Sort by: Newest, Oldest, Most Relevant
(#3ezvila) In addition, supporting "Range" requests _should_ be possible from a client's perspective with fallback options to do a full request. I don't think this needs to be part of the spec at all. The way I would implement this is to keep track of the last position, and re-fetch from that position minus a few hundred bytes just to be sure, throw away any initial garbage. I'd probably also try to detect if the feed is append-only or prepend and track this too (somewhere). If the Metadata preamble changes, seeking back a bit _should_ in theory work or fallback.

matched #22aayga score:0.08 Search by:
Search by 1 tags:
(#aajeezq) @mckinley a few points: Technical: 1. Single point of failure: Relays, though decentralised, could be targeted for censorship or sabotage. 2. Message integrity: Messages are not inherently linked, raising concerns about spoofing and manipulation. 3. Data storage: Clients may need to download large amounts of data, especially historical messages, impacting performance. 4. Limited functionality: Currently focuses on text-based communication, lacking media sharing or advanced features. Usability: 1. Steep learning curve: It is still young and requires technical knowledge for setup and use. 2. Limited user base: Finding an active community and familiar faces can be challenging. 3. Unintuitive interfaces: Client applications may not be as user-friendly as established platforms. Also, full of crypto bros, crypto bros wanna be, and, well, worthless crypto (mixed with some porn, nazi crap, etc.). But go ahead, go through the same phases I went (I even ran my own relay), and see it for yourself. :-)

matched #22ay73q score:0.06 Search by:
Search by 1 mentions:
Search by 1 tags:
This is twtxt search engine and crawler. Please contact Support if you have any questions, concerns or feedback!