Is x/exp/shiny dead?

[fazal]

It was one of the things I was looking forward to, but I see almost no activity there, which makes me sad. 

Is it because it is mature enough for general use? There is a road-map for widgets but non are complete. I really hope it is not abandoned because Go really needs a GUI library that is idiomatic. 

Regards,

[Ian Lance Taylor]

[ +nigeltao ]

> --
> You received this message because you are subscribed to the Google Groups
> "golang-nuts" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to golang-nuts...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[leonsal]

You could take a look at G3N which is a 3D OpenGL game engine but

it features a basic GUI with several widget types.

You could check the available GUI widgets running the G3N Demo and
selecting the GUI tests folder.

[Nigel Tao]

On Tue, Aug 1, 2017 at 6:25 AM, fazal <mohame...@gmail.com> wrote:
> It was one of the things I was looking forward to, but I see almost no
> activity there, which makes me sad.

It's certainly not very active right now. I work part time, for
uninteresting reasons, and shiny is unfortunately not my primary
project.

I'd like to work on it again at some point, but I can't promise when
that will happen.

FWIW, there was some recent discussions of a new, multi-person Go GUI
effort at https://groups.google.com/d/msg/golang-nuts/JyrN8cxWCrU/H6oZpv02BQAJ

[Daniel Skinner]

As one of the persons reached out to for that effort (and I'm not a shiny contributor), my current unpublished work will eventually serve as a basis for some proposals to shiny for review that maybe after some testing and input from others will stand on its on merits, for such a proposal.

Even as a side gig for people, I think shiny is pretty far along; but I think shiny has reached a critical point where every decision may get cemented and people want those decisions to be the best possible ones.

It wouldn't be entirely difficult, just time-consuming, to pump out all those widgets by cementing in a bunch of other decisions; but then we just sort of have this Go version of every other GUI toolkit with the same every-other problems. There's a good opportunity to research new ideas now.

Shiny is already doing and demonstrating a lot of important things; but event and configuration, and a single unified layout and animation step are two things I think we could all be doing better.

[fazal]

Thanks for the great work on exp/shiny. 

Where does shiny need most attention/contributions? 

I think a status report of some sort would help others pickup where it's left off now.

[fazal]

This sounds great. So it is not entirely dead.

[as....@gmail.com]

It's not ready for general use, but it certainly isn't dead. I think a lack of examples for some of the more-complex widgets is probably where it needs the most help, but considering that only one person is working on it the result is a very impressive addition to the Go ecosystem. I've not used the widgets much, and ended up porting/reimplementing Plan 9's libframe due to my inability to understand how to get editable text on the screen. This was over a year ago, so things may have changed. I agree with the general consensus, standard widgets are the biggest obstacle for exp/shiny, but those require effort to be ported, and maybe copying what already exists isn't the best approach either.

[Andrew Williams]

Hi,

I know it's been a year since this thread was mentioned but I thought I would add an update for anyone finding this topic.

I wanted to try using shiny as it's got some great foundations and is well thought out. Unfortunately it's a few steps short of being really useful.

Looking at the demos those that look most complete seem to implement functionality to make them complete that could be in the toolkit for reuse.

For example the Button type in one example is not present for developers using shiny.

When I asked about this it was noted that there's been no activity for a couple of years.

This makes me think that, unless someone steps up to really push this forward, it's pretty much writing on the wall.

Apologies if this was obvious to people or if I've misunderstood the situation.

Andrew