Dahl in his blog post did not cite a specific source of the criticism. But UK-based web developer and consultant David Bushell had heralded Deno’s decline in an April 28 blog post. “The future of Deno Land Inc. is not looking bright,” Bushell wrote. “Their commercial product Deno Deploy claims to be ‘edge’ hosting with ‘massive global scale’…. Except that’s a bit of a stretch if we’re being honest.” Bushell noted that Deno Deploy had dropped from 35 regions to just 12 by January 2024 and went on to complain that development of Deno Fresh had slowed, that development of Deno KV had stalled, that Deno packaging is a mess, and that Deno releases are “nothing but Node compatibility fixes” and “an endless chase.” “Yeah, Deno is done,” Bushell warned.
But Dahl defended Deno and its companion technologies. “Deno 2’s robust Node compatibility effectively removed a major adoption barrier, unblocking a wide range of serious use cases, for one,” Dahl wrote. Regarding the scaling back of Deno Deploy regions, Dahl said most applications do not need to run everywhere. “They need to be fast, close to their data, easy to debug, and compliant with local regulations. We are optimizing for that,” Dahl said. Deno Deploy started in 2021 in 25 regions, then scaled to 35 regions, and now runs in six, he noted. The reduction in regions was driven by both cost and usage.
As for Deno KV, described by Dahl as “a zero-setup, globally consistent key-value store with a simple API and real-time capabilities,” Dahl acknowledged that it does not solve everything. “It’s not a general-purpose database, and it doesn’t replace relational systems for most applications,” Dahl said. “Developers love it for what it is: a zero-config global store that just works.” Dahl noted that efforts are under way to address broader needs for state management, and that Deno KV will remain in beta while critical bugs and security issues are addressed.