What is the porn aspect? (Made in Amsterdam? )

Dude, I know Amsterdam inside and out

I was just wondering how you could prevent objectionable content being injected into others' devices...s'all.

Hahaha. I rather see a hot woman than someone tortured. Maybe I'm too Dutch...

Good call tho. Resumee, the sharing of data requires ACL on both input as well as output layer. This can be based on tagging. Eventually, people are going to use a feature like this for bad stuff; you see that with spam on e-mail, deep linking goatse.cx with UBB, etc. And simply banning someone doesn't work well, so we tend to create pattern regexps or block everyone by default.

Btw, I was also thinking about 'friends of friends', 'forwarding'. This can be very useful for networking in some specific concepts mentioned. At least #2.

Hi, I was one of the participants working on the #1 concept, so I can share some additional (my) thoughts in regards to it. Naturally there are as many opinions and viewpoints as there are people.

The original brainstorm name for it was "Content flu". Similar to how the real flu spreads around populations, through physical contact, your content could spread around in the same manner. Every person (with his device) could select to spread around pieces of selected content, and likewise set his preferences over what kind of content he is interested in receiving. (For the porn question also.)

Now, then when walking around any place, whenever two devices with this software get close to each other, they would begin the process of talking with each other: trying to spread content that the other party shows willingness to receive, and receive suitable content that the other party is spreading around. The filters would naturally be there, in some manner or another, be them in the form of automatic learning, or tag clouds (every user would have a tag cloud of words: only content that would be tagged to match one of the tags that the user has set would be acquired etc.).

Now, whether we are talking about the actual files themselves, or just links to files can of course be questioned. It is rather easy to envision that rather soon almost all content will be available in the cloud. When you're passing somebody in the street, it is technically very hard to download full files, but passing links to files would certainly be possible. (And then somebody could extend this with a plugin that would first download links, then continue to send the full files, given that two persons share the space for a long enough time.)

Here the content discovery part kicks in. Out of 10 million songs available, the problem becomes finding the interesting content. Basically this concept is ultimately talking about content discovery. People do not promote random material, they tend to promote material that they feel most passionate about, i.e. "the best material". Now that I'm in Amsterdam, I would automatically acquire content that people here feel most passionate about. After a long day, I can come back to the hotel, review the content, dump most of it away, and then when coming back to Helsinki, I can be "spreading some of the germs" of best content that people in Amsterdam like.

Although you could also use this for "Nokia Sensor" type activity, of reviewing what are the interests and passions of everyone in the same room (which is also certainly a cool use case), I personally think that ... Well, if I'm in a room with an interesting person, I'd rather talk with them than begin surfing their most interesting content on my device here and there. And more so then review his content afterwards.

This physical dimension is something that no desktop experience can provide. For a new song, I would see "I got this in Amsterdam in 09-Oct-2008 from ThisCoolDude".

This would be the primary layer. Then of course based on all the real world interactions you could extend this information to the internet, for instance by visualizing "the spread of memes" (how this can be also seen), as really nice timelines, heatmaps, network maps etc.

The "flu" aspect is an interesting parallel, because it has a well-known pattern of action: "a new germ" can be very powerful at first, then based on how the first people react to it, we can set how quickly it would spread forward. Diseases fade over time, people can set their "personal immunity", for when they are interested in new content and when they are not etc. etc.

Needn't be an issue.

What I envisage is that it will be probabalistic. If you go to a party, you only get the content that was generally popular amongst that group of people. You have no way of knowing which individuals were interested in that content.

So, if you go to a Metallica concert, you might be exposed to Metallica content. No privacy issue.

That would also take care of the porn aspect. Unless the majority of your friends are into the same kind of porn, you simply won't see it. And if you do see it, it can only be because you have deliberately hung out with a group of people having that as their common interest, in which case you shouldn't be surprised to see it.

Let the numbers of the crowds take care of the filtering and privacy aspects.

It can be any kind of content used on your mobile device: music, video's, photo's. Others can see the 'heat-pool' of what you are using and pick it up. It is a way of sharing and a serendipitous way of finding new content.

Thanks for the input. The poll is about what concept gets you most enthusiastic to work on in the future. What concept has the most potential to become a Maemo success?

To which concept are you referring: Contagious Content?

Thanks and: they are put here for you to comment & built further on. We hope they inspire you to come up with better solutions and get out the stuff that will/does not work! So also keep the 'black hat' stuff coming!

You assume popularity is the only viable method for filtering. It is not the only viable one, and its not fault tolerant. Ie. instead of going to the popular McDonalds I rather go to a restaurant a friend who is chef recommends. Chances are also, I already know McDonalds, because of its popularity.

Therefore, some kind of algorithm which combines various factors (such as popularity, authority) and based on intelligent profiling is required (OK, initially for fun project perhaps not, but if you want something scalable and usable...).

The question is which factors? This is difficult...

I'd say, assume not same weight for every person you meet, give friends (and friends of friends) more influence than strangers. And, allow to set 'good friends' (typical a human has 0-5 of these). This, together with popularity gives initially an OK result IMO. (But there is more necessary.)