won't conclude based on my individual case.


Walking away from AT&T was hard because of the expense of building a phone transmission network, and (I suppose) because of some patents they held.

Walking away from Twitter is much easier, because multiple global instant or near-instant messaging systems exist, including completely open and user-controlled systems, and even building a new such system is reasonably easy for a group of competent people.


The cost of getting everyone you wish to communicate with using the new system is non-trivial.


That's looking at it from a purely technical perspective. A 20% better Twitter or Facebook (from a technical perspective) is certainly comparatively easy to do, but would still have an approximate value of nil. The value of Twitter or Facebook is not technical (though running services at this scale is technically very challenging): it comes from the people using it. Building another Facebook, even with marginal improvements, is never going to be enough to make enough people switch.

So is the case for AT&T: even if you could build a network competing with the AT&T one it would still have been a largely useless endeavor, unless you could also force AT&T to be interoperable with your network.


Thanks for posting this. It's all too often people (like in this article) want to lean on bureaucracy and legislation to fix their problems as opposed to actually doing something about it. Not using Twitter and Facebook has a positive step for my sanity, much like not watching 24 hour cable news was another positive step. I don't need the government to break up CNN or News Corp to solve that problem.


The problem is far broader than that. Other people are still watching cable news or read Facebook 24/7, and what they take away from it guides their actions. When those actions start affecting you, it kinda becomes your problem as well.


No. That's just more software engineer hand waving to avoid taking responsibility for the way their products have changed the emotional landscape for real people in the last couple of decades.

It's like saying the CIA wasn't trying to destroy communities when they were handing out crack, because "moral people wouldn't have taken it". Every single person building these platforms knew how addictive they were. It was the entire reason these companies were valued from the start. It was the end goal.

Yeah, a junkie can decide not to go find another needle. But if that is your response to the drug epidemic, you then have marginalized people who you've decreed "not worth saving". And I don't like what that has looked like over the last 60 years.


This is correct: if you see that you're taking part in building something you know (or suspect) is a bad thing, you are responsible if you continue to work on that.

This only reinforces my point: if you think it's a wrong thing, walk away from it. Leave that job at Facebook data science division; there's a number of other well-paid data science jobs.

But I'm not trying to talk about what "they" have done wrong, and could have done differently in the past, or what regulators should be doing. I'm talking about something you and me can do right now, and what is completely within our own power to do.


+1. Totally spot on.


It's great to be able to stop

When you've planned a thing that's wrong,

And be able to do something else instead

And think this song:

I can stop when I want to

Can stop when I wish.

I can stop, stop, stop any time.

And what a good feeling to feel like this

And know that the feeling is really mine.

--Fred M. Rogers


There's so much right with your take here, IMO.

It flies in the face of so much opposition from reality – much of it credible – like how the hell do you make money when you eschew walled gardens, engagement boosters, and adtech friendly practices? Do people even want to make the tradeoffs they'd have to? Is regulation too heavyhanded?

Will it be that for as long as these systems can be turned into money machines, old money will use them to print new money? Perhaps.

But the mentality of "open at all costs, profits be damned" feels like one we in the tech industry have gradually let slip from our ethos, and it's so core to the old web. I miss it.


I can't help but feel that the Internet is hamstrung by overreliance on advertising as a revenue model ...

All of the opposing forces you describe simply evaporate if you charge a fee for access to the API.

Something small and unobtrusive but which would at twitter scale add up to an awful lot of moolah.

Because you're not trying to monetise your users beyond that it doesn't really matter if loads of people stop using because the network effect isn't important.

But at the end of the day it's all about monitoring the activities of mass populations as much as it is about direct advertising revenue, and that's too priceless for the big corps to let go of.


It matters if loads of people stop using it to the point that you can't keep the lights on.

I'd suspect that "loads," in fact the majority, of twitter users would not pay anything out of their own pocket to use it.


These remarks don't really address the points I was making.

Firstly, if people want to use it, and pay for it, even small charges at mass scale pay off. Something like twitter doesn't cost a lot to run in particular at unit cost.

If they don't, GOTO 10: "the Internet is hamstrung by overreliance on advertising as a revenue model"

How else do you monetise your service?


Sounds like it isn’t very valuable then.


Why would network effect not be important? Even without the advertising model, your profits would still depend on user count, no?


No, that's not the network effect I'm thinking of.

To wit: The value intrinsic in the relationships between your users ("The Network") - to the point that you want all users whether paying or not.


That would be just as important for a subscription based service. If your friends use the service, you're much more likely to continue using it and keep paying the monthly fee.


Yawn


> Something small and unobtrusive but which would at twitter scale add up to an awful lot of moolah.

That only works if there is something else preventing competitors from monetizing end-user attention instead. That's possible, but (I don't think) probable.


"like how the hell do you make money "

Well, you could charge for it. But of course you wouldn't be able to compete with free - you'd have to have competitors who needed to charge too.

They do exist but are pretty niche. The WELL comes to mind. www.well.com

If it's any help I applaud every newspaper that puts up a paywall, and more are doing it. It only came about when the industry was (is) about to be completely wiped out and there weren't really alternatives though. But if you're not paying your journalists, someone else is paying them to tell you what to think.

I miss the old internet too :-( . It had so much promise.


I just had this same conversation with a colleague. The old internet was really the civilized wild west. All the idealogical virtue of sharing and learning was plentiful and you had to have some basic technical skill to contribute or be involved enough to learn how to do it. Now, any curmudgeon can post whatever brain farts pop out with a frequency that only rivals that of their actual mouths.


Exactly this. Before you could ignore the lone racist person in the neighborhood. Now, they can go online, create a forum and gather together...

so much for the promise of connecting the world...


Maybe the nice thing about the old internet was that it felt distinct from the rest of life, instead of an integral part of it - or, frighteningly, much of the foundation. Bad things happened on the internet (or AOL, Prodigy, etc.) in 1994 but then I just signed off and went about my day. Also, it felt like something most people hadn't quite figured out yet and, well, of course it was going to attract eccentrics of all stripes - including some less savoury ones.

I loved the internet, and Prodigy, when I was a kid in the 90's and 12 year old me would be shocked to hear that sometimes I dream of going to a cabin with just a landline, radio, and desk for writing letters for a month.


Wow, you exactly nailed how it felt back then. You mentally switched modes “ok, now I’m online”, and it felt distinctly from “real life”, mostly because accessing it was hard (compared to the present). I think the problem with current internet is that life outside is mostly the same as in 20-30 years ago, and we have artificially overvalued internet presence (youtubers, influencers, online tracking, data leaks, etc)


When was that "before", though?

Stormfront was probably the best-known forum like that before Gab. And it has been around since early 90s, first as a BBS, then as an actual web site.


Mostly before Facebook and Twitter gained mass userbase, before they made it easy for anyone with a smartphone post all their thoughts


>Well, you could charge for it. But of course you wouldn't be able to compete with free - you'd have to have competitors who needed to charge too.

Are you sure about that? Twitter's annual revenue is 2.4 billion dollars, and they have 300 million users. $0.55/month per user is all it would take to match their current revenue.

Sure, when a service is getting up off the ground, you can't beat free, but for one that's established, like Facebook or Twitter, the subscription fees can be made miniscule.

As long as the advertising model is used, a set of perverse incentives exist that encourage social platforms to treat advertisers as more important than users, to inflate their user numbers, to lie about views, to tolerate fake accounts and harassment, and to close their APIs.

A $1 premium ad-free option with 50% adoption would increase twitter's revenue and remove the perverse incentives.


And as soon as they do that somebody else will start a free clone to siphon off their users, and people will abandon Twitter. Getting people from $0-$1 is a lot harder than $1-$5. Humans aren't very rational.

For that matter, note that mobile game companies found out it was harder to get all your users to pay $5 for a game than to get 1% of your users to pay hundreds.


This assumes that all Twitter users are actual people and that one person doesn't have multiple accounts all over the place. I would imagine that, although still not a major issue, the cost would be a bit higher for users and that's only factoring in users that find enough utility in Twitter to think that it's worth paying for.


> A $1 premium ad-free option with 50% adoption would increase twitter's revenue and remove the perverse incentives.

It would. A $1 ad-free option would see nowhere close to 50% adoption though.


This feels like a uniquely engineering perspective. My parents won't filter twitter, they also won't go searching for some complicated front end.

In my opinion, the problem of hate spreading is not a technical one, but a social one. These platforms are just better exposing it, but if they didn't exist, or were somehow filtered or modified, then the people would just go elsewhere.

You already see this with reddit, where they cracked down on a lot of the hate based subs. The result is that new sites popped up to cater to the hate, message boards, news sites, and others. Plus, these hate groups operate locally too. In Canada there have been many cases where they setup outreach groups in person in neighbourhoods trying to recruit.

I'm not sure what the solution is, but the core issue is that at this point in time, hate sells. Maybe people have lived in such a good time that they forgot what wars were like. Maybe inequity is at such a high that those not in the top 10% are just filled with bitterness and rage wondering why this sort of divide should exist.

But I'm willing to bet strongly that this is people problem, not a tech problem.


Social media is basically gossip.

The only difference is scale. You had to be a prolific gossiper to spread a message to several dozen people before. Now, it's one click button away for most of us.


This looks a bit optimistic or underestimating real life tradeoffs.

There can be an API, yes. But if you want to filter by positivity, you have to expect the company to actually implement that in the API. Otherwise, do you expect to get all billions of tweets over an API and then add positivity on top?

You can't have sustainability, performance and "open for everything" on one single API, there are tradeoffs and constraints involved. Even GraphQL needs certain fields to be filterable even before the API call returns, otherwise one can't just loop over all commits in GitHub universe without exhausting both consumer and provider systems!

Atleast that's what I think. Would be happy to be proven wrong.


I think these are good questions. But I also think there are good answers.

One solution is for my client to sentiment-analyze as a filter after the api-call and downrank that way (much like reddit/HN downrank controversials).

Maybe a more generic solution is to allow parties to "tag" twitter accounts. Clients could then make intelligent deductions based on tags (e.g. "political," "controversial," "satirical," "pg13").

I'm sure many other people have many other smart ideas.


I wonder if this is one of those few genuine uses for a blockchain based system. Comments (tweets) would be posted on a very thin layer of technology, which would only support say: - comment id - comment text - datetime of comment - commenting user id - list of [upvoting/downvoting] user id - list of reply comment ids

As you say, third parties could then come along and build whatever front-end they see fit. Don't want to show or allow clients to downvote? Go for it! Only aggregate upvotes from user ids over a certain age? Be my guest! The underlying information would never change.

Operating costs would be funded by transaction fees, whatever that specific blockchain model may be.


But why blockchain?


Because the underlying message service wouldn't be controlled by any single entity. This would mean that once the code was out there, you could be confident that no-one would change the service when there was a change of CEO or new censorship laws brought in or something like that; we'd have a neutral piece of underlying infrastructure that anyone could use to build a robust discussion platform on top of.


or just stop using twitter?


I've used this strategy consistently for things I don't like, works great! Don't understand why more people don't consider it.


I know, right? The only time I encounter Twitter is when others link me to it. I’m okay with that and sometimes I share forward, but I’m not fed a daily stream of tweets. I left google services a few months ago. It was a long weekend of work, but I’m happier than I was before with otherwise no noticeable impact (except the emails that write themselves for you in gmail). I exported all my data from Facebook around the same time but I’m still mulling over how to resolve deleting 10 years of my digital life against my core principle that things I’ve said and done should not be allowed to disappear even if my 30 years future self would look back in disgust.


A lot of people don't really have this option since their audience is on twitter. Many artists would lose basically their entire livelihood if they just stopped using twitter and stopped being able to communicate about their work. You may have this option but it's not true for everybody.


at a certain rate of adoption doesn't that kind of become like "stop using email"


My daughter is in a youth choir, whose director insists on communicating with parents via their Facebook page, which is of course not publicly viewable. I have told her that I will not join Facebook solely to be able to read communications from her. I offered her my phone number and email address, and so now for important things she will send an email as well. But she was absolutely incredulous that someone was not on Facebook in this day and age, and was not interested in joining.


Nobody uses email any more ...

I mean they use it but it's not a high-engagement medium. I'd guess nearly 90% of emails never get read.


> Nobody uses email any more

That's not my experience.

I use email. At work they use email. My mom uses email.


Yeah I think I made my point kind of stupidly.

Email is just "there" at this stage. Your emails appear in your inbox and you're not really constrained by a provider.

I was taking aim more at the comparison with twitter and facebook which require constant engagement.

With email you just get it the way you want it, on your terms mostly.


> With email you just get it the way you want it, on your terms mostly.

Sounds pretty perfect.


Email is far from perfect.


GP: I’m convinced this is how people with poor email management skills feel. It’s how I felt until I stepped up my filter, unsubscribe, and delete game and really developed a strategy for managing my inbox. It’s not heavyweight but the goal is to reduce your inbox to a task/todo list of sorts—only the things you need to worry about. Might be a direct inquiry or a notification about an upcoming sale, etc. Eventually you get to a point where you realize that _everything_ uses email. The internet _is_ email. It’s just really easy to end up being bombarded by so much of it that it feels useless.


The only people who don’t use email are products.


I don't use Facebook, I presume thats the level of adoption you are talking about.


Many people don't use facebook. I thinks WhatsApp is a better example, it's even more ubiquitous, though I'm sure there are still people who don't use it. Heck, I'm personally thinking of stopping, it's just that everyone is there. Two of my recent orders were negotiated by WhatsApp. It's used almost universally.


WhatsApp isn't even close to as ubiquitous as Facebook. WhatsApp has about 1.5 billion monthly active users compared to Facebook's 2.1 billion.

https://www.statista.com/statistics/264810/number-of-monthly...

https://www.statista.com/statistics/260819/number-of-monthly...


Interesting, Doesn't match my experience. I'm wondering what the age distributions are on those. Thanks for correcting me!


I think that depends on what country you're in. The only people I know who are on whatsapp are immigrants who use it to talk to family in the old country.


Yeah precisely, we need some middle ground between "don't use it at all" or "have 0 control over your data"


Federated systems, like Mastodon (or, well, email), offer a rather reasonable balance, I suppose.


You can be negatively impacted if you do not have an account, even if you do not read tweets at all.


Not using twitter/facebook/google/etc doesn't exempt you from experiencing the negative effect they have on the internet and the world around you, though.


I have been suggesting something slightly along these lines.

I think that the monopoly companies provide the platform with the large network. This platform is desirable. The monopoly control over it is not.

We can replace the monopoly platforms with decentralized peer based platforms running on technologies like Swarm Framework, IPFS, WebRTC, cryptocurrency, etc. Diverse companies can build varied opportunities that leverage the common platform and large network.

Its a challenging goal and most distributed technologies haven't been proven at Twitter scale. However I believe it is attainable and should be attempted.


Will you pay to access that front-end? I'm not a fan of closed systems but I notice that folks want access to other people's system and data without paying for it.


One thing that could help is more client-side tools to enable this. It's remarkable what you can get away with individually that would be blocked en masse.

Social fixer, for FB years ago, made the feed much better and was purely client side. Personally, I got sick of the real estate websites where I live and just made my own frontend for them, crawling with Scrapy. I haven't gotten a 403 yet.

Not sure how to make money off of it, though - short of selling people software in exchange for money.


I'd be curious to see what effect this has on business models.

A lot of useful features are free because they can make money on the backend or through paid promotions (search, email). Take away that ability and a lot of the services people enjoy will be forced to be paid or have annoying amounts of advertising.

I agree there is serious concern and we need to have discussions. I would consider breaking up companies a last resort though.


This assumes that someone creates a free and easy to use front end that has these filters. I can believe that.

This also assumes that a large fraction of Twitter users would use such filters. What makes you think they will?

The reason I have doubts is that people who tend to be influenced by evil Twitter posts are not the type of people who are amazingly self aware and look for ways to reduce that influence.


I haven't been on Twitter in over a year, but lists already exist, don't they? So the filters already exist.


How would providing open access to their public data disallow them from using the network effect to their advantage? I'm not really following.

I get how it would enable alternative front-ends, but just looking at other platforms, these clients don't get used nearly as often as the official clients, so their existence wouldn't have much of an effect on the overall system. Look at Discord, Telegram, etc.; only a tiny percentage uses anything but the default client.

Plus, you can already filter messages like that, but in a much simpler manner: just follow people that feed into the "outrage-attention machine." If you get upset when you read a certain person's tweets, just unfollow them, and only try to follow people that you either personally know or really respect.

One of the things that I've realized is that I become more unhappy when constantly buffetted with bad news that I can't change. Things like country- or international-level news, international forums that feed off of negativity, etc; they all feed off of a number of views, or upvotes, or replies, or whatever, so they constantly have negative news. And because I can do nothing personally about the vast majority of these news stories, they simply make me frustrated or unhappy for no reason.

TL;DR following less people (or not having an account at all) fixes Twitter, not access to a public API.


How about a time limit on how long they can have exclusive rights to the network? Like a patent.


How about a limit on how long you have exclusive rights on a house you built, and did not sell? Okay, how about the exclusive rights on your blog that attracted a million subscribers? Should a government agency regulate what you post there?

I'm afraid this is a slippery slope leading to things far more dangerous than we have now.


I agree it could be a slippery slope but we are nowhere near it today. A published work copyright expires 70 years after the author dies, or something similar. That would probably apply to a blog post. 70 years post death is pretty reasonable. Not sure what a reasonable number for a network would be, but I think "forever" is excessive in the other direction.


I do love the idea of Twitter's API being public but I like it for the opposite reason.

The opposite reason being that it would allow people to eliminate censorship if they so choose.


> Suppose tech companies were no longer legally allowed to use the network effect (i.e. they had to provide open-access to their public data via API).

Facebook and Twitter have public APIs.


Come on. Facebook and twitter APIs are so restricted you can't even build a usable alternative client (look at all twitter clients that we had a few years ago, they are all gone now).


Facebook won't even let you read public events through their API anymore, public events created by businesses that want the public to attend, that are visible to anyone even if they're not logged in, that are easily searchable through Google.

But that's probably because the same day they removed that events API, they launched the Facebook Local app that shows all the events around you. Can't have anyone competing with your app.


Perhaps what we are seeing is not a problem with technology but a problem with the human condition?


It's perfectly both.

Hatred is basic - complex emotions break down to something more direct and blunt when your message gets boiled down to X number of characters.

Then take interpretations of that message in mass, again each one distilled to simpler tones: most will mirror or concentrate the negativity.


Still, rickjr85 is right.

Even in your response you implicitly concede that the root problem is human nature. You're implying that if we get rid of Google, or Twitter, that there is some kind of alternative technological method, (which you haven't elucidated on), by which we can mitigate the effects of the root problem.


violence is a part of human nature, too, but the technology of the atom bomb multiplies the danger for us all. So too with these newsfeed algorithms, which favor engagement above everything else, no matter how base and degraded the content is.

I've been thinking a lot about this interview with Jaron Lanier, and I'll just share an excerpt because I think it provides some insight: "The problem, however, is that behind the scenes there are these manipulation, behavior modification, and addiction algorithms that are running. And these addiction algorithms are blind. They’re just dumb algorithms. What they want to do is take whatever input people put into the system and find a way to turn it into the most engagement possible. And the most engagement comes from the startle emotions, like fear and anger and jealousy, because they tend to rise the fastest and then subside the slowest in people, and the algorithms are measuring people very rapidly, so they tend to pick up and amplify startle emotions over slower emotions like the building of trust or affection." https://lareviewofbooks.org/article/delete-your-account-a-co...!


I suppose the limitation of use of nuclear weapons stems firmly form public understanding of their dangers (even if the understanding is vague), and thus public disapproval.

One thing to do about the blind engagement machines is to make more people more aware of them. Think about ways people handle hazardous substances. Knowing that a substance is hazardous is the most important bit, to my mind.


Yes, rickjr85 is right, in the way that a single point can occupy one side of a Venn diagram.

What's funny to me is that even your response I read as very negatively conveyed. But just reading it word-for-word, there's really no sort of animosity whatsoever.

But to the main point, I didn't mean to imply a side to the argument the article makes. I didn't even read it. It was meant to be isolated to the context of the parent comment. Yes I know that's nonsensical given we're all commenting ABOUT this article.

We should never do something so stupid as to get rid of Google or Twitter. Forces will act as they must to create these technologies. People will run them as they will, and they'll be influenced by other people, which may direct the impact of the technology in question.

That's a long-winded way of saying next to nothing. Rather than there being some kind of alternative technological method, it would go farther to address the issue if people just all implicitly understood that free speech is built into all technology, and built under the constraints of law.


> in the way that a single point can occupy one side of a Venn diagram

I'd say you're a hoot at parties.


In this case technology allows those human conditions to breed. Because they remove many restrictions which are pushing people toward sanity.


Maybe it's not a problem.


I think this theory gets buried with zero thought. It is widely assumed that we are worse off because of social media and mass communication. I don't think this has been even close to being proven correct. People have rose-colored glasses about the past and how progress is actually made. It is never smooth or without severe argumentation and usually violence.

Maybe the magnitude is larger, but that doesn't make it worse. It may make it simply faster.


I doubt breaking them up is targeted to solve this problem at all. I believe the only problem breaking these companies up allows is the emergence of competition and innovation.

As it stands today, it seems that any new technology company with a decent idea/product/service, is heavily incentivized to exit the market by being acquired and left to languish. Nest comes to mind and is only a single example. Another one: I was at ARMTechCon2018 and an up-and-coming RISC-V competitor of ARM was trying to be swallowed by, you guessed it, ARM.

The giants are truly monopolies and there are some of them that are capable of influencing entire economic market segments. Do we want that to continue?

EDIT: Grammar...I sometimes suck at it.


>“Before breaking them up, we should see whether they are not just disrupted by a small player beating them out of the market, but by the market shifting, by the interest going somewhere else,” Berners-Lee said.

The implication is that maybe some of the tech giants will one day fall due to their own Achilles Heel. For example, Twitter's case that you cited.


And if that doesn't happen then he proposes breaking up the tech giants, which solves the problem how?


It's harder for 100 companies to collude undetected/unchallenged with no viable alternative than for 1-5 massive/politically-savvy companies to tweak results/ban users to fit their political biases. The larger a company is, the easier it is for it to blow-off any lawsuit losses over bad behavior (or simply have bad customer service) and continue those bad practices.


It won't. But it will limit effects of bad actors, give users a bit more control over their environment and probably stop some emergent behaviors.


The web is designed to make limiting the propagation of information (and any emergent effects from that) impossible.

How will the effect of 'bad actors' be limited when they can just move elsewhere? You don't need sites of the scale of Facebook or Twitter to have emergent, viral or culturally relevant effects occur.

What makes the big sites like Facebook and Twitter isn't their size per se but their capture of network effect and discoverability... they are content aggregators (among other things) and so the content they aggregate can by definition be found elsewhere.

If those sites are broken up, then users have to put in more effort at discoverability (which was the case with the "old" web), but that problem can be solved by smaller sites, other aggregators, other search engines, apps, etc.

It's like the Force in Star Wars.. the light side or dark side doesn't matter, the Force doesn't seek good or evil, only balance. With the web, you can't reduce the propagation of information, or bias it effectively towards one moral or political alignment or another, just distribute or centralize discoverability. Either way the web will still find a balance.


>The web is designed to make limiting the propagation of information (and any emergent effects from that) impossible.

Are you unaware or ignoring the censorship of thought that Google and Facebook have been doing lately?

You are acting like it’s 2005 internet, and it’s not.


>Are you unaware or ignoring the censorship of thought that Google and Facebook have been doing lately?

Google can only censor Google, and Facebook can only censor Facebook. Neither of them can censor Twitter, which can't censor 4chan, which can't censor Voat, Gab, etc. No one controls the entire network, and no one can censor the entire network.

>You are acting like it’s 2005 internet, and it’s not.

It actually still is, people have just gotten so cynical that they've forgotten how to see the forest for the trees. Being popular on the internet is not the same thing as having authority over it beyond one's own domain.


Yes, just look at how well this worked out with Gab.


Gab is such a great example of this. You have a new player in social media (yet we apparently need to break up the big tech companies because you can't enter the market anymore).

It operates in a successful niche. And it also is home to many far-right actors who hate Jews and minorities. Yet the claim is that breaking up the tech companies would help stop hate.

Tim Berners Lee has no salient point here whatsoever, and I don't think anyone would care about what he's saying if he wasn't who he is.

Not to mention he's calling for government regulation of tech companies at least partially to cut down on hateful speech (which is very much against the right to free speech).


>Not to mention he's calling for government regulation of tech companies at least partially to cut down on hateful speech...

This is the most troubling aspect of his call. "There's too much hate speech, so let's take action against platforms until it goes away."

I'm tired of all the terrorist attacks too, but to my mind, you handle that by having your counter-terror people do what they do. It's not like the terrorists would go away if you break up Twitter.


> You have a new player in social media (yet we apparently need to break up the big tech companies because you can't enter the market anymore).

The existence of competition doesn't mean they're within the spirit of anti-trust regulations. Twitter using their network to eclipse other markets would be the concern.


Exactly. The existence of ubiquitous dial-up does not mean that Comcast doesn't have a monopoly on Internet connections in my area. As long as Comcast is offering 100Mbps and dial-up is 56k, the choice is obvious. As long as Twitter has 336,000,000 users and Gab has 400,000, Twitter is going to win.

Can you imagine Twitter's hosting provider taking them offline because they have objectionable content? Can you imagine their registrar revoking their domain name? It would be unheard of. It would be a catastrophe. They're too big to fail in the way Gab failed.


Gab (and Voat, and, er, Hatreon) were open attempts to create far-right versions of existing websites. I don't think they're good examples of the inevitability of hatred.

I'd hazard a guess that many more people are angry about hate speech on Reddit than about hate speech over email. The difference is that one is a company and the other is a federated standard. If I've understood Tim Berners-Lee correctly, he wants to replace Twitter with a million Mastodon instances rather than a hundred different versions of "Twitter, but for the far-right/far-left/radical-centrists".


They were not created as such (well, maybe with the exception of hatreon). They were created to be similar platforms, competitors or clones, with a promise of better management/operation/features. See also: Vidme.

When you compete with a monolith, your early adopters will be the people thrown off the bigger ship, and most of those people were thrown off for a reason.

Very few alt-repreneurs seem to understand this


Gab was very much created with that kind of content in mind. Its creator was specifically concerned with "left-leaning Big Social media monopoly", after Facebook and Twitter started censoring alt-right accounts. Which is not surprising, given his past history:

https://techcrunch.com/2016/11/12/pro-trump-ceo-gets-booted-...


They were created to be uncensored versions of existing websites. (I don't think there're any examples of Gab limiting left views.)

It happened that the existing websites were censoring far-right viewpoints.

So...by default the predominant users were far right.


Yeah, this specifically is quite the top down approach. The realization to have is that social media/the internet is a tool, and how it's used (or becomes most popularly used) is simply a symptom, a sign, a reflection of the current state of society; yes, if we are reactive and not responsive with reasoning, critical thinking, then propaganda can flourish more easily.


Yeah, a dedicated actor can trivially just make meta-posting frameworks and tools for any number of services. It's not as if any of them are magic, at the end of the day it still all boils down to https/POST/GET and such, even if no official API is provided. The exact same message could be posted to as many places as desired for the same amount of effort. Bots will be able to function just as easily. If anything one would expect reduced scale would increase the difficulty of handling a very large scale adversary like a state actor. It would only seem to disadvantage smaller players.

And the article also lumps together wildly divergent companies in the same bucket while ignoring where scale is necessary. If you're talking about Apple for example that necessarily means talking hardware since they're primarily a hardware company, but hardware capex is getting ever higher. What's the plan for getting technologies to market where combined upfront expenditures are in the many billions of dollars? Fabricators and state of the art SOC design aren't getting cheaper. Neither are blue sky but critical next generation technologies like future display improvements (microled, direct retinal projection, or whatever).

I mean, a lot of this can be argued as a failure of government. Ideally there would be way more basic R&D happening there with way less red tape with the results then available for all. Government (at least in a democracy) is the natural place to be doing public service security help, and to handle authentication services when two strangers want to have a trusted identity foundation between them. But in reality government is not doing a great job of any of this, nor does it seem set to do in the near future, but we need it anyway. I don't even think giant scale corps have actually been all bad by any means for things like privacy vis-a-vis government, because while sometimes they can cooperate they can also oppose and be impossible to trivially knock down. Would a bunch of tiny random generi-phone corps really have been more effective then Apple at opposing the FBI for example?

Our system is very, very far from perfect. But it's also got us some frankly amazing and wonderful stuff. I think everybody should be instantly and inherently suspicious of "revolution" type change proposals. There is a real risk of tossing the good with the bad and unintended consequences and even opening the door for something way worse. Incremental, focused changes with clear specific goals in mind and then seeing how they work (auto sunset would be nice to see more) normally seem like the more effective way to go.


>Bots will be able to function just as easily. If anything one would expect reduced scale would increase the difficulty of handling a very large scale adversary like a state actor. It would only seem to disadvantage smaller players.

FAANG companies are incredibly bad at detecting bots. It’s not how they make money so they don’t invest much in it.

There are companies that specialize in this that are used _everywhere_ behind the scenes and are growing quickly.

You’re not going to see that kind of development in house. You’ll see a new market for protection which actually already exists.


> FAANG companies are incredibly bad at detecting bots. It’s not how they make money

Providing confidence to advertisers that the ads they pay for reach particular humans (which requires providing at least as much confidence that they reach actual humans) is exactly how the F and the G, but not the AAN in between, make money; detecting bots (or, perhaps more precisely, detecting humans and excluding bots) is obviously part of that.

G is somewhat well-known for it's tools for this, including at least one that is widely used by lots of other actors on the web.


>(or, perhaps more precisely, detecting humans and excluding bots) is obviously part of that

Detecting humans is not the same as detecting bots. The algorithms are different, running a simple "negate human" clause isn't how it works. It's well known Facebook has a bot problem and is now working on addressing it.

>G is somewhat well-known for it's tools for this, including at least one that is widely used by lots of other actors on the web.

G's captcha solution is a great freemium service but any sophisticated attacker will spend the time and money to bypass it. If it solved the bot problem the paid services wouldn't exist (they're definitely not cheap). The other issue with G's service is that the automated garbage is still hitting your compute resources. So even if you captcha you're spending money on automated bad traffic.


lower the barriers of entry for competition.


Network effects are hard to tackle when there are established players. The easiest thing is to wait until Twitter and Facebook lose the coolness factor and the next generation is hungry for something else.


The problem is that if Facebook and Twitter are large enough, they can buy the next thing that becomes cool before it has time to beat them--like Facebook did with Instagram and WhatsApp.


This is the reason why breaking them up will help.


Yeah, that worked so well for for the bells. 8 is now, what 3? And while there was some network effect for telecom, it was nowhere near the same degree as with social networks. No open API's means you're never going to see splits have any desirable effect, as the consumers will simply flock to 1 or 2 platforms.


Network effects could maybe be fought via regulation. Force the companies to let others participate in the networks.


Via open APIs? I like the idea. But then it puts the burden of development/cost on the bigger existing network. If you want to go this way then we should force APIs for all web services, not just social networks.


Network effects are valuable to users too. Are we sure we want to stifle them?


Are most users happy to be locked into a network? What are the positives?


The size of the barrier is unrelated to anonymous human nature, which is what I read is the unfixable problem.


Aren't there already some competing social media platforms out there? Twitter, facebook(Whatsapp and Instagram too), Snapchat, G+ used to be a thing. You may say they do not compete head on to each other, but which one of them was immune to this problem? The fact that any internet based social platform will have a very large connectivity and will provide people a very easy way to communicate to an exponential amount of other people, that fact won't change by lowering barriers. The fact that all the people reading something online neither have the time nor the patience to research everything is not going to change.


There is no barrier. There are OSS alternatives that are just as good as the real thing which you can spin up on an AWS instance in like 15 minutes. The cost to entry is $10 for a domain and a $1 credit card authorization to get started with AWS.


There is no barrier to entry. But there is a huge barrier to grow and establish something new. Whatever a smaller competitor comes up with Facebook can quickly copy and dominate the market just by sheer size. There is a reason why so many startups get acquihired by one of the big companies and shut down their product instead of establishing themselves on the market.


I think it's inevitable that Google EU will be split off from Alphabet and Google US, into a stand-alone entity that the Europeans will regulate and mold as they see fit. It's the only possible positive end to the way Europeans largely view Google today (an abusive, dominant, foreign invader they can't compete with). After this many years, it's obvious the EU can't mount a competitive threat to Google's search monopoly, leaving only the alternatives of trying to cripple/restrict it through increasing targeted taxation & regulation (bad for the EU in the end, cutting their own nose off) or forcing a split-off that retains all the core capability & technology and can better adjust to what the Europeans want it to be (their culture, preferences, etc). The Google US and Google EU would then compete globally. Google EU would be operated from the various offices Google has planted around Europe. It would also immediately become Europe's largest technology company, surpassing SAP. I suspect that it would quickly begin having a ripple effect across the EU's tech landscape, in terms of investment and spurring more development.

Short of search becoming a lot less important soon... we had the baby bells, we'll have the baby Googles. The same goes for Facebook.


In some ways, Google is already more powerful than the EU. I am not totally sure that the EU could ultimately force Google to do anything it didn't want to do; many of the things that have been implemented putatively to limit the Goog have had the (un?)intended consequences of raising the costs and barrier to entry for new competitors and further cementing their dominance...


Google controls what 3 billion people see when they search for things -- what information they are exposed to. This is more control than any government of any sovereign nation in the history of humanity (more than the top 3 governments today combined) has ever had.

And please note that there are well-known processes by governments are held accountable by countries' constitutions and courts. A user of Google has no such recourse.

I don't think it's unreasonable that a company with impact larger than any single government in the history of humanity (yes, even larger than the EU) should have more than a little accountability. (Honestly, I think the Google board of directors should be elected at this point -- just like we elect leaders of governments -- but I know that's going to be a contentious view.)


It will prevent the hatred to be easily propagated to a large population — control the reach of the polluting voices. Scale of Twitter has given a microphone to these voices which they are exploiting. Hope is with the power split, it will just not be a problem in the first place.


I agree. Having trillion dollar valuation companies dominate the internet is akin to letting Standard Oil dominate a resource, or monopolies dominate the emergence of railroads. You have to recognize these technological emergence's when they occur, because typically they will be unlike anything we've seen before. I would argue Social Media is one of those emergence's. Smartphones arguably another.

Ideally, I would like to see tech giants split into 3 companies, with 2/3 residing outside of SV so as not to be subject to the group-think of that region. This coming from someone who grew up in the heart of SV, got into tech, and eventually left.


I think we generally should discourage growth of any company beyond a certain size. I don't see any societal value in these large companies but a lot of disadvantages.


Big companies add huge value to all our wellbeing. 1000 companies with a revenue of ten million dollars can afford to spend little to nothing on research, while one company with a revenue of ten billion dollars can spend a lot. How many startups comes up with new medicines, for example?

Furthermore, you are implying that all that people do has to be done with the societal value in mind. What about the value for individuals? Look at yourself for example. Do you do everything you do for the societal value or to pursue your own separate interests?


Those $10M companies don't do research because they are research. The founders have an idea, they try it out, if it works they get acquired by a big company that doesn't do nearly as much of its own research as it used to because this is easier. Sometimes the big companies deliberately let groups of employees leave to try something, with IP and funding and legal indemnification, then buy them back if it works. So plenty of research actually occurs, and it's the big companies that don't need to be part of funding it.


Sure, but if these big companies with lots of money to spend did not exist, there would be no one to buy the smaller ones.


Would that be so bad? If the small companies funded by other means (let's call that "venture capital") could grow into medium-sized fish themselves instead of being gobbled up by the big fish, that might actually be a healthier ecosystem.


Companies benefit from protections given to them by society. E.g. limited liability, copyright, patents and lots of others. So I think it's prudent to ask ourselves as a society from time to time if these privileges are still useful or maybe should be modified or revoked.

I agree that for a lot of research you need to be of a certain size but I don't think you need to be a multi-hundred-billion dollar company. Apple was pretty innovative in the 2000s when they were much smaller. From casual observation it looks to me that they are doing less innovation now than they did back then.


Well, you don't think the society benefits from the companies? Where else would you get all the stuff we use for everything? Where do you get the factories? Where do you get the machines? Where do you get the software?

If you don't think companies bring societal value, ask yourself what the alternatives are. There were a bunch of attempts in the 20th century, however it never ended good for the population.


Society overall gets varying amounts of utility from companies. Mostly they exist to help their owners harness the wealth creating power of workers. By some methods of accounting, many of them are a drain on society; they centralize wealth and create much bigger external costs (resource depletion, pollution) than they create value (engineered obsolescence, disposable crap, useless products & services or ones that serve artificially created demand)

Car dealers, patent trolls, for-profit health insurers, security-theatrics firms, content-distribution conglomerates (ISPs in the US) all are exploiting regulation or lack thereof to profit at societies expense. A lot of finance industry behavior is about unproductively extracting money from other participants. (for example 401k holders)

It's my opinion we should aggressively curb all counterproductive business models, but they tend to be good at protecting themselves. They either win over the masses through PR and deceit, or use more direct corruption of regulators.


I said "companies of a certain size". Don't make this into a socialism vs capitalism issue. Yes, companies benefit society. But I believe trillion dollar companies are a negative for society. They get way too powerful.


Only if they use their power to more bad things than good things.


Which they do all the time with no consequence.


By the same line of thinking dictatorships are also only bad if they use their power for more bad things than good things.


Yeah, so?


Same with governments and the attempts at state capitalism you're referring to. Everything always is good unless it's bad, the question is how easily can it be constrained when it goes bad (and monitored for the signs of it going bad).


> Big companies add huge value to all our wellbeing. 1000 companies with a revenue of ten million dollars can afford to spend little to nothing on research, while one company with a revenue of ten billion dollars can spend a lot.

There's a lot of space between companies with 10 million dollars in revenue and ones with 10-100 billion where lots of R&D can happen.

Smaller companies can be very good at new product development, and are probably better at supporting those products once they catch on and become established. The really far-out R&D is probably better handled by university research labs than by big-company monopolies and semi-monopolies, which then will spin off small product development companies.

> How many startups comes up with new medicines, for example?

Seems like quite a lot, actually:

https://www.chicagobusiness.com/article/20180509/BLOGS11/180...

http://fortune.com/2016/05/13/big-pharma-biotech-startups/:

> A crucial part of the allure: Pint-size ventures are driving pharma innovation. The majority of drugs approved in recent years originated at smaller ­outfits—64% of them last year, according to HBM Partners, a health care investing firm.


There is only one medicine in the past 100 years that cured a chronic illness besides penicillin and its derivatives. That medicine was developed by a small biotech company named Pharmasset and bought by Gilead.

So to answer your question, one might say that 100% of medicines that are not addictive pills to treat symptoms were developed by startups.

To be fair, a lot of the addictive pills to treat symptoms were also developed by startups.


> How many startups comes up with new medicines, for example?

Lots, and lots try and fail (the ones that succeed, or even make it past the first few big hurdles, often get acquired by the big players, because buying work that's done is less risky than early research...)


> How many startups comes up with new medicines, for example?

Is that a serious question? The answer is plenty, you just don't hear about them because they're bought up once the FDA approves their medicine.


Would a bunch of small companies have invented the smartphone? I feel like you often need to have concentrations of a lot of capital at once for big things to happen.


A bunch of small companies invented the smartphone. Who do you think invented it?


I’d say a bunch of companies tried to invent the smartphone...but Apple actually did


Apple was not nearly the giant it is now when it invented the smartphone. What revolutionary ground-breaking products has that company produced now that it has become a giant?


Exactly. they are kind of coasting on previous innovations made a long time ago. Same for Google: Since search, Gmail, Android and Maps I don't see much innovation coming from there either.


Waymo? Duplex? Photos? Even Chrome is newer than the things you listed.


Waymo got bought, I don't know Duplex, Photos is not exactly innovative, Chrome is just another browser, they existed before


> Chrome is just another browser, they existed before

That's a weak argument. Email existed before Gmail, web search existed before Google Search, maps even online ones, existed before Google Maps, mobile OSes existed before Android.

> Photos is not exactly innovative

That's what I used to think - another photo sharing/storage thingie. But it has all sorts of crazy machine-learning powered features.[1]

1. https://www.csmonitor.com/Technology/2016/0324/How-Google-Ph...


Waymo wasn't bought


It was, actually, from the guy who then tried to sell the tech to Uber and created this massive lawsuit:

https://spectrum.ieee.org/robotics/artificial-intelligence/t...


Even Android was bought, when you think about it.


Apples stock price is over 16x higher today than it was when the iphone was released. By todays metrics, I might call 2007 apple a small company.


That's nonsense. Apple looked at what everybody else was doing and packaged it very well.


Or just a couple of smart, dedicated people in a garage


What's wrong with smartphones?


I think the problem is not with smart phones, but with the ecosystem around them. Eg, Apple, and iTunes. Android, and google docs


I think a good example is cars. Imagine having a Ford car that you could only get repaired at Ford dealerships, that only took Ford bumper stickers or other accessories, and only could get filled up with Ford gas at Ford gas stations. I think most people would agree that would be a problem, and we are getting dangerously close to that with smart phones.


We're going in that direction with cars as well (EVs specifically).


What do you think of Tesla as a business and car company?


I'm not super keen on it, though I like Electric cars. I think that they will, or at least should, transition to being more open as electric cars become more common and in a larger variety.


You only have 2 choices for mobile OS, dominated by Apple and Google.


I'm willing to admit now that I actually liked Windows Phone and wish it had become a viable platform.


I'm on my Lumia 920 right now, due to my Nexus 5x finally biting the dust. I still love the OS. Most of the apps no longer work, my signal is crap, etc. But I still like the OS.


Same here. I was disappointed when they scrapped it.


Silicon Valley is responsible for most of the world's innovation, so as a consumer I'd rather we not dictate where firms wanted to set up shop.


Silicon Valley is not responsible for most of the world's innovation. Not even within tech. If the internet started any one place it was at BBN around Boston. The web started at CERN. Microprocessors started at Bell Labs and Texas Instruments. Cellular phones started at Motorola, Ericsson, Nokia. Amazon's up in Seattle, next door to Microsoft, and even Facebook is an import. SV isn't even a player in biotech, clean tech, materials science, etc. The cult of Silicon Valley is getting ridiculous.


You're not going back far enough though — personal computers, the transistor, microprocessor....


Liquid Crystal Displays at the university of a former fishing town in the north of the UK (my home town) [1].

[1] George William Gray - https://en.wikipedia.org/wiki/George_William_Gray


Um, I did mention Bell Labs (transistors) and TI (microprocessors). PC is a good addition, though.


Facebook is responsible for most of the world's social networking, so as a consumer, I'd rather not dictate how it wants to run its shop.


> as a consumer, I'd rather not dictate how it wants to run its shop.

Think about that statement. Think about it carefully.

You're a consumer. You and people like you are the entire means by which this company earns its money. You really have no interest in how it does so? Even when it has massively significant (statistically speaking--if not you then hundreds of millions of people like you) influence on your life?


Silicon Valley was essentially funded by the US state, and particularly the military. GPS, ARPANET, computers, microwaves, networking, radio transmissions, all of this has been invented for the military, financed and pushed ahead by them. SV isn't a libertarian paradise.

Read the series https://steveblank.com/secret-history/


Which doesn't change my statement in the least. In the internet era, Silicon Valley has been the crown jewel.


I am not sure, but sometimes I think splitting the companies is not even what we need. Instead, companies with > $1 billion annual revenue should be bound by law to built federated platforms and provide the software to participate in such a network.

I mean, you could split up Google into Search, Youtube, and Mail. But in the end, Youtube would still use Analytics and provide Search with all the data they need for their Ad business and Mail would use the Ad tech from Search. With Facebook, it is mostly the same story.

Federated Systems, on the other hand, let you build a distributed network of providers where no provider has all the data in one place, and you can choose which provider you want to trust (or if you want to be provider yourself).


Instead, companies with > $1 billion annual revenue...

That would just turn the issue in to an accounting problem. Facebook would license their logo from Facebook Logo Inc, the like button from Thumbs Up Ltd, their servers from Facebook Servers PLC, and so on. Every one of the companies would make, at most, $999,999,999 in revenue.

A much simpler and easier legal solution would be to tax companies based on the number of users (based on an independent audit). For example, the first ten million are free (yay startups!) and then the company has to start paying $1 per user for the next 100 million, $5 each for the next 500 million, and $10 each user for the rest. Companies would be encouraged to actively limit the size of their platform by ejecting users and bots who don't make them any money, or to charge users for access.


> That would just turn the issue in to an accounting problem.

...A type of problem which government agencies the world around have extremely powerful tools to manage and prevent abuse in. Sure, abuse still exists (some of it significant--see "too big to fail" etc.), but such oversight is massively better than nothing, especially where direct to consumer social media/marketing platforms are concerned.

> A much simpler and easier legal solution would be to tax companies based on the number of users (based on an independent audit)

That's just another metric upon which you have to trust an oversight agency. Users might be a better metric than revenue, but it's still going to result in highly imperfect (but, again, better than nothing) regulation.


Because only rich people should get access to computing resources.


Would you say that bread should be free because it's not fair that only rich people should have bread? How about Ferraris? Houses? Why should social media be different? Note that I did suggest some smaller social media services should be free of this tax, so they could still operate free for user if they chose to. Social tools are useful. It just doesn't have to be a huge network run by a single company.

Besides, there's nothing stopping social media companies charging right now other than the fact they make more money by not charging, arguably to the detriment of everyone in society. Who do you think should be paying to solve that problem? No one?


I think you are on the right track, but we can used more advanced distributed systems rather than just federation. For example build off of things like IPFS or Ethereum etc. or custom protocol networks. Then that becomes the platform and network that companies provide physical infrastructure for and build off of. Using more advanced decentralized technolpgies means we would not need to trust all of our data to one entity and also allows for better collaboration on the platform level.


Why do you trust governments so much?


To be fair, all sides trust the government.

We're trusting the government to go in and properly split up a company. Or you're trusting the government to properly regulate a company.

The people who don't trust the government, don't want the government to take ANY action. They believe the market will solve the problem.


So all sides trust the government except for the ones that don't


And the ones that don't, have not stated their position in this discussion.

So all sides presenting in this discussion, trust the government.


> To be fair, all sides trust the government.

> The people who don't trust the government, don't want the government to take ANY action

Which is it?


You don't have to have a lot of trust in the government, just more than you do in corporations. That's not a terribly high bar.


> You don't have to have a lot of trust in the government, just more than you do in corporations.

Corporations (sans legal monopoly) have power over their constituents insofar as their products are desirable.

Governments have power over their constituents with physical force.

It is 1000x easier for me to delete my Facebook account than for me to move to another country and change my citizenship.


For what it's worth there are some corporations I'd trust a lot more than the government.


See how stable that remains over time however, your data is out there forever, even when $CORP turns evil.


Where did I say something about 'trust' and 'government'? Seriously, I don't get the connection.


Your suggestion that large companies ought to be "bound by law" implies regulations by the government. Whether or not that is your intention is still up in the air. However, if the government owns the regulation then they get to set the standards of what a big company is on a per criteria basis.

If the government is being attacked by a specific 'group', the government will then have the power to impose specific, targeted regulations on that group's web presence. Without the legislation, this is a matter of freedom of speech and is quickly resolved by the first amendment. It would be incredibly difficult to fight that off.

However, with legislation in place that affords the government the power to impose such regulations, they can quickly and most importantly (read terrifyingly) silence people on the web.


Platforms are not people. Regulating the way that companies offer their services, tax them, and how they are allowed to be structured is not a limitation of free speech.

We already do it with every other industry, social media only enjoys a lassaiz-faire regulation environment because of how new it is, the law hasn't caught up.


"bound by law"


That doesn't imply that I 'trust the government'. I mean, trust is not binary. I trust politicians as far as I think that what they do fits their agenda. So I don't have any trust in 'the government' without context.

The reason why I think a law would be the appropriate solution is

- that the companies have no interest in changing anything towards an (e.g., privacy friendlier system) as it would endanger their market position and therefore their profits

- in the current situation, the majority of people has few alternatives, so they can't 'buy their facebook somewhere else'

- for new competitors, it is quite hard to compete even with a single product not to mention a whole set of them

So it doesn't look like the market will find a solution anytime soon. I am not a fan of too much regulation either. Many laws have problems addressing the real issue and take too long until their are changed appropriately, but most laws cause costs anyway (not just financially).


Who in the world would trust Donald Trump to manage search across the entire internet?

You'd think that after the Trump administration the nation would have a greater appreciation for federalism and a smaller administrative state.


In 1940, Lee de Forest, (semi-self-proclaimed) Father of Radio, sent an open letter to the National Association of Broadcasters in which he demanded: "What have you done with my child, the radio broadcast? You have debased this child, dressed him in rags of ragtime, tatters of jive and boogie-woogie."


> “If you put a drop of love into Twitter it seems to decay but if you put in a drop of hatred you feel it actually propagates much more strongly. And you wonder: ‘Well is that because of the way that Twitter as a medium has been built?’”

No, it's human nature. Stop blaming companies for the imperfections of the human race. If you put us into large groups, we will inevitably gossip, bicker, argue, and kill each other. We've been doing it literally since we were banging rocks and sticks together.

The delusion that technology feeds us, and it is especially potent in America, is that there is some higher power leading us all to be terrible people online, and that technology should be able to fix the problem.

But technology can't fix the problem. We are the problem. We also evolve incredibly slowly, and we're competing against technology that evolves in a matter of decades. We are too inflexible as a species to be compatible with the social effects of technology.


This is incredibly fatalistic.

Why can't technology be created with human nature in mind? Forms of democracy with checks and balances became so dominant because it acknowledged the deficiencies in human nature.

We can certainly blame companies for producing bad/harmful technology just like we can blame forms of government for producing bad outcomes for its citizens.

---

On a side note, based on influence and entrenchment, comparing FAANG companies to governments actually seems more appropriate that I initially thought.


> Forms of democracy with checks and balances became so dominant because it acknowledged the deficiencies in human nature.

And those systems have tons of flaws as well. Have you seen the voter turn out in the U.S.? How many Americans are making informed voting decisions.

You're shifting the blame back to the technology. I get it, as a technologist we'd like to think it must be a mistake in the product. But the reality is that our idealistic notions of how the world should be do not match the capabilities of human nature.


Compare the U.S. government to Feudalism or a similar form of government--not the U.S. government to Utopia. You're not seeing the point.

This is actually the exact opposite of idealism. Try to account for the reality of human nature in systems and they will have much better results for the humans they affect.


There are both human and technological problems at play here.

The human problems, as you stated, have long been well known. These technological communications problems are new.

It seems pretty self-evident that the character limits, the incoherent way conversations are rendered, and how easy it is to take tweets out of context are all detrimental to the quality of discourse and understanding between users.


"But social media, Berners-Lee said, was still being used to propagate hate."

This is one of the side effects of an open web, and one I feel we should live with in the interests of freedom. The only alternative of which I am aware is to police speech through some kind of authority. I'm not sure that there is an authority in existence that I would trust with that level of power.

In a perfect world, we would be able to de-platform those that incite violence against others (for any reason), but we can't even come to an agreement on what that means. Some people argue that criticizing a marginalized group is hate speech and further argue that hate speech is violence. Should some people be shielded from valid criticism because they identify as a certain group?

It sounds like a simple concept, but the more you explore it, the less simple it becomes.


>This is one of the side effects of an open web, and one I feel we should live with in the interests of freedom. The only alternative of which I am aware is to police speech through some kind of authority.

I strongly believe this is nor correct. You can mitigate a lot of this stuff through better design, without censorship. Most important point: Open Web is the idea that people can find info they're looking for, not that anyone on the web can bombard everyone else with information. Modern social media is optimized for the latter. Algorithms chose for you, and they chose badly, and you have no control over it.

Jaron Lanier has a lot of talks about incentives and why current social media is bad for you. His hypotheses seem plausible.


That just means the person writing the algorithm becomes the authority. As we recently saw with Amazon's hiring practices, algorithms have bias as well. Granted, a better designed system might not have made this mistake, but since humans designed it, can it ever be perfectly unbiased?

Without actual censorship (which I'm against) the offending material still exists, but would only be found by those who seek it. I do like that approach personally, but I don't think it would be enough to satisfy many people who prefer to eradicate offensive speech through authoritarian measures.


>Algorithms chose for you, and they chose badly.

The algorithms don't choose badly. They actually choose very well. It's just their optimization function is different -- their optimization goal is more clicks and more ad revenue, not the society's good.

Click-based revenue model encourages pretty dark sides of humanity.


>In a perfect world, we would be able to de-platform those that incite violence against others (for any reason)

Should we really do it for any reason? Are there not some groups of which a certain level of violence (enforced by the state) is good? I can think of many groups of people who engage in actions which we agree it is good to respond with some violence to enslave them in cages. Well, that is one interpretation of what we do when we put someone in prison, but I think we cannot divorce imprisoning someone or putting other punishments on them that are backed by imprisonment from the threat of violence that is put on prisoners to comply with the rules of imprisonment.

At the core, laws are enforced by violence or threat thereof and are largely based upon subjective morals. Talk about adding laws is thus talk about bringing violence to some group of people. And sometimes (say laws against murderer or theft), it is required for us to have a civilized society.


There's nothing wrong with inciting or carrying out justified violence—which is to say, a violent response to similar violence practiced by others. Your examples of murder and theft, for example. A murderer cannot rationally object to being put to death; a thief cannot argue against being fined without self-contradiction.

The problem is unjustified violence. And I agree with you that most proposals for new laws (including the subject of this article) are advocating for unjustified violence.


The problems arises not in the easy to agree laws (murder, theft) or the easy to condemn laws (ban on homosexuality), but with the ones which are debatable on if responding with violence is justified. Many would say that violence against pot dealers isn't justified, but what about heroin dealers? What about opiate dealers? Pot dealers who sell to teens?


If it's "debatable" then it isn't justified. The only universal, objective justification for a violent response is that the other person already did the same thing, and thus cannot claim that responding in kind is wrong without simultaneously condemning their own actions. If you claim that your response is justified on subjective grounds then they can just as easily claim that their actions were justified from their subjective point of view, at which point the whole situation devolves into a case of might-makes-right and justification becomes irrelevant.

None of your "debatable" examples would be justified because the actions you are proposing a violent response to are not themselves violent. Even when the original action is violent, the response must be proportional: applying the death penalty for theft would not be justified because it isn't a response in kind.


I believe that inciting violence against others is wrong, but at no point did I argue that we shouldn't have laws against aggressive behavior. Those are two different concepts.


I think laws are a subset of inciting violence against others. It is more controlled than random mob violence, but it is still violence. For example, some extreme groups support adding laws that will require minorities to leave the country or be imprisoned. That they argue for doing this through the legal framework instead of random mob violence doesn't excuse it in any fashion.


I respectfully disagree that advocating for changes in the legal framework constitutes inciting violence, unless you use the most broad and generic definition of advocating violence. Thanks for the civil discourse though.


Complaining about an open web is just a proxy for complaining about establishment ideas being challenged. You see in Britain with the complaints against the Leave Campaign; you see it in the United States with the rise of Trump; you see it in the European Union with the complaints against far-right candidates. The governing elites don't like it when you challenge their assumptions, and instead of making an argument about their positions or governing better, they complain about Facebook, Google, and Twitter.

I'm American and racists have always been here, Trump isn't new. And in Europe, it isn't insane to want to control your borders and not have rules made in Brussels that you must follow made by bureaucrats that aren't accountable to your nation. It's a crazy idea I know, but politicians have problems with these notions and blame the tools the opposition used to organize instead of being better.

Facebook, Google, and Twitter have problems and they need some reforms, but the rise of companies is not nearly the sinister event people make it out to be.


People seem to be downvoting you without replying, which doesn't seem very helpful.

I'm sure there are some people who fit your first paragraph, but I don't think that's the major issue.

The ideal of free speech is that people can honestly disagree, and state their positions, a discuss the issues. Then the best ideas will win out. But our current social media environment does the exact opposite.

It appears that the way social media now works is that it drives people farther apart. The Red team is moving farther to the right, the Blue team is moving farther to the left, and very little real discussion of issues is happening between them. Much of the cause of that seems to be built into the way our social media is designed.


The "red vs blue team" thing is a US thing, and it is created by the structure of our political (and, especially, our electoral) system. First-past-the-post voting, combined with widely open primaries, result in both parties gradually drifting apart over time, and dragging their constituencies with them. That this erupts into verbal (and, occasionally, non-verbal) violence is not surprising, but what you see on social media is the symptom, not the cause.


I suggest the title be changed to remove “Father of Web”.

“Tim Berners-Lee says…” is better.


I actually have a big issue when the news runs stories like this to begin with. They would do the same btw if Woz said it or any other well known person. Sure the fact they can tag on 'father of the web' gives a strong pull but that is part of the problem. His thoughts on this aren't any more important than thousands of others that nobody has ever heard of. This is a total 'news as entertainment' story. And this has nothing to do with whether I agree or disagree with what he is saying. I just don't like prominence given solely based on some prior achievement which quite frankly has no bearing on what he is saying today.


I agree that his opinion should have no bearing on the way that the modern WWW is run - but considering he invented the platform I think his opinion is interesting and well worth hearing when it didn't turn out the way he originally envisaged.


> I think his opinion is interesting

Hence 'entertainment', right?

> when it didn't turn out the way he originally envisaged

Sure but who cares if that is the case? And he didn't imagine and invent the modern web nor did he even popularize it or put much effort into making it what is is today.

By his statements I take it that he is jealous that he feels that he has not played a bigger role in what happened after his 'invention' (which was based on a great deal of work prior to his involvement). Many of us used the arpanet in the 70's not to mention the US Government's involvement. Honestly (and nobody will agree I am sure) if there was no conflict with Russia the Internet would not exist (most likely).

And I am not seeing what he had a vision for any more than solving some simple issue he saw which was:

> Instead, it was hard work, the experience of working in computer science and an attempt to overcome the frustrations of trying to share information with colleagues and students

That is not the web today or even close to it and for that matter it wasn't the web even in 2000 or 1998.


Yeah. Honestly after his advocacy for the EME/DRM stuff it's hard for me not to be skeptical out-of-hand to any of his suggestions.


It's sad he didn't see that EME/DRM is ultimately a tool that massively benefits incumbents to the determent of literally everybody else.


Make that "Sir Tim Berners-Lee" and I agree.


"Tim 'supporter of DRM and now mostly trying to just stay relevant' Berners-Lee"


Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: