Ideas

Sounds good. but how is this different from google?

Then I'll give you an update to my comment on your previous idea:

Thanks for the video, which at least shows one possible thing I might want to search for on your site. Can you now explain in general terms what sort of things I could find a round-up OF? I feel this is a major omission from your pitch and stops us understanding what it is that you're proposing.

Your video shows the user entering free text into a text box and searching; that implies either a vast database of topics or some way of generating the round-ups as queried. Assuming that you don't have the latter (as your mention of reviews suggests), perhaps a directory/tree structure might be more appropriate?

It's nothing close to Google or other search engines at all. This site works by relations, and hand-picked custom grouping. There is a much more depth to the descriptions on exactly what. Plus, other features like ranking, images, reviews, ect... With Google can search for everything and it uses a wide spectrum. This site is like a magnifying glass with a more narrow range but, highly focused. Google can light up the world but, it can't burn any ants.

"Can you now explain in general terms what sort of things I could find a round-up of?"
Pretty much any topic that is in the general area. The only thing it will not work for are very specific search terms. Like if you wanted to find all of the The New York Times story's on that have to do with widgets. It would not be very possibly on this site, that's more of a thing that Google is for. But, if you wanted something more general like tech news. Then it could point you to all of the good news sites that have to do with technology and it might point you to the technology section of the New York times.

The video, shows the user entering "free email" as the user types an active live search is also preformed simultaneously with pulls up all list of possible available specific variations. So, if you typed just email it would pull up all of the possible things you would be searching for starting at most popular.

This goes to the second page, which is a ranked ordered list list of all of the "free email" services available. With pictures an a description of the site features. From here you can chose to go the the site directly by clicking on the link or see more in-depth ranking as well as user reviews.

The system will work a bit more like a neural net then a top down directory/tree structure. Since, you don't know where the user is starting off. It works better then the the top down directory. Since, you don't have to sort through all of the folders, and if you wanted to find more services related to what ever topic you were looking for it would allow you to do it quickly. An example of this structure is: http://www.visualthesaurus.com/ except this project will not require load times for java since a visual view of the back-end is optional.

What do you need us to do?

Crowdsource just like any other website or software idea on CambrianHouse.

I like this idea. when you say that the results are hand picked, are they done by the user community or from a mod team? if done by the user community what steps are being taken to make sure the system isn't easily gameable?

When I read this idea all I could think in my mind was that Jason Calcanis's Mahalo (http://www.mahalo.com) was already doing this. Basically organizing information by hand for the major topic areas. Since I've always thought that Mahalo was a failed model with a dynamic (and jerk) of a leader, I guess my feeling is the same about this.

As I continued reading the idea it's almost like you're trying to do what Guy Kawasaki's Alltop (http://alltop.com/) is doing for news. Not exactly the same since you are dealing with rich media along with text, but it sounded a lot the same.

All in all, I don't really see strong differentiation that would make me use this type of site over all the other methods of obtaining information.

Roguestartv: There will for a fact be moderators over the system. The community can submit there sites. Which will then be reviewed and put on the list. We could also have it so it is active where user links would be instantly put on the list at the bottom, with the link set to "no follow" to make sure page rank still stays high and the links are not just spam. Then have it reviewed later.

Techguy: Its a different in many ways as far as architecture and focus. I can see why you don't like the modles set up like mahalo there is not much structure or good organization. It feels so off base on mahalo. I can assure you that is not what I am going for in this site.

Plus, more eye candy pictures, images and actual collaboration with users. I would not go so far to say that it is a totally unique idea to have focused lists of links on specific things. I have seen like a page or two done on the web on some sites. But, without dedication and a site geared towards it. Which is why I think this might be a good idea because of prior work that seems to be accepted by users as helpful.

I will be working on a live demo that I might be able to get online in the few days the contest has left. It will not do everything but I think it will show off some key points.

how is this different from mahalo or using other aggregators?

zudos?

* Difference form mahalo & zudos. Night and day. This is not really the same thing. Its not really intended to be a replacement of Google. There are none of the random articles or just random links. This is a functional search for quality related sites, not some free for all link farm.

You notice all of the junk parts on those sites:
"Find the right cat for you...take the Feline-ality test"
"Obama responds to "appeasement" jab"
"Can we switch faces, daddy?"

This is not a news site there are no articles, this is not a random post random links everywhere, once more this is not a replacement to take over the job of Google on searches like that.

* This is not an rss aggregators site.

How about something that could actually find what you are looking for? No aggregation nor irrelevant search (type 1 errors) but locate relevant stuff that we never get (type 2 errors)?

that would win the world ....soon.

LivingtheDream
Turn the Key and Smile

It would be nice if there was a specification which identified the criteria which would be used to classify and group these sites. Then maybe you could write some tools, to be used by site authors, which would enforce the rules layed out by the specification. Perhaps something which would embed metadata.

@LivingtheDream:
"How about something that could actually find what you are looking for?"
That is what I am going for.

"No aggregation nor irrelevant search (type 1 errors)"
Well, was never a need or a proposal to have rss aggregators. Its, just not what I am going for in the site.

"locate relevant stuff that we never get (type 2 errors)?"
Well, thats the plan. Highly relevant linking to sites that have exactly what you are looking for and spot the ace sites that are excellent and wonderful to use.

@BlueStorm:
I spent some time on classification, if you look at the end of video you will see nice map of some of the different groupings how the search was related.

Accepted sites are rated on a number of levels function, appearance, user interface, user reviews, ect... The sites are free to follow there own rules specifications and layouts.

It is possible to have websites add in something like a review our site link. But, other then that I don't think there is a need to force them to follow any specific layouts or anything.