Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

HN: I'll run a marathon to build whatever you decide on tweet.in

news.ycombinator.com

Post reply on HN
Print view View original

11–20 of 24 posts

Re: HN: I'll run a marathon to build whatever you decide on tweet.in

#11

Post by SwellJoe »

Arun2009 wrote: post #10

"Tweeting, inverted". From my knowledge of twitter (have NEVER used it), it allows one person or entity to publish messages for everyone to see. What if several people want to publish messages TO someone or some entity (like a company), and want them to be seen publicly? You create a channel for that entity on tweet.in and tweet your message into the channel. Others will be free to join in. It'll be like one of those…

Hashtags (#brand, #name, #topic, etc.) have become the de facto means of doing this. I'm not sure what a tool could do to make that more effective.

I think the magic of Twitter has partly been its ability to be anything people want it to be. Just like Zombo.com.

Re: HN: I'll run a marathon to build whatever you decide on tweet.in

#12

Post by plaggypig »

erikwiffin wrote: post #8

Does Wolfram|Alpha have an API? if so, you could do a pretty cool question/answer thing. erikwiffin: @tweetin why is the sky blue? tweetin: @erikwiffin The sky's blue color is a result of the effect of Rayleigh scattering. Shorter-wavelength blue light is... http://bit.ly/3qaEo tweet.in could be a searchable/rankable/commentable database of all the searches that it has performed.

There is an API so I believe your suggestion ought to be possible. Just to play devil's advocate though; wouldn't people overestimate Wolfram's capabilities and expect the bot to be able to answer all sorts of difficult questions that it can't grok, particularly about ongoing political/cultural events which seems to be the lifeblood of Twitter, thereby making the service look kind of lame and useless..?

I'll do whatever gets the most votes but it seems this thread was forsaken. Would have been nice to see more discussion/refinery!

Re: HN: I'll run a marathon to build whatever you decide on tweet.in

#13

Post by plaggypig »

chaosprophet wrote: post #2

You could build a twitter search app, which searches and returns real time news. It should be capable of categorizing tweets about an event as headline(MJ's dead), more info at(latimes.com), comments. In short it should search twitter for real news and display it in a traditional blog style. Perhaps, it could even display excerpts from the info source site. The comments display should be real time so that you can in…

Interesting idea and I can think of a novel interface for this. I upvoted you, and that's where by influence will end.. for now :)

Re: HN: I'll run a marathon to build whatever you decide on tweet.in

#14

Post by chaosprophet »

hotpockets wrote: post #9
tlrobinson wrote: post #3

Some sort of user quality measurement / recommendation engine thingy. There's a lot of bullshit on Twitter these days, we need tools to sift through it to find the good stuff. Not Twitter's "Suggested Users" list, which mostly contains celebrities and companies. There's a lot of metrics you could use: * negative points for "follow-baiting", i.e. following hundreds or thousands of random users, hoping they'll follow y…

There is a real need for a reliability metric. For instance, a lot people recently tweeted rumors about Jeff Goldblum dying. Penalize those people. Other people actually were debunking the rumors, reward them. It will help quality to use an NLP grammar parser I believe. Every time there are rumors/scams on twitter you have a chance to improve your metric.

Problem with an NLP parser is since twitter restricts to 140 characters a lot of people don't post proper grammatically correct sentences. This makes the parser much more complex since the parser will then have to guess the intended meaning. A tweet like "Jeff Goldblum???" would be almost impossible for the parser to understand.

Re: HN: I'll run a marathon to build whatever you decide on tweet.in

#15

Post by notaddicted »

I heard a rumor that people use twitter like a search engine -- ask a question, get answers.

Make a bot that records such questions, and their answers, and then make a search engine over the records.

Expert Mode: discern between questions like Am I fat?? vs. What is a cheap but nice hotel in LA? -- although it maybe funny not to.

Is there some kind of tweet stream you can process for this?? It seems like it would hammer the twitter infrastructure so it may not be coool.

Re: HN: I'll run a marathon to build whatever you decide on tweet.in

#16

Post by hotpockets »

chaosprophet wrote: post #14
hotpockets wrote: post #9

Earlier quoted context omitted.

There is a real need for a reliability metric. For instance, a lot people recently tweeted rumors about Jeff Goldblum dying. Penalize those people. Other people actually were debunking the rumors, reward them. It will help quality to use an NLP grammar parser I believe. Every time there are rumors/scams on twitter you have a chance to improve your metric.

Problem with an NLP parser is since twitter restricts to 140 characters a lot of people don't post proper grammatically correct sentences. This makes the parser much more complex since the parser will then have to guess the intended meaning. A tweet like "Jeff Goldblum???" would be almost impossible for the parser to understand.

There really aren't that many ways to say "I heard Jeff goldblum just died???". You could probably make a list of like 100 different ways and get good accuracy. In fact, question marks alone are a good indication the person didn't do any private verification.

Re: HN: I'll run a marathon to build whatever you decide on tweet.in

#17

Post by chaosprophet »

hotpockets wrote: post #16
chaosprophet wrote: post #14

Earlier quoted context omitted.

Problem with an NLP parser is since twitter restricts to 140 characters a lot of people don't post proper grammatically correct sentences. This makes the parser much more complex since the parser will then have to guess the intended meaning. A tweet like "Jeff Goldblum???" would be almost impossible for the parser to understand.

There really aren't that many ways to say "I heard Jeff goldblum just died???". You could probably make a list of like 100 different ways and get good accuracy. In fact, question marks alone are a good indication the person didn't do any private verification.

If you were going to create 100 different rules for that one sentence, imagine the costs of scaling up. It simply is not viable. The only way to make this possible is to maybe train the system for a corpus consisting of about a million tweets. But the problem with that would be that those million tweets would first have to be individually tagged manually by humans.

Re: HN: I'll run a marathon to build whatever you decide on tweet.in

#18

Post by growt »

I think the ".in" is perfect for some location specific service. You could build urls like tweet.in/London etc. There are already some services that show you twitter-users in some specific city or area, the problem is that these people do not always tweet about their city, so it's almost useless.

I once thought of a service that lets you add a location to a tweet and a short url. That service would create a map with all your tweets composed in that way.

Re: HN: I'll run a marathon to build whatever you decide on tweet.in

#20

Post by plaggypig »

growt wrote: post #19

You might also run into copyright/trademark issues with that domain-name. see: http://www.techcrunch.com/2009/07/01/twitter-grows-uncomfort...

This is totally contrary to what was said last year: http://news.ycombinator.com/item?id=682790

I'm very disappointed by their cack-handedness.

Post reply on HN