Families row over home PC access and, in essence, this isn't any different from the arguments a decade or so ago about who gets the tv remote control. Consider it as a resource allocation problem, and there are two solutions. One is that you distribute the resource over space: Instead of arguing over which channel to watch, you have multiple tvs. The other is you distribute the resource over time: You watch one channel now, and use the vcr to record the other. Both of these solutions create opportunities. The space solution means there are multiple tvs in the house which means instead of the tv having to occupy the safe middle-ground, you can have niche tvs. That is, rather than a tv which is semi-portable, semi-big, semi-expensive, etc, you can afford to have one home cinema tv optimised for picture and sound, another tv optimised for portability, another for cheapness and so on. Likewise, the time solution creates opportunities: the vcr stopped the tv from being a single device that existed in everyone's home controlled by the broadcaster, and turned every person into their own scheduler. The intermediation of broadcast allowed video hire, home movies, and now new iterations like dvd rental and pvrs.
(Not forgetting, of course, that gradual iteration to new models often means old problems reemerge. Dvds have good picture quality but no random access, and recordable dvds are only now getting affordable. Digital cameras and broadband are moving into the home movie niche, with distribution done with files and email rather than handing cassettes over.)
What does this mean for the home pc? Let's try and find both time- and space-based solutions.
The time-based solution for resource allocation is difficult: The pc is being used for homework (you can't delay that) or chatting with friends (you can't record and play back chats). Online content and interactivity means you can't save what's on the pc for later, although using an rss aggregator (or other offline reader) could help. When I had a Palm, I used to read the Guardian and the Onion on the tube to work, but that's still one-way. Okay, let's assume that the primary use of the net is for socialising. Synchronous communication (IM, mainly, although messageboards are sometimes synchronous for time-units of a few hours) can move easily onto the mobile, or is there already: text messaging, and I wouldn't be surprised to see text chatrooms for small groups soon. Asynchronous communication (I don't really mean asynchronous, I mean inscribed), like some contexts of messageboards, LiveJournal, blogs, news sites with commenting, and email, these fit into the offline reader model, if there's a way to make comments offline and have them sync to the server later.
Thinking pragmatically for a second, if the weblog comment scripts answered to an API in the same way as the weblog posting scripts answer to posting and editing APIs, you could write an offline commenting system. Using the BlogLines API, you could manage your subscriptions on your desktop pc at any time (when your sister isn't doing her homework), and have the same subs list on your mobile phone.
Okay, this has some way to go as an idea because it depends on standards and people evolving software. The dark side of network effects are in action here. Getting a large group of people to work in the same direction is like trying to start a whirlpool in a bathtub of custard using a teaspoon. You have to embed your development, and so this solution will take time and compromise. I think we'll move in this direction though. It's a software solution, and somebody's bound to try it.
The space-based solution: The problem is that home pcs are expensive. They've got into people's homes because they're so cheap--and you pay more for broadband in a year than you do for an absolutely bottom-of-the-range pc. People can't afford multiple pcs in the home and, besides, don't have the expertise to set up a home network to share the connection (although wireless internet sharing is making this easier).
The thing is, you don't need to share everything. Just as tvs diverged to multiple specialities, you don't need word processing on every computer in the home. You don't need games. You probably don't need a colour screen or a full keyboard. You just need the most common, least able to be time-shifted content: communication again. Email, IM, blogs (well, journals really): this is all done through a web browser. Okay, so you need web browsing round the home.
Here's one easy way to do it: Something that looks like a usb thumb drive slots into the back of the pc. It runs some software which is kind of like internet sharing but does a bit more. It has a bluetooth (or 802.11b, now that's way cheaper) connection to the viewing and input device. This device has a low quality screen and a plug-in keyboard. Other models don't have a screen, they just project on the wall (secret: your house is full of vertical and horizontal surfaces you don't use), perhaps the keyboard is projected too. There's precious little processing power in this device--no Flash games, no movies, no incremental download of pages even. Perhaps the web content is rendered using the internet sharing software. The scenario is that I use a web-based IM client and reply to my email using this device (but mainly: chat), and my connection piggy-backs on the existing internet connection of the home PC, on which somebody else is doing their homeword. If you could make this device for $50, you'd sell millions.
Given you've got a home web browser device, this leads to a second bifurcation. Low quality screens will be used for email and IM. Higher quality screens have plausible deniability ("but I just want to see the screen better and not get a headache!") but actually will be used to browse dirty pictures. This means you have another use: parental filters. Sites like Hotmail and LiveJournal and services like AIM and MSN will be open, everything else is shut down. Unlike existing filters which don't work (the pc is capable of much more), this device is hardware-incapable of rendering difficult sites. The filters are a side-effect, by design almost.
Further business model for the device: Have it show ads in the bottom half of the screen, and let Gmail give them away. The Eudora email client has a sponsored mode, so why not. Or: Sell it as an add-on to the Nintendo DS, or any mobile phone with a sufficient OS and bluetooth. Even though it might be a fully-functional web browser, market it to answer a single need: talk with your friends anytime and anywhere at home, just like Apple made the iPod answer a single need: listen to your music always. The problem for the space-based solution is not money or technology, it's design and business. In which case, I rate this idea ambitiously: 6 months to release from now, 18 months in total to become popular in non-geek families.