reCAPTCHA eases up on the human eye

Oct 28, 2013 by Nancy Owano report
reCAPTCHA eases up on the human eye

(Phys.org) —Google, assuming you are human and reading this, wants you to know that CAPTCHAs are more readable. There will be easier days ahead than having to put your face against the screen, struggling to figure out if the string wants you to key in a v and u or a single w, or if an indescribable shape is really an exotic r. For humans, deciphering a CAPTCHA string may get easier but a tougher time is ahead for bots because their CAPTCHAs will be designed to stop the bots from getting through. The new, easier reCAPTCHAs are numbers. On Friday, a blog post from Google's Vinay Shet, Product Manager, reCAPTCHA, said that "Humans find numeric CAPTCHAs significantly easier to solve than those containing arbitrary text and achieve nearly perfect pass rates on them. So with our new system, you'll encounter CAPTCHAs that are a breeze to solve."

Google's new update involves different classes of CAPTCHAs for different kinds of users. "This multi-faceted approach allows us to determine whether a potential user is actually a human or not," he wrote. Shet said more about the good news. "The distorted letters serve less as a test of humanity and more as a medium of engagement to elicit a broad range of cues that characterize humans and bots." The team members behind reCAPTCHA nonetheless maintained their focus on how to distinguish legitimate users from automated software. "The updated system uses advanced risk analysis techniques, said Shet, "actively considering the user's entire engagement with the CAPTCHA—before, during and after they interact with it."

Their work is not over. "While we've already made significant advancements to reCAPTCHA technology, we'll have even more to report on in the next few months, so stay tuned," he stated. reCAPTCHA has its roots at Carnegie Mellon, not Google, but Google acquired reCAPTCHA in 2009; roots actually go back further to Carnegie Mellon's CAPTCHA which stands for Completely Automated Public Turing test to tell Computers and Humans Apart. CAPTCHA was developed as a security tool, showing distorted textual images, to be used to ward off bot attacks and spammers. One encounters CAPTCHAs on sign-up pages. The reCAPTCHA Project was built on the original concept and acquired by Google in 2009.

Google defines present-day reCAPTCHA as a Web service and notes that those who run web sites will find adoption is easy, as simple as adding a few lines of code. "For some applications (such as Wordpress and Mediawiki), we have plugins that allow you to use reCAPTCHA without writing any code," according to the reCAPTCHA site. "We generate and check the distorted images, so you don't need to run costly image generation programs." Google also positions reCAPTCHA as a free CAPTCHA service "that helps to digitize books, newspapers and old time radio shows." To help along digitizing books, reCAPTCHA sends words that cannot be read by computers to the Web in the form of CAPTCHAs for humans to decipher. Currently, according to the reCAPTCHA web site, "we are helping to digitize old editions of The New York Times and books from Google Books."

Explore further: New privacy battle looms after moves by Apple, Google

More information: www.google.com/recaptcha/whyrecaptcha
googleonlinesecurity.blogspot.… ier-but-only-if.html

Related Stories

Stanford researchers outsmart captcha codes

Nov 03, 2011

(PhysOrg.com) -- Stanford researchers say that captcha security codes, asking Internet sign-up users to repeat a string of letters to prove the users are human, can be thwarted, and they have successfully ...

Google acquires Web security firm reCAPTCHA

Sep 16, 2009

(PhysOrg.com) -- Google announced on Wednesday that it has acquired reCAPTCHA, a company that produces the squiggly words used by websites to guard against spam and fraud.

Stanford computer scientists find Internet security flaw

May 24, 2011

(PhysOrg.com) -- Researchers at the Stanford Security Laboratory create a computer program to defeat audio captchas on website account registration forms, revealing a design flaw that leaves them vulnerable ...

CAPTCHA evokes sympathetic (aka correct) response

Oct 09, 2012

(Phys.org)—CAPTCHAs by definition (stands for Completely Automated Public Turing Test to tell Computers and Humans Apart) are gotcha tools that are used to spot automated-attack attempts posing as people. ...

Ticketmaster makes online 'CAPTCHA' puzzles easie

Jan 30, 2013

Ticketmaster, often the target of fan anger when tickets to popular concerts get sold out in seconds, is removing an annoyance that can slow down the buying process: the jumble of letters that people need ...

Recommended for you

Say Ello to the new privacy debate on social media

Sep 29, 2014

Ello is new social networking space on the web that is receiving a lot of attention of late – so much that it's caused a few problems with the website out of action from time to time. ...

User comments : 4

Adjust slider to filter visible comments by rank

Display comments: newest first

Sinister1811
3 / 5 (18) Oct 28, 2013
Why the hell didn't they just do that in the first place?
Gigel
5 / 5 (1) Oct 28, 2013
Maybe one day absurd requests on password making will be dropped too, so I can remember the passwords without writing them somewhere. It seems programmers forgot we are humans.
shavera
2 / 5 (2) Oct 28, 2013
Has anyone noticed the "picture" that we're always solving in these captchas are like... address numbers?
Eikka
not rated yet Dec 13, 2013
Why the hell didn't they just do that in the first place?


Google used reCAPTCHA as an OCR tool. The first word was computer generated, and the second word was a piece of text scanned from a book that their automatic readers couldn't make sense of. It was enough to write only the first word right and the second word wasn't tested for.

The problem for google was that people got wise and started inputting profanities or just random strings of characters.

Has anyone noticed the "picture" that we're always solving in these captchas are like... address numbers?


They most likely are. The system hasn't changed one bit. That's how google is making money: they're selling your work without paying you anything.