Wednesday, 28 December 2016

Shape Expressions Community Group

The Shape Expressions W3c community group has been created two weeks ago.

The FAQ for W3c community groups says that it is an open forum, without fees, where Web developers and other stakeholders develop specifications, hold discussions, develop test suites, and connect with W3C’s international community of Web experts. They may produce Specifications; these are not standards-track documents but may become input to the standards process. For instance, a Community Group might gather to work on a new technical specification, or convene to have discussions about a tutorial for an existing specification.. In principle, they provide a more flexible structure than Working Groups and I hope the members attitude to be more open and constructive than in the Data Shapes Working group.

I had the pleasure to collaborate with Eric Prud'hommeaux, Harold Solbrig and Iovka Boneva in the last two years developing the ShEx language. We have had lots of improvised meetings through Skype and we had a face-to-face meeting in August 2015. For me it is the ideal of how a bunch of people can work together. With a very friendly attitude where anyone can participate and share ideas, and with one common goal, designing the best language for RDF Validation that we can.

On the contrary, I have to say that the attitude of some members of the Data Shapes Working Group were just the opposite, some people were not constructive at all, criticising any idea, closed to any new suggestion or trying to impose their already existing implementation. Most of the suggestions that were proposed by people related to ShEx were rejected. The working group was even unable to have a real face-to-face meeting, which I think would help to avoid some of the personal problems that appeared. In general, the WG became a place were new ideas were not wellcome and were people were afraid to participate (at least, that's how I felt).

So my hope is that the new Shape Expressions community group maintains the positive attitude and avoids the problems that the WG had. If it was possible I would recommend any member to read the No Asshole rule book and to keep the goal of designing the best RDF validation language that we can.

During the last weeks Gregg Kellogg has started a new ShEx implementation in Ruby. He has also contributed to a Json+LD ShEx representation which will also give ShEx an RDF serialization and he has also been detecting some errors in the test-suite and preparing ShEx implementation reports (by the way, I need to prepare the EARL report of my implementation).