Live data from Hacker News

Show HN: ProveTrust, a user trust reputation system for the web

provetrust.com

1–10 of 29 posts

Re: Show HN: ProveTrust, a user trust reputation system for the web

#2
The idea is to have a central place on the web to store user reputation and reviews across many sites. You should be able to have one reputation (Trust Score) for all sites and not have to re-build it for every eBay or AirBnB. Further, a lot of sites like Craigslist or OKCupid don’t even have reputation systems, but could benefit from one.

I consider this a minimal viable product and plan on adding a lot of features, but would like to get some early feedback. Thanks in advance.

Re: Show HN: ProveTrust, a user trust reputation system for the web

#3

The idea is to have a central place on the web to store user reputation and reviews across many sites. You should be able to have one reputation (Trust Score) for all sites and not have to re-build it for every eBay or AirBnB. Further, a lot of sites like Craigslist or OKCupid don’t even have reputation systems, but could benefit from one. I consider this a minimal viable product and plan on adding a lot of features,…

great job so far. what stack you using?

Re: Show HN: ProveTrust, a user trust reputation system for the web

#4

The idea is to have a central place on the web to store user reputation and reviews across many sites. You should be able to have one reputation (Trust Score) for all sites and not have to re-build it for every eBay or AirBnB. Further, a lot of sites like Craigslist or OKCupid don’t even have reputation systems, but could benefit from one. I consider this a minimal viable product and plan on adding a lot of features,…

great job so far. what stack you using?

Thanks. Python App Engine with AngularJS.

Re: Show HN: ProveTrust, a user trust reputation system for the web

#5

The idea is to have a central place on the web to store user reputation and reviews across many sites. You should be able to have one reputation (Trust Score) for all sites and not have to re-build it for every eBay or AirBnB. Further, a lot of sites like Craigslist or OKCupid don’t even have reputation systems, but could benefit from one. I consider this a minimal viable product and plan on adding a lot of features,…

Someone who is a trustworthy seller may not be a trustworthy date partner, do you generate different scores based on context or just average them all?

Re: Show HN: ProveTrust, a user trust reputation system for the web

#6

The idea is to have a central place on the web to store user reputation and reviews across many sites. You should be able to have one reputation (Trust Score) for all sites and not have to re-build it for every eBay or AirBnB. Further, a lot of sites like Craigslist or OKCupid don’t even have reputation systems, but could benefit from one. I consider this a minimal viable product and plan on adding a lot of features,…

I worked for a company that would have been a target client for a product like this. In the end, we didn't use one of the many similar solutions (legit.co, trustcloud, scaffold, peertrust, project trust, truly, etc..) for two reasons:

* We didn't want to give our information away. It was a nonstarter. Either to protect it from being used by competitors or because of security concerns of giving private user information to a third party. * Each site and community has different levels of trust, and actions have different meanings. Whether someone commits sexual harassment on OkCupid doesn't matter when you are trying to buy their toaster on eBay. I am much more concerned about someones behavior when they are living in my house, not so much when they are giving me a ride in their car within the city at noon.

It would be nice if everyone didn't have to build their own trust system, but any competent company will. A general solution will either be too general to be useful, or require too much private information to be allowable.

The only way I can see a standalone product like this occurring is if a well established company like AirBnB of Facebook would build it on their existing (valuable) dataset and and guarantee data privacy.

Re: Show HN: ProveTrust, a user trust reputation system for the web

#7
post #6

The idea is to have a central place on the web to store user reputation and reviews across many sites. You should be able to have one reputation (Trust Score) for all sites and not have to re-build it for every eBay or AirBnB. Further, a lot of sites like Craigslist or OKCupid don’t even have reputation systems, but could benefit from one. I consider this a minimal viable product and plan on adding a lot of features,…

I worked for a company that would have been a target client for a product like this. In the end, we didn't use one of the many similar solutions (legit.co, trustcloud, scaffold, peertrust, project trust, truly, etc..) for two reasons: * We didn't want to give our information away. It was a nonstarter. Either to protect it from being used by competitors or because of security concerns of giving private user informatio…

I worked for this same company, and wholeheartedly agree. This type of product is a non-starter. Companies like Legit.co, Trustcloud, and Scaffold have all tried this and failed.

Furthermore, there's really not much value in the scale you're giving. It's scary. If someone is 95% trustworthy, should I trust them? Or is that last 5% the axe murderer portion?

Overall, I can't see any target market for this. Companies would build their own, and consumers don't want to take complicated 3rd party trust into their own hands.

Re: Show HN: ProveTrust, a user trust reputation system for the web

#8

The idea is to have a central place on the web to store user reputation and reviews across many sites. You should be able to have one reputation (Trust Score) for all sites and not have to re-build it for every eBay or AirBnB. Further, a lot of sites like Craigslist or OKCupid don’t even have reputation systems, but could benefit from one. I consider this a minimal viable product and plan on adding a lot of features,…

Looks very polished for a MVP, well done!

I do not have a Facebook profile, so the potential employers that run an "internet background check" will not trust me much.

Does that mean that I am not trustworthy? What I am getting at is that there are multiple levels of real and artificial trust. Being socially engaged on Facebook shows just that: you are are socially engaged and have lots of friends: you are probably a real person. Scammers can be very sociable people too, yet only their closest friends can hope to trust them, for the rest it's a facade: keeping up appearances. Buying followers to increase their Klout score.

Let's say this gets big. Do you expect attacks on your system and algo's to inflate the trust ranking? Would people leave fake reviews for themselves? Would people link up fake or stolen craigslist profiles?

If you happen to get the snowball rolling and the demand gets very high, do you think companies will form to cater to that demand? Scammers could create profiles that score well, and sell them priced according to their Trust Score. A luxury problem to have, but what is your view on this?

Finally how do you deal with people sending you e-mail or letters complaining their score is too low, or that your algo's are unfair, or that they didn't create that public profile, or want to remove a negative review? There seems quite a lot of (legal) hassle involved in the user problems that arise with such an application.

Re: Show HN: ProveTrust, a user trust reputation system for the web

#10
post #6

The idea is to have a central place on the web to store user reputation and reviews across many sites. You should be able to have one reputation (Trust Score) for all sites and not have to re-build it for every eBay or AirBnB. Further, a lot of sites like Craigslist or OKCupid don’t even have reputation systems, but could benefit from one. I consider this a minimal viable product and plan on adding a lot of features,…

I worked for a company that would have been a target client for a product like this. In the end, we didn't use one of the many similar solutions (legit.co, trustcloud, scaffold, peertrust, project trust, truly, etc..) for two reasons: * We didn't want to give our information away. It was a nonstarter. Either to protect it from being used by competitors or because of security concerns of giving private user informatio…

Thanks for the feedback. In this case, the target clients are regular users, not companies. Craigslist doesn't need to share any of their data for this to be useful to Craigslist users.

"Whether someone commits sexual harassment on OkCupid doesn't matter when you are trying to buy their toaster on eBay."

You can choose to ignore this information, but for some users it may be a relevant signal. I personally would rather buy a toaster off of non-sexual-harassers. Also, perhaps there will be less harassment if there were consequences, i.e. a worse online reputation.

Post reply on HN