Ask HN: feedback on our startup project | Hacker News Reader

Ask HN: feedback on our startup project

8 pointsageofish10 comments

Hi everyone, I’d love to get some feedback on our startup idea, we just launched it for limited preview.

For many of us it’s hard to find good and fresh content on the web, especially for something that’s new and upcoming. This is a common problem with many popularity based recommendation websites such as Digg, StumbleUpon, or Youtube. The problem stems from the fact that they rank content in a single global bucket, where it’s hard for new comers to compete with established ones.

We came up with the idea to allow viral web content to spread like real viruses, by real people across real locations.

Benefit number one is that by associating content with locations, each point on earth has its own virtual bucket, in which we only expose content that is spread to, or created at the location. With this breakdown, new content has a better chance to be noticed. Then because users on our system picks (Facebook “like”) and facilitates the movement of content across locations, we create a crowd-sourced filtering system like the natural selection process. So you can trust what reached your location is good.

With these, we hope to bring excellent content to consumers, and at the same time democratize content promotion.

Our blog has more details in case you are interested: http://blog.infactio.us.

Our question is, would you give it a try? Does our app (infactio.us) live up to its promise? Your feedback would be really valuable for a young startup!

Thanks HN community!

Loyd

I'm not entirely convinced that the problem you're solving is a serious one, but I still think that your results could be fun and the locality aspect might well result in neat things bubbling up that's particularly relevant to a specific audience (I'm thinking the NASCAR crowd in middle America, or anarchists in Oregon). I am interested to see how it works out, and would participate.

Exactly, some content might not be globally popular, but will be very viral locally - and good examples!

Seen the demo video, and I'm a little skeptical about content spread having to depend on real-life location. I like the current global bucket model, because it guarantees best content (even if it hides other potentially good submissions).

virus originated from Redmond, WA? I'm going to guess you are or used to be a Product Manager at Msoft.

Right on with your guess :) And I agree with "it guarantees good content" part. However, one point we didn't mention is the staleness of the global bucket. For a generic search "funny" on Youtube, most results are 4-5 years old.

We want to focus on upcoming talents, which means we want to bring in new content to the mix, to some people, "fresh + good" is better than "best", and I believe that's why StumbleUpon model is successful.

I'm rather skeptical that there are many interesting classes of information for which geography is a good filter. I don't read my local community newspaper because it's boring. On a larger scale reddit/r/australia is more interesting to me.

For most types of content I just want the best in the world and location is irrelevant. If you took the best from just my area I can imagine the quality being way below that and I live in a good area!

The best thing about the internet is how we now instantly have access to the best content and people that exist on earth. It's no longer based on who you happen to be lucky enough to live near, bump into etc...

Yes, internet made information instantly available, but it also creates overflow. We need information to be curated (news or portal), or automatically filtered (recommendation sites, Digg, StumbleUpon). I hope our filtering does not stop the best in the world content to reach your location; we need to tweak our algorithm if it does :)

By the way, we also implemented some "facebook wall", so you can see what your friends are "infected". This helps break the barrier of needing people to travel to, say, a remote location.

I really like the idea of modeling viral content after real epidemics.

Some questions:

What's the advantage of a limited preview? Do you plan to gain critical mass, say in Washington, then open it up only there? Similar to a facebook style growth pattern.

Why did you decide no downvote? Every social bookmarking site I can think of has the option.

I don't like the fact that I can't see stuff for my area without logging in. I usually lurk for a long time at a site before I decide to join and contribute.

Why don't you just grab my location from my IP address? Why do I have to turn it on in FireFox?

Hi, good questions, I will try my best to answer them:

Preview? To gain critical mass around an area was our initial intention. By the way, we are not very restrictive at the moment, as long as you request, we will give you a preview invitation!

Downvote? To stay truthful to the real-world model, we hope to let things die down quietly. In the next version we will be implementing flagging, and maybe consider downvote as needed. This is a good point!

See more before joining? This is a chicken-egg problem for us now: without large adoption, our content is suboptimal so we are reluctant to open up the experience to everyone; but most people want to see more before contributing...

So at the current stage, we are hoping to target the early-adopter crowd (for example, who would wait 6 hours in the line to get the first iPad), then open it up! I would love to get some discussion around this if others would like to chime in!

Get location from IP? Good idea, we should give it a try! However, we probably will stick to browser's native support - I think this is cool and might be the future, no? :)

I think that it has some real possibilities especially in keeping local items relevant. Where the previous poster's argument comes to play though is in content that has no specific geographic tie but rather a conceptual tie (e.g. Linux) etc. However if that is not the type of content you want then maybe it is not a big deal.

I would suggest having someone edit your About page though to clean up some grammatical issues.

Thanks for the suggestions and support!

Reply on news.ycombinator.com