A dapplet (re-)storing tweets on the fly from/into IPFS and seamlessly decentralizes Twitter operations on UX/UI level.
The Tweery project creates a dapplet (an augmenting application) running on top of the twitter.com UI. This dapplet stores all tweets seen by the community on the fly into an IPFS storage. It also restores deleted tweets on the fly from the IPFS storage and enables retweet functionality if disabled by Twitter.
The project demonstrates a generic technology, how to enforce a decentralization for (almost) any existing web-based community, including its content, accounts and operations. Furthermore, it shows how communities can become self-sovereign and resistant to censorship and forcible profit extraction by the centrally owned hoster, even if this hoster is the creator of the community.
The project transforms the current paradigm of the owner-based web (where the website owner has total exclusive control over its community) into the new one. The new web paradigm balances the power of the website owner by the ability for a community to "fork" its data and services and "redirect" processing to another backend.
DISCLAIMER: some ideas behind the Dapplets technology are new and require some time to grasp.
The project utilizes the Dapplets technology to create an augmenting application and IPFS for data storage. There is an existing Twitter Adapter, which parses the twitter website and exposes the website's semantic structure containing tweets and insertion points. The tweery dapplet utilizes the Twitter Adapter to crawl tweets and perform operations on them. The dapplet platform connects all parts together.
We have created the tweery dapplet, running on top of the tweeter website, and we have extended an existing Twitter adapter with necessary widgets and insertion points. A mapping between tweet id and the IPFS cid was temporarily implemented as a centralized service. It should be reimplemented later as a decentralized writable hashmap (or database).