Re: adding hypermedia to spatial data best practices

hello joshua.

On 2015-08-04 5:16 , Joshua Lieberman wrote:
> I hope we don’t lose sight in the discussion of REST and “webbiness”
> that most of the present Web only works because there are humans reading
> web pages and interpreting link semantics from the textual / visual
> context of both the link and the target. A link that a machine follows
> from one collection of data elements to another collection of data
> elements in performance of a task has a significantly different set of
> requirements.

yes, but the question comes in two parts:

1) do you want to follow an architecture where m2m clients should always 
follow the same principles to achieve application goals, without 
necessarily assuming that they can do that automatically and generically 
across specific domains. this is what REST is doing.

2) do you want to follow an architecture where semantics are formalized 
in a way that m2m clients not only can follow the same principles, but 
actually can interact with domains that they have not been built for, by 
learning about the meaning of those domains. this is what semweb and 
linked data are doing.

personally, my expertise and preference is usually getting the first one 
right and maybe then try to tackle the (much more challenging) second one.

> It would be lovely to have a universal hypermedia data
> format, for example, but so far the only technology that has come close
> has been RDF, and that has definitely been an uneven and incomplete
> success.

RDF is not hypermedia, it is just a graph data format that uses URIs as 
identifier for graph nodes. a couple of hypermedia formats (some of them 
RDF-based, others not) have been proposed, but none of them can claim 
victory so far. here's a simple list (that i should start fleshing out):

https://github.com/dret/hyperpedia/blob/master/formats.md

> One of the nice aspects of a RESTful API approach is indeed the option
> for a person to be able to browse holdings / capabilities to get a sense
> of the content or how best to write clients. It doesn’t mean that same
> hypermedia approach is really effective for machines to use.
> I agree that some mixed human / machine use cases might capture actual
> usage patterns more effectively.

from the REST perspective, the goal has never been to build a completely 
automated m2m web that machines can traverse and learn about the world 
while they are doing it. the goal is to follow a simple, robust, 
decentralized and extensible set of principles so that all kinds of 
services can be exposed, and that building clients for them (and, most 
importantly, *across* them) is relatively easy.

cheers,

dret.

-- 
erik wilde | mailto:dret@berkeley.edu  -  tel:+1-510-2061079 |
            | UC Berkeley  -  School of Information (ISchool) |
            | http://dret.net/netdret http://twitter.com/dret |

Received on Tuesday, 4 August 2015 17:45:36 UTC