Unfortunately, and unlike a QR code, the data is not encoded in the drawing itself, which means an Internet connection is always needed. For URL sharing the Internet connection requirement is not so much of a problem (you'd always need that connection for browsing to the URL, unless it's a local or data URL), for everything else it may be.<p>Also, once the Meshtag service goes down or closes for good, all tags become impossible to read, which pretty much puts me off of using this for any permanent purposes. I suppose it will also have to deal with copyright infringement issues (and there would need to be a mechanism for submitting DMCA notices), but don't take my word for it. Related: it's easier to censor.<p>Whether these disadvantages are enough to make up for the fact that the codes can be drawn by hand, and can expire, only time can tell.<p>By the way, if this could be made to use squares instead of triangles, I guess it would be easier to draw on grid paper.
This is nothing more than a (pointlessly!) complex URL shortener... Meshtags are exactly equivalent to the following hypothetical service which I call "SuperNumericShortener":<p>1. Pick any random number, say 1234. This is your identifier.<p>2. Register this identifier and your URL at SuperNumericShortener.com<p>3. Write "SNS-1234" anywhere you want.<p>4. Install a fancy smartphone app that can take a picture of a writing and recognize "SNS-1234".<p>5. The app opens your browser to <a href="http://SuperNumericShortener.com/1234" rel="nofollow">http://SuperNumericShortener.com/1234</a> which redirects to whatever URL you chose.<p>That's it. Oh and you can pay to reserve short identifiers (since they are likely to be claimed by other users of the service).<p>And for some reason meshtags use triangular shapes instead of numbers. Maybe it makes the concept look fancier than what it really is. Or maybe the author thought it would be easier to recognize a meshtag than detecting the shape "SNS" and do OCR to extract the following identifier? Either way, "SNS-1234" is easier to memorize, faster to draw, can be communicated in writing and speech. All advantages that meshtags lack.<p>Edit: here is something I would really want: in my Android Chrome address bar, next to the microphone icon, I want a camera icon that lets me snap a picture of a written URL, and opens it directly. No obscure QR code, meshtag, or whatever.
It's a confusing solution for a problem that does not exist, which is how to use a pen to draw readable shapes, a problem many of us learned how to solve in elementary school.[0]<p>[0] <a href="http://www.auburn.edu/academic/education/reading_genie/letters.html" rel="nofollow">http://www.auburn.edu/academic/education/reading_genie/lette...</a>
Thought I'd clarify generally a little bit. QR codes definitely have a lot of advantages over Meshtag - and they're perfect for what they do. I'm by no means suggesting this could "replace" QR codes.<p>Where Meshtag fits in is that it's aimed at users and not companies. I think a large reason most people don't scan QR codes is because they assume they're just trying to sell you something. But if you saw a meshtag that somebody drew on a wall somewhere, wouldn't you be curious?<p>Also, thanks for all the feedback! I have dreamed of being on hacker news since I started coming here. Any UI/UX designers in the Boston area, hit me up if you want to get involved!
Even though everyone else is talking about this as an impractical solution, I think it's clever at least.<p>I don't know that I would ever use it, but I like the novel concept of it. If this was just a for-fun kind of project, I'd think it was pretty awesome. If you're trying to market it as a business, you're going to run into all the problems everyone else has already mentioned.
Sam, Cool stuff. I get what you're doing here, but from these comments I think QR code reference really throws people off. While comparing to QR might explain how it works, I would do my best to position this as nothing like QR codes (and highlight use cases that make it unique from QR) to avoid the comparison.
> No. Unlike QR codes, meshtags of any color on any background can be scanned. There just needs to be enough contrast between the meshtag and its background, and the background must be a solid color.<p>qr codes can be any color, just need to have enough contrast, also depending on the redundancy level of it you can actually omit a significant fraction of the qr code and still have it scan
Damn, whole lot of hate in these comments. Chill out guys, nobody's trying to kill QR codes here.<p>samfpetersen, you made a cool thing. Expiration of small tags is an interesting way to solve the landgrab problem. I think these would be <i>fun</i> to use and I hope I see some around.
I don't really get why I'd use a Meshtag over a QR code.<p>Pretty much everyone has a QR reader already, whereas this requires people to download a new app (for no real gain on their part either). It also requires them to have an internet connection, which for any thing other than a URL, is really not necessary.<p>The problem it says it's trying to solve, making them drawable, isn't really a problem at all. I've never wanted/needed to draw a QR code, that's what we have text for.
I think everyone here who's pointing out that this is equivalent to a link shortener is missing something: the majority of QR codes in existence right now... are encodings of link-shortener URLs.
This is cool, though as said in the other comments, it's unfortunate that the readability of the codes are tied to the availability of the service, like a URL shortener, i.e., not very mesh.<p>Did you do much research onto the actual "serialization format" of the tags? Aesthetically they are nice, but I wonder how deep the rabbit hole goes for drawable-but-not-necessarily-human-readable dense formats, which is a slightly different utility function than regular written languages. I imagine there's some overlap with the psychology of memorizing passwords which are opaque semantically, but still need to be remembered.
I much prefer QR Codes as there is nothing to buy and their embedded data makes them standalone. Meshtags seem more like a graphical version of bitly or tinyurl.
I think a lot of people are missing the point here.<p>What we basically have is a pictograph URL shortener. This may be complex from the computer side, but from a human side, this could be vastly simpler.<p>E.g:
- Draw a Triforce and go to Nintendos Zelda page.
- Draw a big triangle, with a line across the top to make a mountain, and go to some page about Everest or something.<p>And since it's the segments that matter, the possibilities are endless, spokes of a bike wheel, a pizza, pacman, airplane. It also easier to remember than a url, and if your memory is a bit fuzzy, just start drawing and you can figure out if it looks funny.<p>Good job Sam!
hey, this is wonderful. it would be great if street painters adopted this to communicate with their audience. i would love to go into a tunnels with cool street paintings and scan these signs and get access to playlists, videos and messages that the painter wants us to visit. these cool triangles could become symbols of an underground movement if promoted in these communities.
As a tag system, this doesn't seem to have any advantages compared to QR codes, and it has several disadvantages (central authority, possible conflict, only encodes the key rather than the direct value).<p>However, this system does have an advantage for other purposes: those where the ability to reproduce it from memory is important. QR codes aren't memorizable; meshtags seem more so. Consider how many people prefer Android swipe-unlock patterns rather than PINs or passwords. Many people have good spatial memories, much more so than their memory for arbitrary strings. You could use this in cases where the property of being able to reproduce this from memory is important. Figure out how many bits of entropy this can represent, create and document a perfect 1:1 mapping between meshtags and an encoding, and target use cases QR codes can't handle.<p>For that matter, how robust is this against minor errors, and how high-resolution can you make it and still reliably resolve it? Can you pack this more densely than a QR code and still reliably read it? If so, you've got something particularly interesting.
I've never in my life wanted to draw a bar code or QR.. Also I can hand-write a URL which both a human and OCR can read. So this all seems pointless.. outside of it possibly being a fun exercise to par take in.
<a href="https://www.youtube.com/watch?v=kW39Mt5kscQ" rel="nofollow">https://www.youtube.com/watch?v=kW39Mt5kscQ</a><p>Another interesting alternative. Prettier than QR and IMO meshtag.
After playing with your app for a while, I have following feedback:
1. It could be some kind of niche fun, for example in case of urban/alternate/augmented reality games.
2. Drawing properly recognized tags requires significant skill - about 30% of my drawings weren't recognized at all.
3. Some kind of scanning-progress-bar or other indicator wold be helpful for impatient users :)
Taking into account points 2 and 3, it would be beneficial to improve users onboarding if you think about mass adoption.
As many have already discussed that this is akin to a URL shortener, and not stand alone (like QR is ), I have a different question.<p>Your app has in-app-purchases. Please provide examples how you are monetizing this?<p>Also, I notice you use a G+ identity to tie all codes to, yet your pictures shows "Anonymous" as creator. Is that really anonymous, or fakenonymous?
So this is a program that can analyze an image for a structured drawing, encode it as a unique bit-stream, and translate that bit-stream into a URL.<p>I can very easily draw "<a href="https://goo.gl/e0aeeD"" rel="nofollow">https://goo.gl/e0aeeD"</a> on a chalkboard. Android users could use Google Goggles to grab the text without typing it in, and paste the URL into Chrome. If I had to, I could leave off the "<a href="https://"" rel="nofollow">https://"</a> part, and it could still work.<p>I get how it's easier to process the triangle-based image into a usable bit-stream than it is to do OCR, but how does that make things easier for the user?
There was a Media Lab project that did something similar 2 years ago -- and is open source <a href="https://github.com/JeremyRubin/Graffiti-codes" rel="nofollow">https://github.com/JeremyRubin/Graffiti-codes</a>
So, even though this is kind of fake (it’s more like a hand-drawn UPC barcode alternative), a <i>real</i> hand-drawn QR-code alternative is something I’ve been thinking about doing for a while. In particular, I was thinking that matrix parity codes like the ones schoen at the EFF discovered in printer dots <a href="https://w2.eff.org/Privacy/printers/docucolor/" rel="nofollow">https://w2.eff.org/Privacy/printers/docucolor/</a> would be an ECC that’s practical to compute by hand, and then you could draw ECC-correctable matrix barcodes with a pencil.<p>Has anybody done this?
Although I agree that this doesn't seem like the most useful idea, I do think it is really cool and it sounds like some interesting image processing.<p>To the OP, how did you come up with the idea? Is the algorithm from image to URL complicated?
I really like this - I think it would be an awesome way for people in cities to leave ephemeral easter eggs with chalk on a brick wall, for instance. Just because a lot of folks don't see the business value or purely technical use doesn't mean it's not awesome.<p>Or maybe make a meshtag scavenger hunt, with each tag leading to the next clue. Sure, you could do it with a QR code, but these look cooler, are ephemeral by nature, and have more of a fun feel to them.<p>The image processing code was probably really fun to write, too. Kudos on a really awesome project.
A problem I see with this is recognition. A QR code can be printed onto a piece of paper with no further information attached, and even some less-techy people will know what to do with it, because it's clearly a QR code. These drawings are pretty ambiguous, and unless it catches on with the less-techy crowd you'd have to add an attached note "Go download the Meshtag app to read this!", which would probably not happen too often.<p>They are however, much prettier than regular QR codes. :)
Has anyone tried anything like this, but instead of posting the information associated with the tag in a private database, to post that info into the Bitcoin Blockchain?
Doesn't it look like a replacement for the (dead) Microsoft Tag [1] ?<p>[1] <a href="https://tag.microsoft.com" rel="nofollow">https://tag.microsoft.com</a>
I think this is a great idea. I'm surprised to see so many negative feedback. To me the great thing about it compared to a QR code is that you can draw it by hand which also means you can cheaply make it as big as you need it.<p>The aspect that it depends on a central service make it less dependable than an QR code and I'm wondering if this could be fixed. Still, I think it's a really interesting idea.
QR codes and such I imagine will fade as more applications support live OCR. Its only a matter of time before phones will ship with the native app using text detection to make hyperlinks clickable from the camera app. There's already text grabber and the like on iOS/Android. This seems like a stop-gap. It's cool but I just can't see using it.
<p><pre><code> "A meshtag is a "drawable barcode" that you can create
by drawing straight line segments at 6 different angles."
</code></pre>
I count only three angles. The entire grid can be constructed via a series of parallel lines extending infinintely: one horizontal, one diagonal left, one diagonal right. There aren't any other angles.
This could be useful for dead-drop style communications. Should be easier for two people to memorize how to draw a reasonably complex shape than an arbitrary string of bits, numbers, or letters. If you can do that there's no risk of leaking the drop location: just draw it, scan it to check the drop, then destroy the drawing.
I would only be interested in the retrieval of the raw data of a meshtag. It's of course smaller than the shortened url. I always wanted to make a simple method to mark my handwritten notes, scan them, then the scanned pages "organize themselves" by the hand drawn tag (in this case meshtag) at the top right corner.
Or you can colour a QR in by hand (I did this to test the concept of storing 128bit salts on custom credit cards).
<a href="http://imgur.com/ClCeHAj" rel="nofollow">http://imgur.com/ClCeHAj</a>
I mean this honestly. Do people actually even use QR codes?<p>Filling up your phone with a bunch of hi-res pictures of bar codes, not to mention having to load:
1. QR code app
2. camera app
3. browser app<p>These things are 100% in the gimmick column in my life.
Personally, I think this really cool and hope it sticks around for awhile. If I saw meshtag graffiti, I would absolutely scan it, and that's the kind of use that's exciting to me. Awesome stuff!
How many bits can it encode? I bet not many.
And I bet it is far easier to just write down two or three characters like "Am3" or "IX8" and use OCR to extract the information.
I think it's a mix of QR code plus Microsoft Tag. Tag suffers from the problem of requiring an internet connection but they do provide stats on how many people scanned the tag.
This is really neat! It does seem like it is much more prone to error than QR codes, though. Would a meshtag that is rotated 120 degrees represent the same one or a different one?
Very novel, but I feel like its fundamentally flawed; for most people it would easier to remember a URL than it would be to remember a 'meshtag'.
I still remember how this went.<p>Connect the dots… la, la, la-la.<p><a href="http://i.imgur.com/FZrXlee.png" rel="nofollow">http://i.imgur.com/FZrXlee.png</a>
As people have pointed out, this tag is reliant on the service operating. What about using Bitcoin's blockchain to store URLs/bytes? Make the tag a digest of a public key. Physical and digital graffiti.