Show HN: DittoKey - Door keys copied from a photo app.(dittokey.com)
dittokey.com
Show HN: DittoKey - Door keys copied from a photo app.
http://dittokey.com
10 comments
What about an opportunist though? Someone you know could snatch your keys from the table at a bar, but snapping a picture of your house keys leaves you unaware that anything has happened, until they rob your house.
Of course you'd have to be hanging around with some pretty shady people for this to happen, but you don't always know everyone at the table well...
EDIT: That said, I really like the idea. I wish something like this was available in Europe.
Of course you'd have to be hanging around with some pretty shady people for this to happen, but you don't always know everyone at the table well...
EDIT: That said, I really like the idea. I wish something like this was available in Europe.
Right now the computer vision isn't really well optimised it takes about a minute to go through the process. Give me a wad of blu-tack and I can get an impression of your key in a couple of seconds.
Valid point though, I've thought most of the security concerns through from a logical angle and I'm pretty happy with the situation, but potential customers will just make a snap judgement based on 'feel' .
Valid point though, I've thought most of the security concerns through from a logical angle and I'm pretty happy with the situation, but potential customers will just make a snap judgement based on 'feel' .
Next, DittoKey could sell special keys that warn you by mail when they are being copied via DittoKey.
The "in another life I could have been a spy" part of me LOVES that this can be done.
I've read that with zoom lenses and consumer level dslr you can get high enough quality images to be able to copy someone elses keys without getting within 100 feet of the key.
I've read that with zoom lenses and consumer level dslr you can get high enough quality images to be able to copy someone elses keys without getting within 100 feet of the key.
Love this. Marketing is going to be critical to your success - it's a one-off, low-cost transaction that people rarely need, and you're competing against the keycutting booths in supermarkets that people see at least once a week.
But you've got great novelty. Perhaps create a theme around the idea that you're harnessing CIA technology (well, they've definiely done something similar at some point in the last 50 years). That's likely to get you a decent amount of mileage in the media, and make you stick in people's minds.
But you've got great novelty. Perhaps create a theme around the idea that you're harnessing CIA technology (well, they've definiely done something similar at some point in the last 50 years). That's likely to get you a decent amount of mileage in the media, and make you stick in people's minds.
Thanks Alex. Yeah my thinking is that the name is critical you will have to remember it for the 6-12 months between hearing about the app and actually needing a key.
The other angle is key 'backup' so you scan all your keys as a backup and if you loose it I send you a PDF you can take to the locksmith and get a key cut whilst you wait. My friend suggested 'KeyBrother' for this service, from the Seinfeld episode :-)
The other angle is key 'backup' so you scan all your keys as a backup and if you loose it I send you a PDF you can take to the locksmith and get a key cut whilst you wait. My friend suggested 'KeyBrother' for this service, from the Seinfeld episode :-)
Great idea! It still needs to be tested of course for an accurate feedback, but the idea itself is really cool.
Only thing that I am concerned of is the lack of a policy and terms of use. Basically we are sending you our address and key and there is no legal policies or terms of use on the website concerning the use of these pictures, how they will be used and stored. I think this is the main concern I see.
Only thing that I am concerned of is the lack of a policy and terms of use. Basically we are sending you our address and key and there is no legal policies or terms of use on the website concerning the use of these pictures, how they will be used and stored. I think this is the main concern I see.
How do you deal with the profile of a key?
Right now I read the code at the top which generally (>95%) gives you enough info, there are some keys I can't do yet.
I've trailed some heuristics to infer the profile based on the side view which look feasible but will require a fairly large validated dataset.
I've trailed some heuristics to infer the profile based on the side view which look feasible but will require a fairly large validated dataset.
Yep security is the biggest user concern, my plan for this is centered around keeping e-mail addresses and real addresses apart.
Ultimately though through my research on this I've discovered how trivial it is to get through most locks so it would just be way too much effort for a thief to use DittoKey.