These are in a couple of pubs around me for the last 2 years and I have never seen anyone uses them (Maybe restaurant are better for them). Its the type of thing I thought would be a great idea but in reality....

What are the names of the pubs? I'll ask Andy if he did those installs.

(i think they only have one now did have 3 before)

http://www.lockebar.com/

and

http://limerick.ratemyarea.com/place...ter-pub-125944

I think this is a great idea and is going to become a reality sooner than we think.

Google has an interface for search trends compared to market trends. If you can figure out a good way to use this information, and integrate it with twitter and facebook it may take a lot of research (tons and tons) but I think there are legitimate results to be found.

My question would be for all of these research metrics, there seems to be a lack of information available, and proprietary information that you cannot gain access to. You can figure out how many people "like" something or how many twitter followers a company has, but how do you know a company is using these numbers effictively, and there are a lot of ways to "fake" twitter followers, facebook likes, etc.

Yeah, I know this idea is cool in theory, but, as crazy as this sounds, I don't think there is the necessary amount of legit data on Twitter to get any meaningul info from it. Sure there are millions of tweets a day, but people just aren't using it for discussing products that much, the amount of people saying "Loving my new Reebok Crosstrainer 20s" or whatever is so small. I just searched for 'Sodastream', which is apparently a somewhat hot product and on the cover of a Bed, Bath and Beyond circular right now, and almost all the mentions are spam, or people entering contests, and one guy saying "DO WANT" in the first 40 results.

If you just look at total number of mentions, you're not going to get any idea at all of how the product is truly trending in consumers' minds, because there is just way, way, way too much spam and noise on Twitter.

Facebook, I have no clue, it may be a lot more useful if you could get to all the walled parts (which you can't since they're walled I guess haha).

Now I actually write all this out, know what I would actually do if I was going to launch this business: I would actually scour message boards and forums, I think THOSE would have a wealth of great info and you could extrapolate buzz (good or bad) on a product very easily and accurately from those. But then if I was marketing it, I would emphasize the Twitter part, because everyone thinks Twitter is some amazing magical hose of information and useful.

So okay, 3 paragraphs into my reply I realized this *IS* a super viable proposition, just with additional info on top of twitter and FB. So I like it!

This is gross. The only college hoops tourneys I've been to, pac-10 and March NCAA, were dry. fml

This is the exact same way I felt replying to it.

The twitter stock analysis is definitely a legit idea! You gotta be careful about this kinda stuff though.

http://www.huffingtonpost.com/dan-mi..._b_830041.html

http://www.theatlantic.com/technolog...s-stock/72661/


Obviously this isn't any kind of conclusive evidence, but there are going to be a LOT of edge cases that you'll have to be pretty good at to get around.

Hedge funds using twitter for trading have been around awhile now. If you think you can compete with DE Shaw's quants and make a better model than the one I'm sure they've been working on for years lol.

http://www.dailymail.co.uk/sciencete...re-prices.html

"It works by first sweeping a swathe of random tweets for positive and negative comments and then processing them with a Google-designed program that works out if the tweeters were in one of six moods: calm, alert, sure, vital, kind or happy."

This is weird, what does Google-designed program mean exactly? Is it related to Google Zeitgeist, or something similar? Perhaps some Google have open sourced some code related to that?

Anyhow, from the paragraph I quoted, doesn't sound like they're using some crazy program they've been working on for years, sounds like they are just scraping a bunch of tweets, seeing when people feel happy/kind/vital/sure/alert/calm (using said Google-designed program), and then checking the Dow. They say that the moods hit and THEN the Dow moves, but I have my doubts - could any of this be because of things that happen after-hours, like there's a tsunami in Japan, a million people on Twitter say "oh my god this is horrible", the program makes a note, and then the market opens up 10 hours later down?

I am just skeptical of the quality of data you're going to get from a site that is at it's core just a huge chatroom and ~25% of accounts are bots. That article quoted the results from 1 month of trading as if it was a meaningful sample size to judge this thing's effectiveness by.

Either way, I do find it interesting that they chose those 6 moods. If you asked me to guess what 6 moods they care about most, "sure" or "vital" wouldn't pop into my head. Would love to know more on how they came to choose those moods.

I had 3 ideas in my life.

1 - the year before the NFL put it in, I was sitting on a friends couch in the basement and said we should find a way to put infrared on the sidelines and have a line going across the field on the TV so we can see the first down. Woops.

2 - wishlist. An online registry where you could pick all the items you wanted during christmas or whatever occasion. It bothered me trying to figure out what to buy for people. People could create an account and pick all the items they wanted in different price ranges and categories. A Kiosk could be put into each mall and you could walk up, look through your friend's wishlist, and the kiosk would direct you to which store in the mall has that exact item, and the price. We could make money by charging the store a fee for the service and all we needed were the kiosks.

This idea has already been patented from what I heard.

3 - A hinge that is self-rising so we could install it in public bathrooms. Pretty much men's toilets are disgusting, so why not attach a hinge and spring that would make the toilet seat rise up when let go, and you could push it down if you had to take a crap. That way, it stays clean and men would pee all over the seat.

Never followed up with that one.

I had some other ideas too.. but those were my 3 good ones.

Oh, it's already a reality. The current output is just sub-optimal though that the managers I know who used this info are really just using it on like micro-caps that get a mention - like Justin Bieber was spotted wearing something from some $85m mkt cap firm, so there's x% chance the stock will rally based on that alone [even before sales and profits pick up] etc.

There's no question at all that you get legit results with which you can create alpha. With something more sophisticated, you can get a leg up on sales trends or pricing points, and to make real bank on Wall St you only need the figure out what that market 'edge' is in 13 weeks or less. If someone is gonna come in weaker or stronger than expected, you will get paid.

You're confusing twitter with twitter + FB + google search trends.

Also, you seem to be confusing Quant trading with value-oriented [or growth/momo ] trading.

I'm well aware David Shaw uses these methods for pure quant trading, I have multiple clients in his funds.

http://www.google.com/finance?q=NYSE%3Askx

Look at the last 2+ years of SKX: The rally from 5 to 45, and subsequent drop to 12 all has to do with Shape-ups and 'toning' fad [higher sales followed by the inevitable inventory writedown and margin crash and stock implosion]. If you were able to mine that info via social media, and figure out what sales were going to do via expectations, you made bank. [either on the way up or the way down]

Extreme quant skills not necessary. An expected delta in the GM was all you needed to figure out long v short.

There are several funds a lot smaller and less quant-driven than DE Shaw that are doing this.

I can be a pure L-T value investor but if this analysis tells me that Skechers is gonna sell 30-50% toning shoes because the fad is over, I'm going to sell my longs at 40-45 and possibly rebuy at 10-12. Etc.

Anyone who thinks anyone has figured out how to optimize the data flow from FB + Goog/Bing + twit is an idiot. David would be the very first to tell you he has not.

But hey! You skimmed a daily Mail article, congratulations brahsome!

You don't need to compete with DE Shaw to make money. I'm sure there would be lots of companies lining up to buy information from what you could scrape from twitter. This information would be helpful for marketing purposes, elections and I'm sure many uses other than investing.

Actually I just googled hedge fund uses twitter and linked you the first article. I didn't even read it tbh. There are several other articles dating back over a year ago detailing the phenomenon. This is when people started reporting about it, not when people started researching it/using it. All I'm saying is

1) This idea is not new
2) There is lots of big money behind scraping information off twitter/fb/google and while you and your two programming buddies might be able to find a small niche to make a little money - if you think you are going to become a major player charging hedge funds 100s of thousands of dollars for your info I think you're dreaming. But hey, feel free to prove me wrong.

http://www.wishlistr.com/

no mall kiosks, but they do a good job with your starting point.

Kantar Media has various "heat" indexes used for this sort of planning, plus a lot of other datasets.

The NFL idea would have made me gabillions. It was by far the best idea I've ever had, but it doesn't matter now because I wasn't in a position to do anything about it.

And I was late on it. But don't let facts get in the way of a good story.

Someone should make a website or a service that caters to amateur/semi-pro (IE - getting started but w/o the support team of a big company) iphone and android app developers. There are so many times late at night I would have gladly paid $100 for someone with experience to spend a few minutes on my question that I've been beating my head against the wall over for 8 hours.

There are hundreds of thousands of amateur and somewhat paid app developers out there with varying levels of experience. Often your problem is something trivially stupid to an experienced developer - but for whatever reason it's hard to google (EXEC_BAD_ACCESS - and the app just crashed). This is a brand new ecosystem that we've never seen before and I think the market is being underserved.

I've been a web developer since the late 90s, and I know the difference between web development, client-server development and mobile app development. The latter is much more prone/conducive to smart kids/adults working completely on their own based off nothing but googling and youtube examples. In a lot of cases these people are already generating revenue with their work, or have modest self funds.

Find some experience iOS app developers in Russia or India (prob Russia/Ukraine etc.) - do $150 per support ticket (or buy packages etc.). Make it clear you're what you will and will not handle. IE - we're not going to write your code for you (well or that's a different package). But we will handle the stupid vexing questions like why you can't find ARM6 architecture on the new xcode, even though the instructions clearly tell you that you can still use it. Etc.

Either that or make some kind of "common gotchas" wiki-type page for mobile app developers - and sell the advertising. All the forums out there now are horrible, usually out of date, no one replies. No one wants to wait hours or days for their answer anyway. A wiki is much better.

Ok so that's my idea. Someone feel free to run with it because I'm never going to have the time or energy to set it up and market it.

heres an idea i havent seen.

thumbless gloves. ive seen fingerless but not thumbless. great for using a cell phone. someone invent this so i can buy them tomorrow, or link me.

they have conductive gloves, so you can use your mobile phone.

Personal audio recorders that live on your body and record every conversation you ever have. Remember where you heard it first.

Heard this idea mentioned on This Week in Startups a few weeks back. The guest was asking someone to make an app where you take a pic of a physical key you want to replicate. The service would then make a copy based on the picture and ship it to you.

Security issues would be a big hurdle to overcome.

I would pay for this.

StackOverflow? I'm sure you would get a good quality free answer in a very quick amount of time.

For reference, if I post an ASP.net/c# question I usually get a top quality reply within 10 minutes.

SO have blogged a lot about paid Q&A sites I think, and I think the general opinion is that these sites are inherently flawed.

Experts-exchange is a prime example.