Ask HN: Review my Startup (Idea) | Hacker News Reader

Ask HN: Review my Startup (Idea)

8 pointsab_initio17 comments

Hey all-

I'm not sure if this type of post is allowed, but feel free to downvote if not. Basically, I was hoping to run an idea I have for a startup by the community at large and maybe find a cofounder in the process.

About me: CS nerd/bad idea generator

So heres the gist:

Create a site that would be a reputation repository (pardon the alliteration). This repository would be similar to gravatars but deal specifically with establishing "universal" rep as opposed to avatars. Rep could be used for any type of crowdsourced site, news sites like HN or DIGG, blog stuff and I think much more. Obviously these interactions would have to be fleshed out a little bit, but I think it could certainly be a nice little addition to some sites.

Thats the basic plan. What do you all think? Worth a shot? Feel free to email me if you'd like to talk about it. daftdg at gmail dot com.

Thanks!

edited to add contact info.

View on HN

>> Obviously these interactions would have to be fleshed out a little bit

When I first read your idea I thought "Hmm... This is to reputation what OpenID is to usernames and passwords". And just then a little light bulb appeared above my head: If you tracked this by OpenID, than any site using OpenID could easily switch over to your system - without having to confirm that a certain username was a certain account on YOUR system, yada yada yada.

I think it's a good idea - the only problem is getting this to catch on with at least a couple of well known sites. That's a hard thing to do, but I think it's a good idea. I like universality (though admittedly, I'm not the biggest fan of OpenID itself)

edit: If you're really serious about doing this - I'd be happy to work with you on it - email me at SHP4HN [at] gmail [dot] com

It's too meta.

It's also impossible. HN and Digg and Reddit are not going to implement your universal reputation system.

This has been tried before by at least a half dozen companies and you've never heard of them because they all failed.

Lastly, Twitter and Facebook already solve enough of this problem with twitter auth and facebook connect and have hundreds of millions of users so you would have stiff competition.

Thanks for your comment. As for your arguments, I am somewhat unconvinced.

1) This is impossible. What exactly is impossible? Getting Digg and Reddit to adopt it? Maybe, but there are a variety of uses that I can think of that have nothing to do with digg reddit or HN.

2) This has been tried before. So what?

3) Twitter and Facebook do this already or something similar. So what? First of all, both facebook and twitter have totally different business models and really have no incentive to include reputation. Further, the mere fact that there exists a large competitor in the market doesn't mean a small venture that may be more simple, targeted, and agile cannot succeed.

Maybe I am being naive, and if so, I fully expect to hear about it. However, I find your arguments unconvincing, and I would be greatful to hear other people's thoughts on them.

Thanks again for the comment.

1) This is impossible. What exactly is impossible?

Getting anyone to use it. How will you get users?

Getting people to use it is FAR from impossible.

Let's look at it a bit of a different way. If I understand it correctly, this could be used to identify trolls and such. Makes me think about the NIN rant of last week.

So, let's say you built your first go of this on top of something like the Disqus API, and convince a few sites to use it.

Maybe create a few simple plug-ins for forum services, etc.

All far from impossible. Don't go the 'boil the ocean' route where you try to grab everything at once. Piggybacking on OpenID is a decent idea (as mentioned already), but personally, i believe that OpenID is far from a success (it isn't mainstream yet). You could potentially create a system that piggybacks on Facebook Connect, Google Friend Connect, etc. and as long as you have a simple enough way for to hook into the service, I think you could get some traction.

I think thats certainly an issue. Right off the top of my head I see two distinct paths by which one might drive adoption:

1) There are a variety of smaller sites that utilize reputation type systems. For examples, take a look at all those tech help sites (one called it exchange comes to mind). These all fit into the Question/Answer model. I think there is some merit to exploring blogging as a second option. Specifically blogs can use reputation threshholds for posting, filtering, and comment prioritizing (putting the good ones at the top). Basically those are the primary areas where I'd like to look to start building a user base, but I think there are many more out there.

2) As one commenter recommended, the service could be expanded to include more than just simple +/- karma type scores and move to more complex stuff like Kiva loanability etc. This is pretty self explanatory, add a service that makes life easier for these sites, and they'll ask their users to adopt it.

Finally, I realize openid is a different beast but it boils down to similar idea, universalize your identity. Login and passwords are, of course, prevalent everywhere, so there is a huge market for this type of service and it seems to be picking up some steam. Our target market is smaller, but if it proves nearly as useful to users within that space, it has a chance of being adopted as well.

I think this is a good idea. I've actually thought of something similar myself. Why not have each person's reputation be based on a variety of attributes -- discussion forum insight, financial reliability, technical expertise, etc.

This could eventually replace the credit ratings infrastructure of experian, equifax, etc.

Ideally lots of sites would participate -- e-commmerce sites could rate you as a customer (headache free, not trying to cheat on promotions, etc.), kiva could rate you, etc.

It would also be great if reputation could be tied to biometric data to prevent transfer and fraud.

Such a system would ideally be open, cryptographically sound, and free.

I think that a properly designed system would end up removing lots of the transaction costs associated with more mainstream reputation systems.

* This could eventually replace the credit ratings infrastructure of experian, equifax, etc.*

Wow - talk about smoking the kool-aid. We went from "similar to gravatars" to replacing Equifax all in about 1000 characters. That's a bit much for me lol.

Well consider that those companies charge $10 per credit rating lookup...

Credibility/trust tracking is huge. Ebay's seller/buyer trust data, if tied to SSN (or better) would be worth Billions.

It's an interesting idea. There are a lot of details that you would have to flesh out.

What is reputation? How would you explain it? How does it aggregate?

Draw some pictures - what might it look like, what might it be integrated into a site? Etc.

In my mind, reputation is the summary of a series of transactions. Ebay's user reputation comes directly to mind. How they will interact across sites seems challenging -- I have 1000 points on news.yc and 2000 followers on twitter; how do they interact? How do you present this, etc?

I recommend you cut the idea to the very minimal core to see how small you can make it, and see if you can validate that minimal concept.

Joshu - Thanks for the comment. Curious as to what exactly you mean by "cut the idea to the very minimal core to see how small you can make it, and see if you can validate that minimal concept." Are we talking about hashing out some specific feature and then seeing if that, on its own, will be successful? I guess, if you could explain a little about what you mean by "validating a concept," it would make things more clear to me.

Thanks

People have a tendency to overcomplicate things.

You are looking to build a system that is a) as small and as simple as possible that yet b) still retains conceptual integrity.

You want the first because it's easier to build and get to market, and the second because it's not done till all the necessary parts are in.

Remember that the vast majority of your understanding of the problem space is not accurate, and will remain so until you encounter customers and users. So do as little work as possible; just enough to get you to those customers and users. Then you listen to them to find out the next parts to build. (The ideal case here is that you are also customer #1, incidentally.)

So no, not just some specific feature; you need the right coverage. But make sure you don't go overboard anywhere in particular, especially if it might not be needed.

Awesome. Thanks for the clarification. I wonder if you (or anyone) think it would be more fruitful to take a holistic approach i.e. factor in ebay, twitter, HN and create an overall score, or stick with a simpler, more streamlined karma counter type thing that would be "widgetable". Or does the latter necessarily lead to the former?

You're still dealing at levels of abstraction too far away from an actual product. At a distance, all these things are similar, but up close they are quite different.

How would you "factor in" ebay and twitter into one number that made any sense at all?

I like the idea, i'd allow something like that on my website.

Will you call it whuffie?

You can buy the naming rights for 1m whuffie

Reply on news.ycombinator.com