Hacker News new | past | comments | ask | show | jobs | submit login
Idea for killer mobile app? (battellemedia.com)
7 points by newton on Sept 21, 2008 | hide | past | favorite | 10 comments



These are people who JUST DON'T GET IT. You see, texting is a completely different form of communication.


Exactly.

Moreover there are reasons why texting evolved as the popular application of the underlying technology SMS, which supports many other forms of broadcasts. (including the killer idea)

Excerpt from Wikipedia

Most thought of SMS as providing a means to alert the individual mobile user, for example, of a deposited voice mail, whereas others had more sophisticated applications in their minds, such as telemetry. However, few believed that SMS would be used as a means for sending text messages from one mobile user to another.

Read on - http://en.wikipedia.org/wiki/Short_message_service


Exactly. The guy writing the article misunderstands why people send messages with text. It has little to do with convenience, or because there is no alternative to it - it's just a form of communication that many people like for whatever reason. I won't make the same mistake and claim that I know why others prefer texting sometimes, but personally, I use it when I want to make a neutral query to someone, when I want to communicate with someone without revealing the tone of my voice, when I want to give a person the opportunity to ignore what I am saying or to delay a reply, etc.

Communication is wide and varied, and SMS is a choice in communication. Trying to replace SMS with voice is like saying that one should replace email with telephone calls. The applications are different, and people use it for a different type of communication.


texting is a form of asynchronous communication who places the greater burden to the sender who has to spend more time typing a message, while a phone call can be faster.

But, text messaging is considerate to the receiver, as it lets the receiver consume the message at their own time, while a call would have to interrupt the person whatever they are doing.

Text is better in very noisy environments, bars/clubs/concerts.

In Europe at least, texting as always been a little bit cheaper than a phone call, and the receiver usually doesn't pay for received messages.


But it's more than just convenience. It's also the impersonal part of the text that makes it so popular. You can write stuff in text messages that would be wierd saying out loud.

Just like me writing this message to you - I can think about what I am writing, revise it, etc.


Already done. And successful enough to be acquired. Yet by no means 'killer'.

Rapid Messaging Service (RMS): http://www.heyanita.com


I think Jott already does it as a premium service-- but who wants to call a flippin phone number just to send a quick txt?! I've heard there are some people who swear by jott, but I think they are all old people because if you are halfway competent with a smartphone it just seems like a pain in the ass.

You could try to port http://www.speech.cs.cmu.edu/pocketsphinx/ to the iphone...

Or even easier just stream the audio off to a sphinx-4 server and push the txt results back to the phone in near-realtime for visual confirmation/correction and sending. Very seamless for the end user and compensates for voice recognition sketchy.

It would be interesting to then run the text through some kind of txt-speak filter that would shrink the text. I'm always impressed by people who can cram a small novel into 140 characters.


Voice recognition would obv be very cool if it was instant and involed machine tralsnation so you could talk in english and on the other end it comes out in chinese(or whatever language you choose).

Also, some phones already have voice recognition, if minimal, right? You can say a name and the appropriate person is called.


That is (as far as I know) only voice-matching. It matches the name you said to whatever pre-recorded read in name you have assigned to a contact.

So it's not like it actually understands what you say, more that this input "pretty much" matches this pre-recorded file.

There's a long way to from there to full, usable voice-recognition.


So basically he thinks a killer idea is to 1. use a phone, 2. to do voice recognition, 3. to send message as text, 4. to recipient which somehow will get the same message read as if spoken originally.

So what he is saying that every phone should have voice recognition, voice regeneration and blah blah. Colour me stupid, but for this purpose, why not, you know, send a voiceclip over MMS or just make a damn call?

This is a hammer looking for a nail.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: