Comment by sbov | Hacker News Reader

Comment by sbov | Hacker News Reader

Parent Full thread

How tightly bound we are to just a few large tech companies is... unfortunate.

I would love it if it were easier for the common person to host their own stuff. Between all the concerns involved (security, hardware, demand, etc.) I'm not sure how feasible it is. I've given it a lot of thought, and have slowly unwound my custom tech setups - from domains, to emails, to hardware. Mostly for the "bus" factor. When my dad died, my mom let his computer sit untouched for years. At some point I transferred the files on it to my computer so they wouldn't disappear forever. But still, a decade later, and it's just sitting there. She never bothers to turn it on.

I have a family, and my wife is not technical in the slightest. If I were to be gone tomorrow, there's no way she would have a clue as to what to do. My #1 barometer for everything I use in life is how easy it is for her to use things.

It's interesting to me how temporary and inaccessible our digital lives actually are.

This is a great point, because I think it shows that people who are trying to redefine the contours of the standard free speech conversation (which had historically been about freedom from government imposed limits) are attempting to solve the wrong problem.

The problem is dependence on huge monolithic tech platforms, such that we feel that the societal value of free expression is threatened by the standard forms moderation that we had accepted as normal for most of the history of the internet.

Now, the ability to moderate carries a significant amount of power in a way that didn't use to be the case. And some people, looking over the available options and the effort required to pursue them, would now prefer to resign themselves to permanent dependence on huge monolithic platforms and just focus on making sure those platforms are not engaging in moderation.

But this is hugely problematic for a number of reasons, not the least of which is that moderation at least sometimes offers a genuine good to users because toxic, manipulative, illegal behaviors that undermine the platform are removed. There's a "societal value" in having flourishing discourse that isn't compromised by the toxicity that would naturally occur without moderation. And there is good reason to not want to make that sacrifice.

Either side offers a distasteful sacrifice. So maybe the answer is to try and bypass the dilemma by making non-centralized communication more accessible, so that people like your wife (or my mom, or my dad's entire side of the family) don't have to make that kind of choice.

There's also a very blurry line between a common carrier and a publisher. Social media platforms don't want to be liable for content published on their site, but they also want the right to delete content at their own discretion. Somehow, the courts have granted them this wish.

The USPS has a similar monopoly to YouTube, but they're obliged to transport any printed material that is legal. There's no hidden algorithm that shreds letters containing the wrong keywords. There's no team of Indian outsourcers searching your mail for content that breaches an arbitrary set of rules. Nobody but a court can prohibit you from using their service.

I think we collectively need to decide whether social media services are publishers or common carriers. The present ambiguity over their status gives them an unreasonable degree of power with unreasonably little responsibility.

> The USPS has a similar monopoly to YouTube

No, the USPS has a legally-protected monopoly, where alternative services are prohibited from competing with certain core functions.

Alternative web video distribution services are not prohibited, and there are several that specialize in content that is unwelcome on YouTube.

> The present ambiguity over their status

What ambiguity? They are active distribution agents (neither publishers nor common carriers, in that regard, though they are often also publishers, more like bookstores).

It only seems “ambiguous” when framed with a false dichotomy.

The ambiguity is that facebook and youtube can't be sued for hosting illegal content. They are not responsible for user created content.

Yet, even though they have many of the legal protections of being a pseudo common carrier, they have none of the obligations, such as the obligation to treat all content equally.

For example, it is illegal for your phone provider to eaves drop on your conversations and censor your phone conversations.

But it is NOT illegal for facebook to do the same thing. Why are they treated differently?

> The ambiguity is that facebook and youtube can't be sued for hosting illegal content.

Taking that as unvarnished truth [0], for the sake of argument, that's not at all an ambiguity, it's a crystal clear status.

> Yet, even though they have many of the legal protections of being a pseudo common carrier, they have none of the obligations, such as the obligation to treat all content equally.

Again, taking that as true for the sake of argument, its not an ambiguity, but a crystal clear status.

Its a conflict with the false dichotomy of the idea that the only two roles in information dissemination are fully-liable non-neutral publisher and fully-immunized neutral common carrier. That model isn't true offline, and it's not true online.

[0] Which it isn't, because while the CDA and DMCA (and others) provide some limited/conditional protection, they are not a total bar to liability; but the still the status is not ambiguous, merely distinct.

You are comparing a postal service (1:1) with a broadcasting service (1:many). ISPs should be neutral carriers, but websites should be free to offer whatever services and terms they like. Given equal services, supply and demand should kill the ones with the most unfair ToS. The question of liability is more difficult, it might help if users had to register as real persons or companies (i.e. providing ID).

Why do you think that the distinction between 1:1 and 1:many matters?

I see no reason why 1:many platforms shouldn't have the same protections as a 1:1 platform.

Because there is no established common carrier equivalent for a 1:many content service. So we are given the false choice of categorizing content platforms as either "common carrier" or "publisher", when they are actually in their own category and deserve their own legislation.

It is in the public interest to prohibit moderation by 1:1 common carriers. It is not in the public interest to prohibit moderation by '1:many common carriers'. But by virtue of being 'simple' carriers, they should both be free from liability for the content.

Note, however, I don't think it is very good to have these companies moderating themselves, but it's better than nothing at the moment. I'd propose legislating the scope of the moderation according to standards set by some internationally recognized ethics committee (not a government decreed scope, they are likely just as biased as a company).

Alternatively, platforms could offer something like a complex filter system that allows users to implement their own moderation policies. But this would increase the burden (costs) of the provider.

Why should a for profit company be obligated to do something for you. Setup an irc or any xmpp server and knock your self out, oh you want to publish to people that find it offensive ! Too bad Ps you is not personally you more the wiser position

They should be obligated to do something for the same reason that it is illegal for a phone company to eavesdrop on your phone calls and censor certain words.

There are tons of protections, already in place, on large networks.

The USPS isn't unregulated though, mail fraud can be prosecuted.

Because they arbitrarily decide they don't like it, or because it is actually illegal?

Are there any companies out there trying to make it easy for consumers to legitimately own/host all their own digital content?

If the price was reasonable people would go for it, as long as setup and maintenance is transparent (that's what the company's services help you achieve).

If the networks were federated .. like e.g. Mastodon ...

How would such a company avoid having the choice of not selling to people they don't want to be associated with.

And then we come to hosting which is a gray zone in most retail contracts actually offering usable upload mostly because of bandwidth concerns and do we really trust the likes of Comcast not to play dirty when put under external pressure in regards to unpopular speech.

I think we've gone well past the equilibrium point between ease of use and computer literacy, to the point that we are now disempowering users rather than helping them. If someone record a video on their phone and want to share it we know it's just a file on a disk, but most users don't, to them it's just a video that appears in a list in an app. We know a file can be copied, modified, shared, etc, but all they can do is what the share button on the phone's UI allows them to do, like upload to youtube. Even if they had a youtube clone owned by themselves or a friend they wouldn't know how to upload it.

I saw it with my mother going through photos she took on her digital camera, she's a slave to the software that came with the camera and the features it provides. She doesn't access the file, she goes through a UI to view the images and manipulate them. If she wants to post them to facebook she does that through the UI. When the camera dies and is replaced she has to relearn a different software package whereas if she'd managed her files through explorer like we probably do then the knowledge would be transferable.

Unless we start teaching computer fundamentals better and expect people to apply that knowledge to do things then we are sliding head first into the world of digital serfdom.

Teaching fundamentals won't change much if the software with the good features is never developed. The app business model is hard on things that are for pro users, and this group is shrinking because it is too much trouble to do things right.

Power users are being wrestled to the floor by locking bootloaders, not releasing source code and banning apps from the popular web stores. Eventually, they get fed up and just join the normal consumers using 3rd party web-apps with biased features.

Indeed, this!

It's true that the process of setting up and managing websites can be daunting. Even using shared hosting, let alone on a VPS. And then there's the additional complexity of HTTPS, which is now pretty much the norm (even when not really needed, with no account creation or payment processing).

Using Micah Lee's OnionShare, one can easily create Tor onion sites.[0] They can be transient (for file-sharing) or persistent. It's based on his txtorcon library,[1] specifically TCPHiddenServiceEndpoint.[2]

That still might be a little intimidating, but hey.

0) https://onionshare.org/

1) https://github.com/meejah/txtorcon

2) https://txtorcon.readthedocs.io/en/latest/txtorcon-endpoints...

I don't believe OnionShare uses txtorcon at all, actually. And I wrote txtorcon, not Micah ;) but they're both decent options for publishing depending on your use-cases.

people arguing (out of principal or to push their views) for some kind of freedom to publish on a for profit company's site are simply seeking that company's audience to spread their message

Sure... Since it is virtually impossible (for normal folks) to publish any other way.

True. It's amazingly hard to get attention, otherwise.

Maybe something like a RPi, made to just be a web server. Something pretty dumb. Also, possibly legislation that forces ISPs to let residential customers use their outbound bandwidth the way they want, within the bounds of the law?

I dunno...

Sure, Pi are great. And if you setup a Tor onion service, the ISP can't block access. Unless they block Tor, of course.

And people don't even need Tor to access it. When high security isn't necessary, it's fine to publish Tor2web links, such as https://m_facebookcorewwwi.onion.link/

Yes, I've posted about this a few times - but it is too big of a project and not novel enough to get many skilled engineers interested, I think: We need a Zoho-like replacement for Google Apps that is open source, and has a host-agnostic ecosystem around it. Someone should be able to launch the apps in their own docker containers, with a button click on AWS, and evolve features like protonail et al. Not all of Google's apps need replacing, but it should include custom DNS, calendars, collaborative editing of documents, email, browser-syncing, storage, and possibly even a mobile data management and decentralized chat! Who wants to start building a team to work on it? :)

Have a look at https://sandstorm.io/

Seems kind if close to my description... Need to try it out. Wonder why I never hear of people using this; maybe my ignorance.

Along with proactively making my wishes concerning life support known to my family, I have made certain that they understand that a close friend is to be responsible for dealing with my data. He is capable of purging private data and filtering it down to family photos and the like. He knows to physically destroy all storage devices after processing the data and my family knows that he has accepted this responsibility.

I make sure to mention this to all concerned parties every couple of years so that everyone is clear on the matter. This doesn't include online services (email, etc), but it's enough for me.

So what do you think of LBRY:

https://lbry.io/

"LBRY is a free, open, and community-run digital marketplace.

You own your data. You control the network. Indeed, you are the network.

Hollywood films, college lessons, amazing streamers and more are on the first media network ruled by you."

Reply on news.ycombinator.com