Slashdot Mirror

How is the parent offtopic? It's about charity and helping them. Their reward site is here and if you don't want to use the reward points yourself, you can give them to charity.

They did offer a defense: it's the customer data

Let me see, they put a routine in the customer's computer that collects what the customer types and what is sent to the screen when the customer uses a third party application.

That is usually considered a crime, not a defense. It would be the weirdest form of alibi if someone claimed he could not have robbed a bank because at that exact moment he was murdering someone.

It's only a crime if MS installed said third-party program. If the customer installed said toolbar and left the checkbox that basically says "Microsoft can watch everything I do in my browser," then no, that wouldn't be a crime because the customer explicitly opted in to it.

You can thank competition: http://www.discoverbing.com/mobile/411/

OTOH, msft copied Google on this one. Does that mean they will copy Google on discontinuing the service? ;)

Microsoft still is in the 411 game with Bing 411. Link

1-800-Bing-411
(1-800-246-4411)

Back in the early 90's (92, I believe), I was co-op'ing for IBM and was lucky enough to get to go to COMDEX provided I man a booth for a while. The product I was demo'ing was voice independant voice recognition (it was all the rage at the time). There was no training required, random guy from the street could walk up and interact with the computer by voice, regardless of dialect or accent. I got pretty good with it, but I noticed that some people did have to repeat themselves (but not more than twice) to get it to work -- again, early times in terms of speech recognition. But the reason I was good at it was that repeated practice actually trained ME to speak the way it wanted instead of it being able to adjust to how I spoke. Speech recognition has become more prevelant since then (BING 411 anyone? http://www.discoverbing.com/mobile/411/ ), and I'm sure you've made adjustments to how you speak to computers just to get past the voice prompts. You speak slower with more distinct pauses between words.

Behavior modification is an effective way to improve computer input.

Hah, that reminds me of an old phone that I had. It has voice dialing and it would always speak back the name it thought i meant in it's own "voice". The funny thing is, it basically sounded like an asian with a thick accent, which is probably because that's who programmed the english into it (it was an LG, and I'm pretty sure they're made in Japan, right?). Anyway, I basically learned that if I faked a thick accent it would understand me better. When i wanted to call my friend Zack, it would only work half the time, until I learned to say "Sek" instead, which is how it said it back to me when it worked.

It also wasn't really programmed to handle acronyms, so I had to speak those out.

Once i learned how it worked, it was perfect, but it definetely trained me!
-Taylor

Speech recognition has become more prevelant since then (BING 411 anyone? http://www.discoverbing.com/mobile/411/ )

Goog-411 anyone? Been around for years too...

Back in the early 90's (92, I believe), I was co-op'ing for IBM and was lucky enough to get to go to COMDEX provided I man a booth for a while. The product I was demo'ing was voice independant voice recognition (it was all the rage at the time). There was no training required, random guy from the street could walk up and interact with the computer by voice, regardless of dialect or accent. I got pretty good with it, but I noticed that some people did have to repeat themselves (but not more than twice) to get it to work -- again, early times in terms of speech recognition. But the reason I was good at it was that repeated practice actually trained ME to speak the way it wanted instead of it being able to adjust to how I spoke. Speech recognition has become more prevelant since then (BING 411 anyone? http://www.discoverbing.com/mobile/411/ ), and I'm sure you've made adjustments to how you speak to computers just to get past the voice prompts. You speak slower with more distinct pauses between words.

Behavior modification is an effective way to improve computer input.