Slashdot Mirror
"Pull" Barcode Scanning Could Be Android's Killer App
Seor Jojoba writes "The release of T-Mobile's G1 Smartphone is shifting focus away from push-based barcode scanning, where embedded URLs send you to locations of a vendor's choosing. There is now more interest in pull-scanning, where product information is retrieved from user-specified sources. It may be that QR-Codes and other 2D barcodes will have their thunder stolen by 1970s-era linear barcodes. On the iPhone, scanning a 1D barcode is slow and unreliable. But the G1's improved optics and Android's improved access to image scans has made 1D scanning quick and useful, opening the gateway for killer apps that help people make spending decisions."
Let's just hope Google (and her telco partners) don't fuck it up.
How we know is more important than what we know.
Or a CueCat. We know how big of a killer app.
If this takes off, it'll result with me waiting in the supermarket checkout line for 5 minutes behind some idiot arguing with the cashier because his phone says a different price to the register. As if phones in supermarkets haven't caused me enough grief...
Commodore64_love: I don't comprehend people who're so frightened of death that they'll bankrupt themselves to stay alive
Oh wait, it wasn't.
This is spot-on. Yes, many years ago there was an attempt to invest heavily in barcode readers - the Cuecat, in particular, was a well-funded attempt to bring barcodes to the masses. But due to a major error in their business model - a grave error - the 'cat lived an extremely short life.
Jump ahead to 2008. People are buying fancy telephones, and there are barcodes everywhere. Google is in a unique position to read and process these barcodes on the fly - using a well-connected application living on a mobile phone. Next thing you know, you'll be able to go to the store, pick up a six pack of Bud, and scan in that barcode. THEN you can find a cheaper vendor - maybe down the street. YOU WIN due to CHEAPER BEER.
And we know that the world, with its flailing economy, will certainly needs cheaper beer. The cuecat was just ahead of its time.
" On the iPhone, scanning a 1D barcode is slow and unreliable. But the G1's improved optics and Android's improved access to image scans has made 1D scanning quick and useful,"
[citation needed]. 2 megapixel is more than enough to scan a 2D barcode, and the iPhones optics are quite reasonable. If the iPhone scanning is actually "slow and unreliable" (I have no evidence of this) it's simply because of the algorithm that the third party developer is using.
For what it's worth, though, the iPhone has 7 scanners on the App Store when you search for "barcode" and all seem to revolve around one kind of 2D bar code or another (EZcode, DataMatrix, QR Code, ShotCode, Blotcode, etc). The reviews seem to indicate the iPhone is quite good at scanning them.
Basically, the article submitter appears to be another anti-iPhone troll, which is too bad because for me it detracts from his main point about bar code ubiquity.
E pluribus unum
Any money you save will be eaten by the cost of gas...
Anyway, who needs beer when you can make your own moonshine?
(Poster not responsible for blindness)
Fanatics are so fun to mess with, the moment you downplay something, even slightly, they will jump on it and try to jump around it to falsify and even rationalize around it.
I think this would be a very useful item. I know that the whole barcode thing is a little played out, but add in the wireless network and you have a way of comparison shopping.
Yeah, but at least he has one thing right:
The !phone sucks.
Mod me down, my New Earth Global Warmingist friends!
Enjoy your negative karma.
"On the iPhone, scanning a 1D barcode is slow and unreliable."
True. Some examples of 1D barcodes.
-
---------
----
--------
The question is, can your scanner tell the difference between barcodes 2 and 4? That's the problem we are currently facing.
Knowledge is power. Knowledge shared is power lost.
I have bar code scanning on my latest phone. It doesn't work. The camera just keeps going in and out of focus. Having never had much to do with barcodes in my IT work, I decided to look at open source bar code readers and scanned in the bar codes on a few things (like my son's birth certificate). I looked and the standards and my own scans quickly found that often the number was often printed right beneath the barcode. Barcodes were made when computers were slow and had trouble doing OCR. They're a lot better now. Bar code scanning is still useful to some degree but to call it a killer app is a bit much.
These posts express my own personal views, not those of my employer
I used to work for a big (biggest?) developer of games for mobile phones and I think we used to sell games in Japan using some sort of barcodes (with squares). We would put game ads in magazines and the user would just scan the barcode with the phone and buy the game. So this is nothing new or that difficult to do, it just hasn't caught up in the west yet. Hmmm, so I just checked and it seems that those were the QR-codes that the article talks about. I never bothered to check before as we were just making the games...
I just did an experiment and indeed the phone does not seem to be able to take a reliable picture of a bar code. I don't think it has to do with resolution as much as the crappy lens inherent in cell phone cameras along with the the fact that cell phone cameras were not made for macro photography, a tricky proposition even with a real cameras. To take back the resolution thing, a higher resolution may let the software extract the bar from a normal, non macro, photo.
So here are my two questions. First, is the lens on the G1 that much better? Second, Isn't this fundamentally a software problem. A bar code is a defined form with a known and rigid structure. Even with a blurry/fuzzy photograph, it should be possible to clean up the bars. For that matter, why are we even dealing with bars. The numbers are there under the bars. Why not use those?
In any case, how many people use this application? This is the first I heard of it. I certainly don't go around taking pictures of bar codes. The only time I thought about doing it was for my library, but a scanner seems like a faster solution.
"She's a scientist and a lesbian. She's not going to let it slide." Orphan Black
Huh? The sentence you quoted is about scanning 1D barcodes, and you say the iPhone can handle 2D barcodes. Guess what? The point was that 1D barcodes are harder to scan than 2D barcodes (RTFA). 2D barcodes were designed precisely for lower-resolution cameras, but the downside is that most products still have only 1D barcodes. The G1 has a higher-resolution camera (3 megapixels vs 2) and can handle them better.
Are we seriously considering a bar code scanner a "kiler app"? To me, a killer app is one which makes you absolutely want it, even if it means making a different hardware decision. You know, like how Halo is a killer app for XBox. A barcode scanner might be neat or even nifty and, to some rare individuals, it might be an absolutely killer app, but for the majority of people I see it being nothing more than a novelty app - something that's cool to have and you use from time to time but, most of the time, you forget you even have it.
Then again, maybe the poster is using "killer app" in a different way than I would...
There's a fair bit of a difference between optics and sensor pixel count... 2 Megapixel tells me almost nothing about actual image quality. Other details are more important here, especially because barcode scanning requires very different features, like the ability to focus on nearby objects, that many phone cameras lack, regardless of the sensor.
Also, while you are speaking of 2D barcodes, 1D barcodes are a very different matter. 2D barcodes work well with camera-based sensors, and are often designed to work well with phones. 1D barcodes are far harder to read with camera phones, and I expect that the iPhone, like every other phone I've tried, is unable to do so well. 1D barcodes require far higher resolution of thin parallel lines, and weren't designed to be scanned by camera; they also tend to need to be in focus. Unfortunately, these are far more prevalent than the 2D barcodes that are easy to read.
That said, I'm doubtful that the G1 will be able to read 1D barcodes well either, unless the optics have been designed to facilitate it. Better optics doesn't imply that the optics are better for such a special case.
I really fail to see what you are trying to say here.
Unless I misunderstand, you are claiming the submitter is a troll for claiming the iPhone has trouble reading 1D barcodes because it can read 2D barcodes?
Are you aware that a 1D barcode is the kind of barcode you find in a supermarket while a 2D barcode has been engineered to be easy to read in noisy pictures using poor cameras?
I don't have any trouble believing that an ordinary cameraphone will have trouble counting the zillions of tiny parallel lines in a standard 1D barcode.
I find it sickening that you have been moderated up for a baseless attack on the article submitter.
Ok... TFA pushes the idea for what would essentially be a product database.
You scan the bar-code, it gets sent to the server, which returns useful data to you.
OK... I can see how that should be useful to consumers as well as a hypothetical company that makes its living out of contextual commercials.
BUT... The TFA goes on and on about how it MUST be 1D barcodes and NOT 2D barcodes - despite the fact that 2D barcodes are easier to read for mobile phones because of redundancy and greater bandwidth.
And since The New PhoneTM has the optics that can FINALLY read 1D barcodes - let us make a database that handles ONLY 1D barcodes.
Cause... there is like a lot of them out there.
Hmm... how about this GROUND BREAKING idea I just had.
Make the "killer app" capable of reading both 1D AAAND... wait for it... 2D barcodes.
HA?! Isn't THAT fuckin' brilliant or what?
At the cost of... umm... nothing... you get a "killer app" that works on The New PhoneTM AND all those phones out there already.
Which it would be pretty stupid to just disregard.
Cause... there is like a lot of them out there.
Mit der Dummheit kämpfen Götter selbst vergebens
Sterno -- the drink of choice for the discerning hobo.
Have you even used one of those apps? Of all the 2D barcode scanners not one is rated more than 2 stars. Ive tried lots of them and none is as reliable as the Zebra Crossing app available for Android.
If you read the WHOLE summary, it says that "focus is shifting away from" URL's of the vendor's choosing, to precisely the situation the person you doubting described, which is URL's chosen to best benefit the consumer.
Finish reading the summary, please. That's a description of push-scanning, while Android and Google can provide pull-scanning.
A barcode scanner might be neat or even nifty and, to some rare individuals, it might be an absolutely killer app, but for the majority of people I see it being nothing more than a novelty app - something that's cool to have and you use from time to time but, most of the time, you forget you even have it.
Then again, maybe the poster is using "killer app" in a different way than I would...
That or the poster has worked in a warehouse. I've worked in one that's completely barcode-driven.
There is an app (Delicious Library or something) for OS X that reads 1D bar codes with the built-in iSight camera. This does 640x480 and has pretty cheap optics (much worse quality pictures than my cheap camera phone in low-light conditions). In spite of this, it is very accurate at reading bar codes. I tried the demo, and it correctly read things even when I held them at strange angles.
I am TheRaven on Soylent News
That's the push model. Embedded URLs are in 2D barcodes. 1D barcodes just list the manufacturer and product.
Android uses the pull model.
Interesting, but I think the real problem is that a 1 dimensional barcode is impossible for anything to see.
"Linear" barcode I can handle, but they still have 2 dimensions. Likewise a 2D barcode should probably be called quadratic. Then just call them L2D and Q2D Barcodes or something.
Just bitching, so ignore me.
If you'd bothered to read it all, he explains that it's NOT what the Android app does. This is how regular 2D push barcodes work.
---- Sig. gone.
Buy some beer, keep the barcode, and scan whenever you need more beer.
The zxing library can read about any 1D barcode in a matter of seconds, at least on the G1, if there is sufficient light.
I quit working on the VNC for android phones cuz I didn't have a phone network expert any more.
Now... they have exactly another product I wanted??? LOLZ... scan, SOLD! lolz.
The submitter is quite right. I have an iPhone, and the biggest challenge with doing as the camera suggests (a coworker of mine had the same idea) is that it uses a fixed-focus lens, set to 'infinity', which means that it cannot focus on near objects - so the barcode has to be far enough that it's within the focal range, but big enough that it can be seen from there.
... for me the killer app of the only-available Android phone is the keyboard, as I don't like touchscreens (I prefer to type than touch a flat surface as I can more easily do it without looking), and also the promise of a full bluetooth implementation and an uncensored marketplace
I suspect 'killer app' is always going to be a subjective & personal interpretation. It's not a good way describe a feature.
I have been a user for about 10 years. This ends Feb 2014. The site's been ruined. I'm off. Dice, FU
Why are you talking about 2D barcodes when even in your quote it talks about 1D barcodes. You physically had to copy and paste that text yet you still completely missed the whole point of what you're talking about.
Yet another apple fanboi foaming at the mouth for no reason. What idiots.
Spot on. My 2 MP phone has frequent problems resolving both 1d and 2d barcodes. Not due to lack of resolving power or resolution, mind you, but becuase of the camera's focus, or should I say, it's lack thereof. Anything that small just comes up blurry, regardless of lighting conditions or focal distance.
I feel like we're getting too excited about something that's already completely ubiquitous in Japanese cell phones.
It seems hardly a killer app if it's already taken for granted on the other side of the world. Am I missing something? Is this somehow different than the Japanese cell phone 2d barcodes?
I got my start doing stuff like this on the PalmOS Symbol handheld scanners back in 1999. I've done this same stuff for years on various handhelds running mobile OS's. As long as you can scan a freaking barcode, you can store that info and hit that website when you sync...whether it's through a wired connection, a wireless connection, it doesn't matter.
You can reinvent something 10 years later that people have done for years, and now it is a "killer app". If Google does it, apparently, idiots pay attention and it is suddenly, somehow, feasible and marketable.
1D barcodes just list the manufacturer and product.
Which gets you what? Manufacturers and retailers have been dragging their heels on providing customers the ability to pull competitive price data over the 'Net. With the advent of loyalty card programs, there's now a retail price for the non member, a posted member price and an even lower price depending on who you are, what you spend, what your zip code is. Maybe even your credit report data. Only the sucker prices are advertised anymore.
Have gnu, will travel.
Oh baby, don't be so negative.
I'm karmkarmakarmakarmkarmachemeleon...
Mod me down, my New Earth Global Warmingist friends!
If these billions of products were instead marked up with 2D barcodes that provided the same unique identifier, that would be an even better situation, because the crummy cams on current mobile devices would have more success reading them. However, 2D barcodes aren't planned to be used that way. Instead, they're intended to push promotions and other vendor-supplied content at consumers. If you want to benefit from the "pull" model, your phone has to read Ye Olde Barcode from the 1970's.
Just because you can put a URL in a 2D barcode, that does not mean you can't put a simple text code there instead.
Mit der Dummheit kämpfen Götter selbst vergebens
The question is, can your camera focus well enough to see as fine lines as bar codes use.
The pixel size used in 3D codes is simply larger by comparison.
Bar code lines are the same length, which merely has to be wide enough to make a linear scan cross the code easily.
I believe the US Postal service uses a varying length code for mail routing.
Finish reading the summary, please. That's a description of push-scanning, while Android and Google can provide pull-scanning.
I think you must have PUSH and PULL reversed. From the article:
In a nutshell, here's how [the PULL model] ShopSavvy works:
1. You scan a product's UPC code. It could even be a sandwich, I suppose, if it was the pre-packaged kind.
2. A list of places offering the product are shown, complete with prices and reviews.
With the "pull" model, you are getting information about products from wherever you choose.
Exactly the same model the grandparent post described.
You may not like barcodes, but if you attack the idea, you should at least know the terminology.
lol -1 retarded.
if this is a killer app my dick just fell off.
the iPhone failed because it had no killer app at launch, so will the android phones.
slashtwats... that was sacarsm.
Barcode scanners are the killer app? Wow, who knew? Someone better tell Steve Ballmer so that he can throw yet more chairs at Microsoft engineers for overlooking this game changing technology.
Then watch for "Store Shelf Explorer" to appear on a Windows Smartphone near you.
I'm trying, really, to imagine why I'd want this on my phone.
If I'm at Best Buy, and I scan a DVD and it comes up $2 cheaper at Wal*Mart down the street, I still don't think put up with driving, parking, and all that hassle just to save a few bucks. Now, if it were 35% or 50% higher, then sure, but that rarely ever happens in the retail environment.
And if it found it $2 cheaper on Amazon.com, then that's great. That at least gives me the choice... buy it now for $15 or get it in a couple days for $13. But I already knew that Amazon.com would be cheaper before I even walked in, so this is moot.
About the only useful application I see here is if the retail store is just hoping to get "lucky" by charging me a super-high markup on something that I'd be better off getting somewhere else. When this happens, I can usually tell I'm being gouged when I see the price. I don't need a CueCat-like device to tell me that.
That all being said, having bar-code reading integrated into employees cell phone can do a lot in the business world. If these APIs are easily tied into, I could see a whole set of custom apps for document/product tracking, inventory, etc.
-David
Google must see a future in bar-codes too. They are great for getting information for BUYING things and a lot of Google's future business will probably be mobile marketing and sales.
One of the $275K winners of the Android challenge was Compare Everywhere which uses a scanned barcode to find an item locally for sale. The idea is that you scan something like a book or a disc that you can then compare prices.
So if that cute girl at the office plays some music for you and you want a copy, scan hers and compare local prices (so you can buy a legitimate copy rather than rip hers). Then the app will give you directions to the store.
webcams are different from fixed focal length cameras.
most fixed focal length cameras are set to infinity. that means if you take a close range picture it's all blurry, beyond the ability of fast recognition. if the camera has higher resolution, the less the blur affects the recognition by software.
most webcams are set to a focal length of a few feet, or come with auto focus, or manual focus..
so a webcam can be lower res and have better image recognition, oh yeah and a laptop has a lot more processing power than a phone. that makes a huge speed difference.
also with higher resolution you can take the picture farther away, and still have enough pixels, it is true there are scanner apps for the iphone, but most likely they have compromised between speed and ability to read blurred photos.
https://www.gnu.org/philosophy/free-sw.html
i want to be able to call my girlfriend and well, give her a buzz
i mean this thing is called android afer all, right?
intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
I was at the T-Mobile/Google launch event last week in NYC, and had a chance to try this. I also have an iPhone.
First, this is not a Google-made app, it's called ShopSavvy and it's from a third party. It will come preloaded on the T-Mobile G1, though.
It's neat. It's very easy to use and returns simple links to product reviews and prices from multiple online sources.
vs. the iPhone:
Barcodes on the iPhone are NOT slow. They ARE unreliable, because the iPhone has a fixed lens that simply cannot focus on something up close.
The G1's "improved optics" is an auto-focus lens that can focus on things up close. That's why this works. It's very slow, though.
"Improved access to image scans" is bullshit. It's the same in Android as the iPhone or any smartphone, at least for something like barcodes.
MANY smartphones have a high-res camera with auto-focus lens and can run third-party software like this (which has existed for a while). It's nothing new. It's only in the news now because Google chose to feature it during their press conference and demo session at the event in NYC last week.
Also, the whole 1D vs 2D thing is beside the point. 1D is the type that's printed on all products at any SHOP, so of course it's the type that a SHOPPING application is designed to scan.
>apparently, idiots pay attention and it is >suddenly, somehow, feasible and marketable.
Idiots pay attention when innovations like touchscreen phones and mp3 players flounder and someone finds a way to market it to them later.
Idiots dont know if some idea is new or 10 years old. They wait for marketing department to tell them what they want.
Remember when they were telling idiots that you needed 40, 80, 160GB music players.
Then they told you that you only needed 4, 8 and 16GB...?
Idiots believe a lot of things because idiots want to believe.
Is it just me, or does the writer of the article confuse push/pull with Web 1.0/Web 2.0? The "pull" option appears to me to be a "service", as opposed to the "push" option, which is a specific result/page tied to a specific barcode. So 90's.
The software for scanning UPC codes isn't all that difficult, the problem is focal length. Most cell phones can't focus on close objects, such as UPC/EAN codes because they have a fixed focal range from about 1 foot to infinity. If you position the bar code far away from the phone, you don't have nearly enough pixels of resolution to get a good reading. One workaround on older phones has been to attach a macro lens, such as the one available for the Nokia 3650.
The Android phone must have one of the newer auto-focus lenses not available on earlier phones. It's also possible that its software can analyze multiple video frames and combine them to increase resolution, but this seems to take quite a bit of processing power which would have only recently become available.
This is not a new idea at all and numerous companies have patents on quite a range of techniques and applications for reading UPC codes on cel phones.
Get back in the car, burn more gas, waste more time in traffic. Whoohoo! Another nickel saved!
And you wouldn't research bigger purchases ahead of time?
This seems to be getting slammed pretty good in most of the comments, but I can see this as being very useful especially when purchasing tech items. Case in point, I have gone through several wireless routers, mostly because some of them really really suck, and some are fine. What am I to do when I don't know what is in stock locally and I want to know if the specific model I am getting is crap or worth buying? Go read reviews on Amazon, etc.
How do you know which of the 10 motherboards in stock at Fry's is a total piece of crap? Go home and read up on them? Sounds like a super way to spend time, going back and forth to read reviews when you need a computer up and running ASAP, but don't want a brick in a month.
While perhaps not a killer app, it would be something I would interested in. All assuming you can focus that close.
This is Mac OS X software, award winning sold 250K in 1st month! USing any cheap USB cam, bluetooth, cuecat modified, or TYPING in the number beneath the code. I found the story of how the spaces between the black lines were simple (after he found out) and his massive CD/DVD/Games/Pr0n were so vast he just wanted to keep track of who borrowed them... then a hot UI like iTunes & iPhone before Apple ever had coverflow. So, since Apple hired this guy, I can see his style all over iPhone, and he already did it! Wikipedia link: http://en.wikipedia.org/wiki/Delicious_Library So, Apple fanboys have been using this for years with FAR worse USB web cams what is that .3 megapixels?
So, it has already been a killer app! Using hardware in every Mac notebook, iMac, and now in any phone.... software only solution...
( Yes, it is a cocoa application which would run easily on iPhone with widget sized controls.... hmmm, that IS a great business plan)
TFA is iPhone hater, which detracts from the article. Of course if he were an Apple fan, he'd see how this gorgeous app works for every day folks keeping track of media...
(don't tell mje it's already up in the iPhone app store)
Like a Pokedex
Capturing data in this way is a killer app that justifies the whole expense of the device to me -- even if the device had no other features at all. Cordless barcode scanners are pretty spendy units.
So yeah the freedom is great. Let's not overlook that it's the freedom to share your killer app and so enhance the utility of this tool for people with similar needs to yours. There will be a lot more way cool stuff presently.
Help stamp out iliturcy.
Actual website http://www.delicious-monster.com/ It was quite a splash in 2004.
No, its still stupid. I already know where the Cheap Beer is. I also know where the Good Beer is. I also know how few people will buy an Android. I also know how small of a percentage will know how to use the feature. I also know how small of a percentage of people who will want to use the feature.
" On the iPhone, scanning a 1D barcode is slow and unreliable. But the G1's improved optics and Android's improved access to image scans has made 1D scanning quick and useful,"
[citation needed]. 2 megapixel is more than enough to scan a 2D barcode, and the iPhones optics are quite reasonable. If the iPhone scanning is actually "slow and unreliable" (I have no evidence of this) it's simply because of the algorithm that the third party developer is using.
For what it's worth, though, the iPhone has 7 scanners on the App Store when you search for "barcode" and all seem to revolve around one kind of 2D bar code or another (EZcode, DataMatrix, QR Code, ShotCode, Blotcode, etc). The reviews seem to indicate the iPhone is quite good at scanning them.
Basically, the article submitter appears to be another anti-iPhone troll, which is too bad because for me it detracts from his main point about bar code ubiquity.
"the Android O/S that runs on the G1 doesn't require a time-consuming storage of a scanned image before it can be processed"
The increase in speed is because of using the image in memory, not the megapixels. It's great that you love your iPhone, but not everyone pointing out a better product is a troll.
Actually, there are several ways to get sharp near-focus images with the iPhone camera - you can use a magnifying glass, modify the optics (has been demonstrated to work), use a pinhole in a piece of black paper (clumsy but works), or buy a commercial case which includes a switchable macro lens (Google for "iphone case lens").
From my understanding Android hardware is not standardized which is a good thing, however there will inveitably be phones with lower res or no camera at all. I think the summary should be ""Pull" Barcode Scanning Could Be The Android based G1's Killer App". I could see immense value in this service to help consumers, can't wait!
I love the example that was given. If Google wants to buy me lunch every day, that's fine by me.
****
"I'd never want to join a club that would have me as a member" - G. Marx
What many seem to be missing is "the bar code image could have been processed by googles 150,000 servers" not the software on the phone. The real power of Android is it's ability to leverage googles services for pre/post processing. For example, lets say I want to add an effect to a picture I have recently taken. Now imagine the effect is way beyond the processing ability of the mobile phone. In the case of the android phone it seamlessly uses a simple image editing front end then for hard tasks it sends the job to Google (servers) for processing. This is were Apple has serious hole in its iPhone model, no back end network processing, or what they do have his seriously under powered, thus Apple will be stuck with ever more powerful processors chewing through batteries faster and faster until they start building huge server farms. Then that produces other issues, how will Apple charge for such services when Google dont.
I know, offtopic, but are their cheap scanners out there now? I'd like to inventory my CD, books, etc.
The built in iSight camera (that the poster referenced) is a fixed focal length camera just like most camera phones.
Uh, people...
Hasn't anyone thought about getting a cheap lens and holding it just in front of the iPhone's camera so that it can focus at macro distances? Sure, it would have to be the right size and thickness but if scanning barcodes with an iPhone is that important to someone then this is one workable solution. One good source for lenses is Edmund Scientific http://scientificsonline.com/.
The problem with the iPhone is that the fixed focal length lens won't take a decent picture of anything close up, so any kind of bar code comes up blurry or too small.
The Google phone is an HTC phone. all HTC phones (to my knowledge) have macro focusing on their lenses. My last phone was an HTC phone and I was able to take a picture of a business card and have OCR software in the phone auto-populate a bunch of fields for scanning into my contacts database. It could do it in English or Chinese.
fuck barcodes, *that* was the killer app. Shame the iPhone doesn't do that, either. It would be even better if you could just scan and OCR any text, like foreign language magazine articles etc.. and have them sent to google for translation or copy/paste bits into your dictionary. I am gagging for an OCR Japanese or Japanese-English dictionary.
Sure I can buy a little lens and stick it on the back.. but it's clunky, and the point is that because the *standard* iPhone doesn't focus properly, nobody is going to bother developing the software.
Oh yeah.. copy/paste.. another 'killer app' apple didn't bother to give us.
j'ai découvert une démonstration vraiment admirable (de ce théorème général) que cette si
yeah, regular people LOVE scanning barcodes. so fun!
If this concept of "i'll just scan this bar code and find out that there's a cheaper one across the street" becomes common practice, stores will just shift to having all unique bar codes. You can already see this with certain products. There will be a store-specific UPC/model number at Walmart vs Target vs Best Buy.
Hardly. 1D barcode means the standard barcodes you see everywhere - the bars are only significant in one dimension (width) so they're considered 1D. They're also designed to be pretty damn resilient, scalable, and many codes (EAN-13, for instance) also have a checksum bit. Code 39 follows a similar concept while allowing for alphanumeric codes.
In any case, all of the 1D barcodes are completely height-insensitive. It's the bar spacing and thickness that come into play, and since it's all based around a multiplier of sorts (UPC/EAN codes have bars of exactly 1x, 2x, 3x, or 4x the thickness of the end bars, which can only safely be within a certain range), it doesn't matter if you scan it at an angle or completely perpendicular to the bars, how tall the bar set is, and isn't too picky about being on curved surfaces.
The problem is that most phone cameras have appallingly bad low-light performance, and that's where this application would generally be used. Decoding the barcode is a fairly simple process, but if the image of the code is noisy as hell (as it will tend to be on phone cameras), the apparent width of the different bars could end up being completely unpredictable.
The problem becomes a hundred times worse with 2d barcodes (datamatrixes), like those that you see on UPS shipping labels. While the scaling issues are about the same, there's much more encoded data to get screwed up by a bad imaging system.
However, seeing that we're primarily talking about fairly short numeric codes and a PHONE as an input device, this whole problem is easily solved with the built-in keypad. It may not be quite as sleek, but it's certainly more reliable.
How are sites slashdotted when nobody reads TFAs?
That gets me the exact model of the product I'm looking at. Next, I use that information to hit review sites to see if it stacks up. Meanwhile I can also hit a couple price information sources to get an idea of the average price for the item. I'll then have some idea whether the item is something I should consider purchasing (alerted to lemons or major design issues) and whether buying it then and there is worthwhile. If I can't decide, I can then move that bit of data to a wish-list for later consideration - especially if it's a more expensive purchase or I'm just getting an idea of whats available in the market.
Would I do that for a six-pack of beer? Maybe not (although I might if I'm in the mood for an especially good beer). But I would for consumer electronics.
First you say it can't be done, then you say it can be done, but it's not worth it... both times you cite vendor reluctance to share prices.. but they have to make those prices publicly available... they generally call it 'advertising'. I really don't see any merit in either of your arguments, i'm afraid.
Fanatics are so fun to mess with, the moment you downplay something, even slightly, they will jump on it and try to jump around it to falsify and even rationalize around it.
Fanatics are so fun to mess with, the moment you downplay something, even slightly, they will jump on it and try to jump around it to falsify and even rationalize around it. That, or they'll just act like idiots. See what I did there?
Or a CueCat. We know how big of a killer app.
I think a lot of people forget that one of the big revenue streams for :CueCat was the ability to identify the purchasing (or browsing) habits of its users. The devices had unique IDs and people actually gave the company their name, address, and ZIP code. Whenever a :CueCat scanned an item and went to find out information on it, that UPC code could be associated with your ID, and -- viola! -- instant profiles of what you find interesting.
For a marketing company like Google, popularizing this would be a dream come true. I have no doubt that they would make use of this information to better target ads to you. Some people like the idea of ads only being things you want to buy. Personally, I don't want ads that are actually *good* at getting me to part with my money, and I have concerns in the wake of how Preferred Shoppers databases at grocery stores were using the wake of 9/11.
This just smacks of privacy invasion to me -- or at least further temptation to abandon one's privacy. Count me as not sold on it, but I don't really matter compared to what the mass market will think about it. I hope it flops, but I worry that it won't.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Somehow missed in the article that it was a 3rd party vendor that was doing this. The point still stands, even if Google is not the marketing company in question.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Let's hope they don't. But really, that's the nice thing about an open platform. unless they absolutely decide to kill it, it'll fly because the consumers want it to. And that's different from any other platform -- American cell phone systems have tried desperately (and largely succeeded) in absolutely killing anything the customer might want, because they see everything as a revenue stream ala ring-tones.
It's bizarre. If the customer wants it, the telcos gleefully KILL IT and give them a crippled, pay-as-you-go version. This when the cell phone manufacturers are begging them to take phone with features, so the manufacturers can get some market cred/traction. But no, the cell phone carriers demand that features in phones be killed.
Sigh. It's been embarrassing. You go to just about any other country and they've got better phones than use. Why? Because the telcos have the American consumer by the balls, thanks to a hefty lobbyist (read as "bribery") budget.
But unless I'm missing something, here, if a telco supports an Android based phone, the consumer gets control and whistles and bells. Period.
Hence, either telcos accept android based phones, or ...
They SAY they will and phone manufacturers make 18 models of android phone, and then the telcos say, "GREAT! We love it! Just disable this and this and this." The phone manufacturers say "Sure!" and the phones go out, and we fix them. This happens for one year, and the telcos start telling the manufacturers to drop Android, or they won't buy their cheaper, crappier phones in bulk. And the manufacturers will get very, very afraid, and mysteriously stop supporting Android.
We'll see. I hope this represents a real change.
---
It's not the acting. When just one actor stinks, that's acting. When they all stink, that's writing and directing. Mostly directing. And it's not that you get inured to it, Straczynski and his helpers got better at it.
Capturing data in this way is a killer app that justifies the whole expense of the device to me -- even if the device had no other features at all. Cordless barcode scanners are pretty spendy units.
So yeah the freedom is great...
How free is this? Am I free to use non-google searches? Can I input the barcode info only into a Google searchfield? Or can I use it with other websites, like maybe Amazon?
The Nokia CC-49 macro lens for their 3650 model does exactly this and makes it possible to decode UPC/EAN codes from software.
It sounds to me like this device will scan a barcode in Google's database and then provide Google search results for the barcode's item. If "push scanning" is bad because you're forced to go the the website of the manufacturer's choosing then this will be a much worse, much more extreme version of push scanning. Every barcode will take you to the same website (Google search), regardless of what product the barcode represents.
The next iPhone hate article will be titled, "iPhone is Farsighted".
Killer app? Yes, I say. Not so much for what it does for the individual, that's just nice. But think about what a killer app this is for economic efficiency.
People will be able to get instant comparisons of competing products. They will be able to find out which sunscreen really does protect even while swimming. They'll be able to know which digital camera provides better color reproduction. They'll be able to find out, in real time, which product serves their needs as a customer.
This will help them, sure. But think about the force this brings to bear to advance the free market. People will be buying products based on quality, instead of packaging. The network effect on product quality could be staggering. Products will improve, which will raise consumer confidence (the natural, healthy way). The economy could surge forward in a way unseen since women remained in the labor market after WW II.
Of course, that assumes we win network neutrality so that we remain able to share information without discrimination. That ability is a cornerstone of all the extraordinary advances that ubiquitous social computing will bring. Help advance collective intelligence: Learn about, and spread the word about, net neutrality.
Stop-Prism.org: Opt Out of Surveillance
Comment removed based on user account deletion
AC fails at reading the thread.
The post you are responding to was responding to this:
Do you think a vendor is going to undercut their distributor by directing you to a cheaper source? I don't think so.
2 megapixel is more than enough to scan a 2D barcode
This assumes ideal conditions. In low light or with somewhat oblique angles , 2MP becomes marginal. 3MP+ makes the difference between taking one casual shot and having it work, and having to try two or three times.
If a job's not worth doing, it's not worth doing right.
And we know that the world, with its flailing economy, will certainly needs cheaper beer.
Sorry to be the one to break it to you, but America != The World.
-- Language is a virus from outer space.
What? The article isn't even about a real app, it is speaking theoretically. Even the theoretical app described uses "user-specified sources."
Your point about making them publicly available isn't true. No retailer advertises ALL of their prices, and even those that are advertised may not be in an indexable, web-accessible way.
I agree with you that it will still be a useful application though.
For consumer electronics,
Scan barcode -> jump to NewEgg reviews.
Sweeeeeeeet.
Come on own up there must be some people doing it ? :-)
If I was to try selling a friend on getting an Android, I'd first get laughed at for the stupid name and then they'd laugh even more when I told them how good it was at reading barcodes!
No this isn't. It's possible to recognize 2D barcode on a 320x240 image taken in bad lighting conditions and at reasonably oblique angle in real time. And I'm afraid you won't find a phone with enough processing power to deal with 3MP image in a sane amount of time. By my experience, resolution is of much less importance than a nice lens and the ability to focus the camera.
Come on, as a programmer/designer this pisses me off. Only a complete and UTTER idiot would include price info in the barcode.
What if you had a price change? You would have to change the barcode on all your products.
As the article explains and anyone on slashdot could expected to know, a barcode (the 1D kind we are talking about here) ONLY has enough information for 10 digits. It is a 'unique' indentifier. The cash register scans this unique code and then looks it up in the stores database to get the price and whatever other information you could require.
To think that you would put the price of a product in the barcode is silly. ONE of the reasons why the switch to barcodes has seen the removal of price-stickers on products is that with barcodes you can easily change the price.
The OP simply meant to point out that he got the PRICE from the INTERNET with the unique code and is arguing that the price retrieved by the cashregister from the stores database is in-accurate.
And this discussion already happens daily in stores whenever there is an mistake made with special offers or a new product incorrectly entered.
My own recent story is of a frozen fries, used to be 1kg packages but suddenly they had 2.5kg packages but no record of it in the database. In the end, I got it for the price of 1kg while they went and sorted it out :) Got to love lousy math skills, a fair price would have been 2x the price of 1kg, but I suppose that was to complex.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
I live in Japan and bar code readers are included with every cell phone here. The bar codes are everywhere as well. In magazine advertisements or print ads on the train. They will take you to the product's web site.
There is even a zoo that has a bar code on the information sheet about the animal. Scan it with your phone and you are taken to the zoos site that has more information about that animal.
As for the quality I have never had any trouble scanning the barcodes that are used here.
Those are 2D codes, this is about 1D codes. So you are not intresting, you are lazy and redundant. Have a nice day.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
YOU WIN due to CHEAPER BEER.
Cheaper beer is not good enough. It must be Free Beer to be worthwhile. Well, that's what the FSF has been telling me.
... and then they built the supercollider.
The application you are using to scan this product is a internet capable gadget with a full browser and a keyboard. If it comes up with an unidentified barcode you could simple GOOGLE the product by its description, then enter the new code at a comparison side. One person needs to do this and then everyone else can use it.
The internet, you might have heard about it. It allows millions of people to pool their resources into common works.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
Nokia handsets have had 2D barcode reading for years now, with no apparent 2D barcode revolution in sight. Literally millions of handsets must have this already, and AFAIK it has only really taken off in Japan so far. So I fail to see why a relatively small number of new handsets would change this.
"I imagine it won't be long before you can search with whatever engine you can think of."
And then Google will release a standard search engine plugin which the majority of non-technical users will then simply use by default. Plus the Google one could even be pre-installed by default. Google then gets the information they origonally designed this feature for. The ability to know what products the majority of users are interested in. This is just like Google's way of profiling searches on their web site, to then workout from an advertisers point of view, what products people are interested in. Now Google will be able to extend that goal onto knowing what physical products we scan in. Its another step towards their goal of total information awareness on people.
It would be interesting to combine this barcode scanning data with GPS data, as that would then give a lot more infomation.
Google wants to becoming Big Brother. Because with total information comes huge power (even political power) and with huge power they have the potential to earn huge amounts of money profiling everyone, to then sell that data to advertisers. But then even political campaign's need to be advertised and marketed, so Google is aiming to become litrally Big Brother.
There are 10 kinds of people in the world... those who understand binary and those who don't.
It is arguable whether the grandparent is an "anti apple" troll, however he is right to say
The iPhone does not have Autofocus or rather, more specifically a "Macro" mode. This makes it MUCH harder to accurately focus in on and scan a 2D barcode, let alone a 1D one.
And I am not talking out of my behind either, as I am a developer who 6 months ago was working on a project for marketing company where we were using 2D barcodes with embedded URLs as part of a promotion.
We tested the system on many phones, including:
- HTC Tytn II (T-Mobile Vario III). The G1 android based phone will have an equal or slightly better camera. (3MP camera with autofocus, no LED, or flash)
- Nokia N95 (5Mp Camera with autofocus + LED light)
- SonyEricsson K800 (3.2Megapixel with Xenon Flash)+ K750 (2.0Megapixel with LED light)
-Apple iPhone (2 megapixel, no autofocus, no LED light)
-Moto Razr V3i (1.2 Megapixel, No AutoFocus, no LED light)
-plus a few others.
Using the same algorithm, (java for the normal phones, Symbian for the N95, MacOS for the iPhone, and Windows CE for the TyTn)
The iphone, n95, and the Tytn (as well as other smartphoens running native code) did a "live"scan, as they scanned on the fly. The Java based phones had to take the picture first, before scanning.
All the cameras without autofocus performed poorly. The v3i with its non live scanning required many attempts at taking the photo before it could recognise the barcode. The iPhone had to be held at the "right distance" (a little further away than comfortable) to have enough focus to resolve the 2D code, and was simply too far away to scan the 1D code in any reasonable resolution to resolve.
The k750, despite being only 2MP did very well on the 2D codes, thanks to the AutoFocus and macro capabilities,Its 1D performance was not so good, despite getting the best possible image, it was often unable to resolve the exact sizes of the bars, to be accurate.
the K800, and the TyTn were much better at both 2D and 1D, and the Tytn had the advantage of live scanning, which allowed quicker resolution, as u just adjust your hand until the phone picked up the code. (The G1 android will be similar to the TyTn in camera performance)
The Nokia N95 was the best at both 2D and 1D, and indeed comes with a barcode scanner as a built in app, which was even better than our app, as it used the accelerometer to judge the camera shake.
So although the iphone is "useable" with 2D barcodes, it was (apart from the V3) the poorest of therest, as it required a steady hand, a specific camera to subject distance. IT was pretty much unusable on 1D, as we were unable to get it close enough to use up as much of the sensor (and therefore the 2MP resolution) without loosing focus. and in general, would be a frustration to use in the supermarket (even with 2D) or on the train for example.
Dotn get me wrong, I am NOT anti Apple, and we WANTED to have a version fo the software on the iPhone, but frankly it just was not usable enough for the average person with very little patience. And its 1D performance was simply not up to it.
So yes, as the experiment with the K750 (And anotehr dedicated 2MP Nikon) has shown it is POSSIBLE to scan 1D on a 2MP, its resolution is, even with great optics, and macro still very difficult. Taken away the macro, and its impossible.
Have a nice day!
I have written a post to this, where I have actually had to test various phones with 1D and 2D barcodes. You right about 1D barcodes, but you missed a third factor, Focus, or more specifically Macro.
- Good Optics ensure the accuracy of the image, reducing barrel, and avoiding distortion (which will kill 1D accuracy more than 2D)
- Focusing, and Macro to get as close to the bar code as possible, to "fill" the image as much as possible.
- Megapixel count, to get as much detail of the line widths to accurately retrieve the info.
The G1 will be reasonable with 1D barcodes (we have tested with a 3.2 MP HTC TyTnII which has either similar or poorer optics to the HTC made G1). IT works better with flat "Standard" barcodes (the ones which are about 1 - 1.5 inch wide). it is not good enough for really wide 1D barcodes.
its 2D performance is excellent.
Have a nice day!
Well, I don't know about him, but I _have_ been stuck in a line while someone is arguing with the cashier what the price should be.
E.g., I wanted to buy some computer component at some point, so I go to a local small computer store. What do you know? Both guys behind the counter are stuck respectively with:
1. Someone who couldn't decide if she wants her new computer without a power supply or without the CD-ROM drive, because she apparently didn't have the money for the complete sum. So she's standing there debating the merits of getting a computer that won't start, vs a computer she can't install stuff on.
2. A couple which wanted to buy some TV decoder card. But they had a price list from another shop, for another product, and were prepared to argue all evening why the product in this shop should cost the same. The talk went roughly like this:
Buyer: "But your competitors sell the same thing for less!"
Clerk: "Well, it's not the same thing in the first place. That model is from another manufacturer and does less."
Buyer: "But they sell it cheaper! Why can't you match their price?"
Clerk: "Because it's a different product and has a different price. See, this one also has <insert list of features>, and the manufacturer sells it for more."
Buyer: "But I don't care much about those features, so for me they're the same. And those other guys sell it cheaper."
Clerk: "So buy it that model from them, then."
Buyer: "But I want this one..."
Clerk: "Well, ok, that'll be X euros then."
Buyer: "Why do I have to pay that, when your competitors sell the same thing for less?"
Clerk: "Because it's not the same thing."
That conversation ran around in circles like a broken record for half a fucking hour. Well, probably more, because it was already going when I arrived, and it was still going when I just left the shop half an hour later. There's a huge line of people inside the shop, people getting fed up and leaving all the time, but the idiot just won't give up. He wants product X at the price of product Y, 'cause they're in the same category.
I can just see this kind of thing happening with a camera phone just as well. Only now the idiots don't even have to get the price list of the other shop first. They can now do it on the fly.
A polar bear is a cartesian bear after a coordinate transform.
What if you had a price change?
Some countries have laws that allow or mandate price fixing. In several European legislation, books - for example - have a fixed price. Can happen with other goods too.
In those countries, some barcode scheme for some of these product can often also contain an infromation about price range or "official" fixed price - beside the few standard info you would expect from a barcode (product categorie, unique type identifier, producer, etc.)
That doesn't stop some seller to find way to circumvent the official price and selling the goods at a discounted price (the cash register can still look-up a databsed using que product's unique ID#). But the official price may get embed on the barcode.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Let's say you are shopping in a used store, perhaps books.
You could scan the UPC on a used book, compare it to the numerous sites that buy books from consumers
, like http://www.betterworld.com/buyback.aspx
, and also scan the sell side of the site
, and know if you are getting a good deal and how much you might be able to sell the book for when you are done with it.
This kind of data access gives massive benifits to the consumer. It could work for books, cars maybe (vin#?), furniture?, clothes?, all kinds of stuff.
btw: Android is a spooky name, it's gotta go. Secratary, Drone, Blib (short for web librarian), so many easy to make, cool, sexy, and fun names. :-)
Ummm, how about, you know, typing in the product name? Every product, even ones without a barcode, have a name. Do we really need technology for this?
And what does this have to do with Android? Certainly no-one's stopping you from making one on the iPhone -- in fact, there is one, I just downloaded it. It sucks.
Ahhh, I guess I haven't sprinkled enough "open" and "platforms" around my workstation, that will magically make it work better.
Maury
Well played!
Of course if he were an Apple fan, he'd see how this gorgeous app works for every day folks keeping track of media... ... and if he were an Apple realist, he'd see just how closed every Apple platform actually is.
iPhone is closed, Android is open. That's pretty much all I need to know to make any buying decision.
Gentoo Linux - another day, another USE flag.
Griffin Technology makes an iPhone 3G case called the Clarifi with a built-in mini-macro lens. Perfect for this sort of thing.
What? The article isn't even about a real app, it is speaking theoretically.
And I was commenting on the theoretical application.
Even the theoretical app described uses "user-specified sources."
So, for every barcode the user scans, the user is expected to pick a website to associate with the barcode and then scan the barcode to go to the website? Or do they pick one website to handle every barcode and bring us back to my original comment?
Haven't I seen this as a "killer" app before?
Oh yes, now I remember.
Presumably the latter, which does not bring us back to your original comment. You would set it up to search specified database for product information. Push scanning is bad because the vendors have no incentive to give you useful comparative information, it is a marketing ploy. User control of the source defeats every downside of push scanning, because it makes it NOT push scanning.
Local (GPS-based) price-comparison also needs to take stock availability into account.
What if your local store already sold-out of something?
What if the product sells-out by the time you've hiked across town?
How much work is it going to be to hook into every stores stock-control system and implement "Reserve this item"?
Two basic problems trying to scan store codes and convert to UPC
1) the store's internal numbers may not be unique (may clash with another store, of they may re-use a number again and again as stock comes in and out.)
2) The barcode will need different decoding software. Taking Borders as an example again, their barcodes contain far more stripes than a standard UPC barcode. More stripes require greater phone resolution and clarity to accurately scan. And different decoding algorithms (the barcode could be alpha-numeric instead of all-numeric.)
BTW: UPC is dead. Every US store is supposed to support EAN now (13-digits world-wide standard vs 12-digit US-centric scheme)
That's one reason common barcodes start out with a "guard band". That is, a sequence consisting of bar-space-bar, which allows the scanner to determine the size of the symbols. There's also limitations on the number of the same symbol you can have in a row. Same sort of thing that's done in magnetic formats and RF transmission.
Now I know why.
(I work on the zxing project which powers the ShopSavvy barcode reading)
Pixel count is not the issue, yes. You only need a clear, say, 320x240 shot to decode a 2D barcode.
Focus is a much bigger issue, all the more so for 1D.
The G1 has a solid auto-focus camera (and yes plenty of resolution) which makes 1D scanning quite a snap:
http://www.youtube.com/watch?v=3LkNlTNHZzE
I was thinking the same thing, but having to buy an external case+lens is a barrier to entry for people. Now, those who already have the phone and want the case could certainly avail themselves, but beyond that, it makes it difficult to design a software product that requires a separate hardware one in order to function.
You could always add a keypad, I guess, so people can punch it in manually.
Yes, that was the right answer. Because Android is open, you can run whatever barcode scanning software you want on it, and use the open cam to capture barcodes to be interpreted in any way you like.
That makes the G1 smartphone the cheapest available wireless barcode scanner. The leverage of freedom is that there are a lot of expensive special purpose proprietary tools that are about to go obsolete because they solve a problem that this device can solve with the addition of software. Since this device is open, every concievable bit of software will quickly be ported. Many more that exploit its special blend of features will be invented.
That's what makes Freedom the killer app.
Recognizing that is what makes the Google guys so smart. I gotta get that stock.
Help stamp out iliturcy.
The main problem when decoding 1D Barcodes is still the image quality and the lack of macro-mode of the majority of the mobile handsets. Snappr.net has a demo for decoding 1D Codes on the iPhone with the help of a macro-lense and it works pretty well (video at YouTube ). Together with the ability to enter barcodes manually I strongly believe that this technology is hitting the spot.
There's an iPhone app called Save Benjis that scrapes a bunch of sites for info (though apparently they got a C&D from Amazon, which sucks) and gives price comparisons. They say that barcode scanning was at the top of their feature wish list, but after working for a month on it, they gave up because of the focal length problem. They tried sharpening algorithms, etc. They were even trying to use OCR to read the numbers under the barcode, but it was apparently really slow. I told them about the Clarifi case, and I just ordered one for myself. (A long time ago I ordered the CC-49 lens for the Nokia 3650 after finding a tech demo app that could read barcodes, but it never went anywhere. I've been wanting to do this for a LONG time.)