This is to create an easier run time environment for our tutorials like how pytorch docs tutorial has:
https://pytorch.org/tutorials/beginner/blitz/tensor_tutorial.html

We can reuse Katacoda's scenarios @dashohoxha is building for this.
Yes, I am working on an interactive version of Get-Started: https://github.com/iterative/dvc.org/issues/664
I don't know whether it will be linked with a button or with a simple link, but this does not seem crucial to me.
I am working on an interactive version of Get-Started
Here it is: https://katacoda.com/dvc/courses/get-started
Actually I did it as a single tutorial first: https://katacoda.com/dvc/courses/tutorials/get-started
But it seemed to be too long, taking more than 60min to finish, so I split it in smaller lessons.
I am not sure yet how to best link it from the website.
@kurianbenoy Please feel free to give a try to a Colab version of the get-started, some people may prefer it better (refering to this: https://discordapp.com/channels/485586884165107732/565759186693128202/628298295063740436)
@dashohoxha this is great! I'll take a look and review the scenarios, also we should start integrating it with the engine (add a button). Any suggestions on how can it look like? I would prefer it to stand out as much as possible, a full text message something like "Run an interactive version". Would be great to emphasize it somehow when you hover code blocks?
@algomaster99 @jorgeorpinel any suggestions on this ^^
Super cool to have all those Katacoda scenarios! I'd love to help reviewing them too. Are they created in a way that we can easily give feedback on @dashohoxha? For example in a Git repo perhaps? (Not familiar with Katacoda.) Otherwise how can we leave comments on specific commands in certain scenarios? Thanks
Google Colabs, Katacoda, Kaggle kernels
So we're discarding Kaggle kernels for now right?
feel free to give a try to a Colab version of the get-started
Er... Not sure we want so many versions of the same material, already it may be a challenge to maintaining both our dvc.org docs and Katacoda updated and synchronized. Perhaps we should also discard that option for now?
...we should start integrating it with the engine (add a button). Any suggestions on how can it look like?
Agree! That seems like the main purpose of this ticket 馃構. As for how it can look, maybe pretty similar to the PyTorch tutorial link provided, 
Are they created in a way that we can easily give feedback on @dashohoxha? For example in a Git repo perhaps?
You may start any discussions here: https://github.com/iterative/katacoda-scenarios/issues
However I would suggest that we keep all the discussions here: https://github.com/iterative/dvc.org/issues
So we're discarding Kaggle kernels for now right?
I don't even know what they are, but for me it is ok, as long as somebody is willing to work on them.
@shcheklein We have already hyperlinked it. https://dvc.org/doc/tutorials/interactive. Where are we planning to put the interactive buttons?
So we're discarding Kaggle kernels for now right?
I don't even know what they are, but for me it is ok, as long as somebody is willing to work on them.
I feel there will be some maintenance challenges to support jupyter-notebook based interactive tutorials in Kaggle with DVC like @jorgeorpinel said for Documentation team. And to be honest a CLI tool like DVC is not best to be run in a notebook.
I have already added the get started tutorial as a jupyter notebook sometime back.
https://www.kaggle.com/kurianbenoy/introduction-to-data-version-control-dvc
Agree, I think we should stick with Katacoda, and also ignoring Colab for now unless a contributor wants to develop and maintain that.
We have already hyperlinked it. https://dvc.org/doc/tutorials/interactive. Where are we planning to put the interactive buttons?
By "interactive button" we mean a pretty link right? Yeah it's just a matter of design. Refer to https://github.com/iterative/dvc.org/issues/670#issuecomment-540079227 and https://github.com/iterative/dvc.org/issues/670#issuecomment-540146127. I think whoever takes on this task should just come up with a suggestion as of where in the docs layout to place this.
Perhaps a first step is to implement automatic recognition of md links to katacoda.com/dvc/..., turning them into buttons, and let the doc writer decide where to put them in each doc. Or <button> tags with value="katacoda.com/dvc/...". Idk, just brainstorming here.
Most helpful comment
I feel there will be some maintenance challenges to support jupyter-notebook based interactive tutorials in Kaggle with DVC like @jorgeorpinel said for Documentation team. And to be honest a CLI tool like DVC is not best to be run in a notebook.
I have already added the get started tutorial as a jupyter notebook sometime back.
https://www.kaggle.com/kurianbenoy/introduction-to-data-version-control-dvc