- From: Ruben Verborgh (UGent-imec) <Ruben.Verborgh@UGent.be>
- Date: Mon, 17 Dec 2018 08:23:54 +0000
- To: David Booth <david@dbooth.org>
- CC: semantic-web <semantic-web@w3.org>
Hi David, >> – build powerful query engines such that app developers do not need to write HTTP requests > (every data access should be a query, read or write) > > Can you please elaborate a bit on this one, perhaps with an example? I'm wondering what kinds of query engines you envision, and how to use them. The usage of such query engines is best illustrated here: https://ruben.verborgh.org/blog/2017/12/20/paradigm-shifts-for-the-decentralized-web/#interfaces-become-queries-p-5 In essense, by not binding applications to concrete (but inevitable changing) HTTP requests, we bind them to queries, which declarative indicate the data operation. These queries are then translated at runtime into concrete requests. An example of such an engine is Comunica: http://query.linkeddatafragments.org/ It decomposes a query into HTTP requests, depending on the capabilities of the servers it talks to. Best, Ruben
Received on Monday, 17 December 2018 08:24:20 UTC