Re: moving forward—with a plan

Hi Markus,

I also completed the survey.
But I have to disagree strongly with the approach you propose.
Why should we now, after years of discussion and slow progress, just 
stumble into implementation
when we are almost done with some parts? This applies to the collection 
design and some other issues.
This would blindly waste resources in my opinion. So what I think is 
that we should first try to resolve
some basic issues and then of course, we can do this -- I'm all for 
having something working.

I'm telling this as we have been waiting for some stuff and have already 
implemented quite a lot.
This also applies to other members of the list as far as I understand 
from the recent mails.

Now let me share our contribution to the overall effort.
We have published a generic Hydra Core Client that we use in production 
already:

https://github.com/pajax/hyjax

It's based on a generic AJAX client in which we have invested quite a 
lot of resources over years and recently have published:

https://github.com/pajax/pajax

It's not complete yet, for example we have not implemented the new 
collection design as we were waiting for its finalization.
It still has other issues https://github.com/pajax/hyjax/issues/3 but I 
would be glad of course, if we would build on it.
The nice thing is that it is based on/ resembles the to-be Web standard 
"fetch".

We already did some stuff towards a new API console/ generic extensible 
Hydra GUI Client here

https://github.com/rhea/

But this is too imperfect yet.

BG, Thomas


On 25.05.2016 22:48, Markus Lanthaler wrote:
> Thanks everyone for sharing your thoughts. It's probably good that I waited a bit with writing mine :-)
>
> I don't think we have a tooling or process issue. In my opinion the biggest issues is we have are missing commitment and contribution... and, probably most importantly, lack of code that is driving and informing the discussions.
>
> Hydra's initial development was completely driven by the Hydra Console and the corresponding server-side library. Along the way we somehow lost that. Our discussions became more abstract and less and less driven by implementations. This made it hard to assess whether a design proposal was good or bad. As a side-effect it also became way too easy to sidetrack discussions with arguments. There have been new implementations but we somehow never managed to start a meaningful discussion or to use them to spearhead the development of certain features in Hydra (I'm likely over-generalizing and drawing a too negative picture here to bring my point across).
>
> So, my proposal is comparatively simple: we start creating a Hydra client (or a couple of them) and use it to drive the further development of Hydra. By refactoring the Hydra Console we already have a first use for it. Perhaps it could also be used for Linked Data Fragment clients and other projects. The server side is obviously important as well, but it is comparatively simple to mock or fake so I would mainly focus on the client-side. If we have enough resources, or in other words, volunteers, we can obviously work on server-side implementations as well.
>
> As soon as we are happy with the implementation of a feature, we will update the specification (at which point, we will already have examples).
>
> I'm fine with having regular calls but we have to keep in mind that it might be hard even for people that are willing to invest the time to attend (timezones, work, etc.). We also need to ensure that we properly archive all the work and the discussions if at some point in the future we want to make this an official W3C standard. There have been concerns at W3C in the past that GitHub may not be the best place for that. I could reach out to the folks there asking what the current state is and whether it would, e.g., be fine if we would automatically export all discussions etc.
>
> So, to start gauging the interest, I created a survey:
>
>    https://www.w3.org/2002/09/wbs/64267/hydra-client-collaboration/
>
> Please let me know if you would be willing to participate in such an effort or not and what programming languages you would like to use. I also included two questions regarding telephone conferences.
> (Almost) everyone on this list should have a W3C account, please use it. All non-responders in the group will get an email on June 1st reminding them to answer the survey.
>
>
> Thanks,
> Markus
>
>
> --
> Markus Lanthaler
> @markuslanthaler
>
>
>
>