Hacker News Reader

> There should be no real reason I can't point my iOS device at my own infrastructure to co-ordinate sync

Well there is a real reason, you never own the iphone you have. You never have control of the device, Apple does. They have final say in what can be built to run on it, what can run on it, what you can install, what OS you can run, whether you get updates, and what the phone does. You have no control over it, at all.

So don't act surprised when Apple takes advantage of absolute power to screw you over.

At that, I know owncloud exists for iOS, so you could just use that, at least for the bullet points you mentioned, not so much for the sensors-aware network of devices.

>>There should be no real reason I can't point my iOS device at my own infrastructure to co-ordinate sync, backup and sharing of all my stuff (instead of iCloud).

I want this as much as you and everyone else who upvoted you, but the problem is there is no obvious business model in it. Sure, techies will pay (app.net?), but the general population doesn't care enough to put out money for this.

There are plenty of ways of making money, including asking your users to pay. Take a look at the number of 'personal cloud' crowdwfunding campaigns to see how much latent demand there is for things like this. Most of those products don't even exist yet.

I think it's a mistake to assume that nobody will pay for such services given the vast number of people online now. It's just a matter of appropriate market segmentation [1]. Business models emerge as people begin to use things, and we should remember that Google didn't have one for a long time (neither did Facebook).

[1] The teenage version of me didn't give a damn and only used free things but the adult version of me has a much better understanding of the trade-offs I'm making (and I also have disposable income now). I don't wish to become a sysadmin, which means I have very few options.

All good points. I think it's certainly feasible in the future, but not today. There are challenges besides technical and business that need to be solved too. For instance, you have a personal cloud server you set up that runs all your apps, and you have an Instagram-like app for your photos. How do others get to view your photos? There needs to be an aggregator that runs across servers to pull all the content and puts it into a newsfeed, with an easy to access URL/app for your friends to open and view your photos. How do we get this? With a centralized server?

Isn't that just a technical challenge? In this world view, you are your own central server and everyone else is their own. Let's take your example and run with it (bear in mind I'm ignoring some rough edges here). You share your instagram-style pics into an app on your personal server to begin with and then from there they get federated out to people who've decided to 'follow' you (or those you have allowed to). Practically speaking, the pic (or rather access to it) is given to the app on the personal servers of those followers. Their server then constructs a newsfeed of all the incoming pics and presents it to their user (via an app or web-interface or whatever). This doesn't happen on the mobile device itself but on the 'personal cloud' of the user. There are many things I've glossed over but there doesn't have to be a central server in this scenario and everything can be encrypted end-to-end.

That's why this is really a distributed systems problem. In order for an app-developer to build such an app, the deeper problems around deployment, sync, and identity/connectivity need to be resolved. Developers just need something to target that provides the appropriate components. That's what the toolstack I previously linked is trying to solve. Without that, we're stuck with requiring someone else to provide the central system/aggregator. Clearly this is a non-trivial problem but if it can be cracked then it'll be as impactful as the opening up of the Web was.

>>Isn't that just a technical challenge?

I supposed it could be, or a UI problem....it all depends on the architecture of the code and how you choose to define the problem. Regardless, it's all certainly feasible and possibly will be done sooner or later, it's just a matter of when. Are you familiar with the Diaspora project from a few years back?

https://diasporafoundation.org/

I wonder how this type of system would get paid for. If we're all paying for our private clouds and we use this 'Distribugram', who pays for server to server transfers? Is it free to the users (handled by provider peering/transit agreements)? Do users pay? Puller or pusher? How does this impact people with many followers/followees? Do they pay more?

Thinking about this, I wonder if the economics would work given the ad-supporter/loss-leading aspects of some of these services. Still, interesting.

Yeah, this is kind of a serious problem. People want to use popular, or at least well-advertised platforms. People also don't want to pay any money if they can help it. So we end up with advertising-funded spy-clouds and addiction-funded IAPs as dominant business models.

What I'd like to see, and hardly anyone is talking about, is the cooperative model for this. Servers run by a community of interest; paid staff to manage the things which really require central clearing (anti-spam and anti-abuse in particular) and develop the system; users paying on an ongoing basis and being given a voice and a vote in the way the system is run. Federation with similar systems.

Bits of this exist already, but only on a very small scale and with the political/organisational infrastructure - basically individuals running Minecraft servers and the like.

I suppose I could gather thousands of recruits for something like this if I said "blockchain" a lot, but in truth it's the opposite: a system that requires users to trust individual other humans.

That sounds like a commune

Ok, I am a techie, but I did pay for that kind of setup. I bought Synology NAS box and access my files, pictures, music, videos with Synology apps and couple of other apps built on open standards (WebDAV), sync my notes through WebDAV and so on. I am building another small Linux server to do things I cannot do on Synology, primarily back up of my social streams. This takes quite a bit of time and expertise and I would gladly pay my way through some parts of this setup.

> The OP conveniently ignores all the other mobile OSs out there such as Windows, Firefox, Ubuntu and also people like Fairphone. It's pretty lazy to just claim we need more, when what he really wants is one more (open) winner.

Yeah, I'm not clear how excited he'd be if Microsoft was a major player in mobile. I think if that were the case, we'd just get the same article but with 'triopoly' instead.

Also, did he say custom boot animations? I don't think that's what is holding us back... better examples would strengthen his argument

Are you talking about having a thin client that talks to a private cloud instead of each client's data residing on the vendor's server? One of the hurdles is that the server side of apps require more than reading and writing files. Full text search, recommender/ml jobs etc. Apps don't just need file systems, they need CPUs.

If an app could be installed on a server and sandboxed (so for example multiple apps could share an ElasticSearch instance without being aware of each other) this would be a big step forward.

I'm only vaguely familiar with it, but Brad Fitzpatrick has a project called Camlistore [https://camlistore.org/]. Probably worth a look. It seems he's trying to tackle exactly these kinds of problems with that project.