Hello, I'm Aravinth.
I find computers fascinating and making them do a specific thing in a specific manner even more fascinating. So I spend most of my time building and breaking software.
Projects that I'm currently working on
One of my biggest gripes with existing CAPTCHA systems is that they incorrectly judge a visitor using TOR or has their cookies turned on to be a bot. I wanted to create something that could effectively ratelimt bots while still allowing users with tracking countermeasures enabled. And I think mCaptcha is effective in that regard.
Also the World Wide Web Consortium reports that existing CAPTCHA systems pose accessibility problems to users with special needs. The proof of work algorithm that mCaptcha uses requires minimal user interaction as it only requires the click of a button!
The original ForgeFed required involvement from forge developers themselves to enable federation. While it's a reasonable approach, such an endeavor from the forge developers' side will require massive commitment. Our approach is different as we are trying to implement federation entirely using the APIs exposed by the software forges. If successful, the lessons learnt from ForgeFlux can be used to externally enable federation in a variety of services.
Loïc Dachary and I wanted to create a 100% Free business, based on the idea discussed here, we have now dropped the business part and are now working towards a reliable, self-hostable software development suite, with a payments processor, that may optionally be used to manage users.
kavasam: Privacy-focused phone and email spam protection. In India, Truecaller is the most popular method to block spam phone calls. While effective, it uploads its users' entire phone book to its servers. This has the undesirable effect of creepy phone calls from stalkers(I uninstalled Truecaller in 2017, my experience might be outdated).
Kavasam's approach is to use cryptographic hash functions and digital signatures to both anonymously blacklist an identifier(phone number, email ID, etc) and anonymously file the blacklist report. And it employs a peer-validation mechanism for effective blacklisting: most spam calls are the effect of database leaks or irresponsible service providers selling their customers' contacts. Usually, spammers work through the contents of the database in their spam campaign. So if Alice, a victim of a campaign, blacklists the spam calls she receives and Bob, a victim of the same campaign, has noticed Alice's blacklist reports against the spam calls he received, can instruct Kavasam to implicitly block phone calls based on Alice's reports, both future and current.
Available for hire
I'm passionate about decentralisation and digital privacy therefore, I hope to make a career out of free software. If you are a free software company and like what I'm doing, please consider hiring me. My résumé is attached below and my contacts are available here.
- Résumé: click here to download