External Tools Dataset Extension

205 views
Skip to first unread message

sean....@berkeley.edu

unread,
Apr 6, 2018, 6:42:49 PM4/6/18
to Dataverse Users Community
Hey everyone!

We are working on a project at UC Berkeley to integrate Dataverse with Binder which is a service that automatically hosts dockerized instances of Jupyter notebooks online. This is part of a larger project to aid with research data reproducibility. It seems like the new External Tools features would be perfect for what we want to do, but currently it only supports tabular data at the file level. In our usecase, we would want to make an external tool at the dataset level so that our tool could download all of the files in a dataset and use those files to automatically create a docker instance to interact with the data. We already have the ability to automatically setup a docker instance, but we need a way to interface with Dataverse to retrieve the files for that instance.

Are there currently any efforts to expand the capability of External Tools outside of tabular data or anyone who would be willing to try to add some basic integration? Our thoughts are essentially having another 'Explore' button show up in the Dataset view just as it currently does in the view for tabular files.

Thanks,
Sean Dooher, Kevin Yang, and Aaron Culich

UC Berkeley

Mercè Crosas

unread,
Apr 6, 2018, 7:03:30 PM4/6/18
to dataverse...@googlegroups.com
The integration of Dataverse with Binder is really exciting! Thanks for sharing it. 

I can see that what you are requesting could work well, but I'll let other in the team follow up with details.

Merce


----------
Mercè Crosas, Ph.D., Chief Data Science and Technology Officer, IQSS, Harvard University

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/bdf4c7d1-7949-481f-81cb-d420a8d386b7%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Jonathan Crabtree

unread,
Apr 6, 2018, 7:08:44 PM4/6/18
to dataverse...@googlegroups.com, Aaron Culich, sean....@berkley.edu
Sean

Odum is working on a similar effort for journal verification

We would love to coordinate

Jus let me know when you would like to talk

Jon

_____________________________
From: Mercè Crosas <merce....@gmail.com>
Sent: Friday, April 6, 2018 7:03 PM
Subject: Re: [Dataverse-Users] External Tools Dataset Extension
To: <dataverse...@googlegroups.com>



The integration of Dataverse with Binder is really exciting! Thanks for sharing it. 

I can see that what you are requesting could work well, but I'll let other in the team follow up with details.

Merce


----------
Mercè Crosas, Ph.D., Chief Data Science and Technology Officer, IQSS, Harvard University

On Fri, Apr 6, 2018 at 6:42 PM, <sean....@berkeley.edu> wrote:
Hey everyone!

We are working on a project at UC Berkeley to integrate Dataverse with Binder which is a service that automatically hosts dockerized instances of Jupyter notebooks online. This is part of a larger project to aid with research data reproducibility. It seems like the newExternal Tools features would be perfect for what we want to do, but currently it only supports tabular data at the file level. In our usecase, we would want to make an external tool at the dataset level so that our tool could download all of the files in a dataset and use those files to automatically create a docker instance to interact with the data. We already have the ability to automatically setup a docker instance, but we need a way to interface with Dataverse to retrieve the files for that instance.


Are there currently any efforts to expand the capability of External Tools outside of tabular data or anyone who would be willing to try to add some basic integration? Our thoughts are essentially having another 'Explore' button show up in the Dataset view just as it currently does in the view for tabular files.

Thanks,
Sean Dooher, Kevin Yang, and Aaron Culich

UC Berkeley

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/bdf4c7d1-7949-481f-81cb-d420a8d386b7%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-commu...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataverse-community/CAPAYmDOzp%3DrV1ZmCm%3DKnh32w1dN1N-KBSFAAiKxZWm%2BMfKUbJw%40mail.gmail.com.

Durand, Gustavo

unread,
Apr 6, 2018, 7:11:31 PM4/6/18
to dataverse...@googlegroups.com
When we first added external tools like this, we imagined (knew?) at some point there would be some cool explore type functionality at the dataset level. Glad to hear about your project!

I already have some questions / comments, but let try to compile a list and I'll send something out next week.

Thanks,
Gustavo
(Dataverse Technical Lead / Architect)


On Fri, Apr 6, 2018, 19:03 Mercè Crosas <merce....@gmail.com> wrote:
The integration of Dataverse with Binder is really exciting! Thanks for sharing it. 

I can see that what you are requesting could work well, but I'll let other in the team follow up with details.

Merce


----------
Mercè Crosas, Ph.D., Chief Data Science and Technology Officer, IQSS, Harvard University

On Fri, Apr 6, 2018 at 6:42 PM, <sean....@berkeley.edu> wrote:

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-commu...@googlegroups.com.

Philip Durbin

unread,
Apr 6, 2018, 8:08:20 PM4/6/18
to dataverse...@googlegroups.com
Hi all,

This is all super exciting and I'm sorry that I missed you at http://irclog.iq.harvard.edu/dataverse/2018-04-06#i_65518 a bit ago. Stupid timezones. ;)

Sean, the external tool feature is new and shiny (I don't think we've even tweeted about the Dataverse 4.8.6 release yet) and you aren't the only one who has expressed interested in having external tools at the dataset level rather than the file (tabular file only right now) level but since you have your own installation of Dataverse (if memory serves), I'm wondering if these two settings are on your radar:

- :ComputeBaseUrl - http://guides.dataverse.org/en/4.8.6/installation/config.html#computebaseurl
- :CloudEnvironmentName - http://guides.dataverse.org/en/4.8.6/installation/config.html#cloudenvironmentname

These settings should give you a "Compute" button at the dataset level. Basically, it's just a link that dumps to another website (an OpenStack environment, for example) with the bucket for your dataset in the URL. I believe you need to be configured for Swift storage for this to work, judging from my faulty memory and what I'm reading at http://guides.dataverse.org/en/4.8.6/installation/config.html#setting-up-compute

Anyway, it might be something to try. I'd hate for you to be blocked because you need the external tool feature to do more.

If you'd rather stick with external tool, and you might because it's much more lightweight and doesn't require Swift storage or anything, as a kludge you could make every dataset include a file that's tabular just so you'll get the Explore button to show up so you can get the user over to BinderHub. Then from BinderHub you can use the file ID in Dataverse to derive the dataset ID an then proceed to download all the files. I can give you pointers to all the API endpoints you'd need on the Dataverse side to accomplish this. Just let us know which direction you'd like to go.

Longer term, we should extend the external tool concept to operate at the dataset level. Please go ahead and open an issue at https://github.com/IQSS/dataverse/issues about this if you are so inclined.

Thanks!

Phil

p.s. In case anyone is wondering, this didn't come out of nowhere. You can check out the "2018-03-09 Meeting with UC Berkeley and Dataverse" notes I took at https://docs.google.com/document/d/12njDDHHfoNdos0By3lKlxYbRWLsAp8Eo-AGpjuBF_rY/edit?usp=sharing and discussion at chatter at https://github.com/jupyterhub/binder/issues/80 and https://github.com/nds-org/ndslabs-dataverse/issues/8#issuecomment-372359052

On Fri, Apr 6, 2018 at 7:08 PM, Jonathan Crabtree <jonc...@gmail.com> wrote:
Sean

Odum is working on a similar effort for journal verification

We would love to coordinate

Jus let me know when you would like to talk

Jon

_____________________________
From: Mercè Crosas <merce....@gmail.com>
Sent: Friday, April 6, 2018 7:03 PM
Subject: Re: [Dataverse-Users] External Tools Dataset Extension



The integration of Dataverse with Binder is really exciting! Thanks for sharing it. 

I can see that what you are requesting could work well, but I'll let other in the team follow up with details.

Merce


----------
Mercè Crosas, Ph.D., Chief Data Science and Technology Officer, IQSS, Harvard University

On Fri, Apr 6, 2018 at 6:42 PM, <sean....@berkeley.edu> wrote:
Hey everyone!

We are working on a project at UC Berkeley to integrate Dataverse with Binder which is a service that automatically hosts dockerized instances of Jupyter notebooks online. This is part of a larger project to aid with research data reproducibility. It seems like the newExternal Tools features would be perfect for what we want to do, but currently it only supports tabular data at the file level. In our usecase, we would want to make an external tool at the dataset level so that our tool could download all of the files in a dataset and use those files to automatically create a docker instance to interact with the data. We already have the ability to automatically setup a docker instance, but we need a way to interface with Dataverse to retrieve the files for that instance.

Are there currently any efforts to expand the capability of External Tools outside of tabular data or anyone who would be willing to try to add some basic integration? Our thoughts are essentially having another 'Explore' button show up in the Dataset view just as it currently does in the view for tabular files.

Thanks,
Sean Dooher, Kevin Yang, and Aaron Culich

UC Berkeley

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-community+unsubscri...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.
Message has been deleted

Pete Meyer

unread,
Apr 7, 2018, 12:32:02 PM4/7/18
to Dataverse Users Community
Hi Sean, Kevin and Aaron,

Sorry I missed you folks on IRC Friday.  I'm glad to see other people interested in dataset-level tools; this is something that we've been thinking about for a while - more focused on integrating results for reprocessing / reusability / validation pipelines than "exploration". Architecturally, it seems like similar Dataverse interfaces should work for both cases, thought.  

Something that we've also been working on has been replicating data files to different storage sites (with dataset-level granularity) prior to computing on them - the dataset sizes (for our particular area, at least) are large enough that downloading them all each time a compute process needs to run wouldn't be a great idea.  So I don't have any suggestions for how to improve in download-all-files-at-once area; but it may be worth mentioning that the existing Dataverse APIs already support downloading all the files in the dataset.  I believe this would require multiple API calls (list files in dataset, get URLs for each file, transfer each file)

One approach that *might* be worth considering would be to pre-stage datasets (from the relevant dataverse installation) to compute-local storage, which could be provided to the docker containers as a read-only volume (which could be shared among containers); with one docker volume for each dataset (or more specifically, dataset version - but it may make sense to start with only the most recent version).  This has enough conceptual overlap with the work we've been doing with multiple storage locations (https://github.com/IQSS/dataverse/issues/4396) and data locality (DLM) - but that work is in the early development stages, so it's not available for any Dataverse installation (that I'm aware of, at least :) ).

Sorry about the wall of text - but I'd be happy to discuss further (here, IRC, dataverse community call, or another venue).

Best,
Pete

Steven McEachern

unread,
Apr 7, 2018, 10:13:13 PM4/7/18
to Dataverse Users Community
Hi All,

Sean and team, this sounds really interesting. We are involved in a project here in Australia that will be looking at Jupyter notebooks so would love to know more.

On a related note, there is a related case for having a DataSET Explore tool, for tabular data.

Many datasets are set up with multiple files with an identifier linking the files (e.g. households to individuals, geographies to persons, schools to students, persons over time, etc.).
At the moment, the link at the data FILE level means that the user needs to move outside of Dataverse and link the files themselves in order to conduct any analysis.
Being able to provide a link between files and then a Dataset Explore would allow users to analyse across all files within a dataset.

Regards,
Steve
Australian Data Archive

Philip Durbin

unread,
Apr 9, 2018, 9:50:17 PM4/9/18
to dataverse...@googlegroups.com
Hi Sean,

Bleh, it's hackier than I hoped because I have to pull the DOI out of the citation from the search result for a file but here's a little script that will download (well, print the URLs to download) all the files in a dataset based on one of the file's database id:

#!/bin/sh
# https://dataverse.harvard.edu/file.xhtml?fileId=3040230
FILEID=3040230
FILE_SEARCH=`curl -s https://dataverse.harvard.edu/api/search?q=entityId:$FILEID | jq .`
DOI=`echo $FILE_SEARCH | jq '.data.items[0].dataset_citation' | grep -o 'doi:.*' | cut -d, -f1`
ALL_FILE_IDS=`curl -s "https://dataverse.harvard.edu/api/datasets/export?exporter=dataverse_json&persistentId=$DOI" | jq '.datasetVersion.files[].dataFile.id'`
for i in $ALL_FILE_IDS; do
  echo "downloading from https://dataverse.harvard.edu/api/access/datafile/$i"
done

Please let me know if you have any questions about this.

Thanks,

Phil

p.s. I don't know if you really sent your email two days ago or not but I only just got it. :(

On Sat, Apr 7, 2018 at 1:29 AM, <sean....@berkeley.edu> wrote:
I'm glad to see that this feature has some interest! That Compute feature does seem interesting and while I definitely am going to take a closer look at it, I think in the long term we'll want to use external tool because, like you said, it's much more lightweight. As a kludge, we can probably hard code in a button on that page until the feature is developed for testing purposes at least. We still have a decent amount of work to do in actually implementing the tool itself so it shouldn't block us up for a while yet. I'd love the pointers to those API endpoints you mention as that's exactly what we had in mind for the tool.

Thanks,
Sean Dooher


On Friday, April 6, 2018 at 5:08:20 PM UTC-7, Philip Durbin wrote:
Hi all,

This is all super exciting and I'm sorry that I missed you at http://irclog.iq.harvard.edu/dataverse/2018-04-06#i_65518 a bit ago. Stupid timezones. ;)

Sean, the external tool feature is new and shiny (I don't think we've even tweeted about the Dataverse 4.8.6 release yet) and you aren't the only one who has expressed interested in having external tools at the dataset level rather than the file (tabular file only right now) level but since you have your own installation of Dataverse (if memory serves), I'm wondering if these two settings are on your radar:

- :ComputeBaseUrl - http://guides.dataverse.org/en/4.8.6/installation/config.html#computebaseurl
- :CloudEnvironmentName - http://guides.dataverse.org/en/4.8.6/installation/config.html#cloudenvironmentname

These settings should give you a "Compute" button at the dataset level. Basically, it's just a link that dumps to another website (an OpenStack environment, for example) with the bucket for your dataset in the URL. I believe you need to be configured for Swift storage for this to work, judging from my faulty memory and what I'm reading at http://guides.dataverse.org/en/4.8.6/installation/config.html#setting-up-compute

Anyway, it might be something to try. I'd hate for you to be blocked because you need the external tool feature to do more.

If you'd rather stick with external tool, and you might because it's much more lightweight and doesn't require Swift storage or anything, as a kludge you could make every dataset include a file that's tabular just so you'll get the Explore button to show up so you can get the user over to BinderHub. Then from BinderHub you can use the file ID in Dataverse to derive the dataset ID an then proceed to download all the files. I can give you pointers to all the API endpoints you'd need on the Dataverse side to accomplish this. Just let us know which direction you'd like to go.

Longer term, we should extend the external tool concept to operate at the dataset level. Please go ahead and open an issue at https://github.com/IQSS/dataverse/issues about this if you are so inclined.

Thanks!

Phil

p.s. In case anyone is wondering, this didn't come out of nowhere. You can check out the "2018-03-09 Meeting with UC Berkeley and Dataverse" notes I took at https://docs.google.com/document/d/12njDDHHfoNdos0By3lKlxYbRWLsAp8Eo-AGpjuBF_rY/edit?usp=sharing and discussion at chatter at https://github.com/jupyterhub/binder/issues/80 and https://github.com/nds-org/ndslabs-dataverse/issues/8#issuecomment-372359052
On Fri, Apr 6, 2018 at 7:08 PM, Jonathan Crabtree <jonc...@gmail.com> wrote:
Sean

Odum is working on a similar effort for journal verification

We would love to coordinate

Jus let me know when you would like to talk

Jon

_____________________________
From: Mercè Crosas <merce....@gmail.com>
Sent: Friday, April 6, 2018 7:03 PM
Subject: Re: [Dataverse-Users] External Tools Dataset Extension



The integration of Dataverse with Binder is really exciting! Thanks for sharing it. 

I can see that what you are requesting could work well, but I'll let other in the team follow up with details.

Merce


----------
Mercè Crosas, Ph.D., Chief Data Science and Technology Officer, IQSS, Harvard University

On Fri, Apr 6, 2018 at 6:42 PM, <sean....@berkeley.edu> wrote:
Hey everyone!

We are working on a project at UC Berkeley to integrate Dataverse with Binder which is a service that automatically hosts dockerized instances of Jupyter notebooks online. This is part of a larger project to aid with research data reproducibility. It seems like the newExternal Tools features would be perfect for what we want to do, but currently it only supports tabular data at the file level. In our usecase, we would want to make an external tool at the dataset level so that our tool could download all of the files in a dataset and use those files to automatically create a docker instance to interact with the data. We already have the ability to automatically setup a docker instance, but we need a way to interface with Dataverse to retrieve the files for that instance.

Are there currently any efforts to expand the capability of External Tools outside of tabular data or anyone who would be willing to try to add some basic integration? Our thoughts are essentially having another 'Explore' button show up in the Dataset view just as it currently does in the view for tabular files.

Thanks,
Sean Dooher, Kevin Yang, and Aaron Culich

UC Berkeley

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-community+unsubscri...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-community+unsubscri...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsubscribe...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Vyacheslav Tikhonov

unread,
Apr 10, 2018, 4:24:04 AM4/10/18
to Dataverse Users Community
Hi Sean and others,

We (DANS) are also very interested in the possibility to run Jupyter notebooks online bundled to Dataverse datasets. We've actually started this integration on Docker level to get tools running in the separate containers as internal microservices, but support of external tools like Binder sounds very promising. Please keep us in the loop!

Best,
Slava

Durand, Gustavo

unread,
Apr 13, 2018, 2:52:34 PM4/13/18
to dataverse...@googlegroups.com
I'm glad to see that there's been a lot of interest here and a lot of conversation.

Some thoughts: 
- the way we decided to develop the External Tools functionality was incrementally in order to add value as rapidly as we could. So we started with supported Tabular Files and one "function", that we called "Configure" - this was for prototyping a tool from a collaboration we have. The next increment was supporting multiple functions, where we added "Explore".

Other ways we want to consider expanding are: 
• supporting more functions (the infrastructure is there, we would just need to think about how each function would interact with Dataverse),
• add support for specific types of files; for example to support either configure or explore type tools for map files, images files, or any other way we can determine we may want to subclass files.
• consider external tools for datasets - i.e. your use case

So the architecture of this functionality has (hopefully) been designed and developed in a way that we can continue working on these (and other) steps.

Some specific comments / questions:
- Compute is something that we would likely want to make work as an External Tool once we have the framework to support this. It seems like a natural extension of Explore. Clearly, Dataset external tools would be needed for this, as well.
- do you plan your tool downloading the files once and then keeping a docker container for any future users clicking this function? Or would it create a docker container for each specific user? Or some third option?
- I think the pre staging suggestion from Pete is definitely something you may want to consider
- another thing to consider is interacting with permissions; at the file level it's simpler because if you have the right permission we can show the button, otherwise, not. At the dataset level, you're likely to have access to some, but not all, of the files. So you / we will need to consider how to handle this (I believe for compute, we took the easiest technical approach, which was don't show compute unless user has access to all files - but I could be mistaken).

OK, I hope some of the above text is useful to you. Please feel free to contact me* and let me know if I can help in any way with architecture, design, or development issues. The more we talk upfront, the more straightforward a merge should be in the future. :)

* I will be away next week on vacation, but back on the 23rd

Thanks, Gustavo



On Tue, Apr 10, 2018 at 4:24 AM, Vyacheslav Tikhonov <4tik...@gmail.com> wrote:
Hi Sean and others,

We (DANS) are also very interested in the possibility to run Jupyter notebooks online bundled to Dataverse datasets. We've actually started this integration on Docker level to get tools running in the separate containers as internal microservices, but support of external tools like Binder sounds very promising. Please keep us in the loop!

Best,
Slava

On Saturday, April 7, 2018 at 12:42:49 AM UTC+2, sean....@berkeley.edu wrote:

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse-community@googlegroups.com.

Vyacheslav Tikhonov

unread,
Apr 13, 2018, 6:09:08 PM4/13/18
to Dataverse Users Community
Hi Gustavo,

Just very simple but important use case to illustrate how we (DANS) are going to use this new feature with external tools: we want to show publicly open documents in viewer (there is excellent ViewerJS framework for that). 

We're running this viewer in the separate Docker container with preinstalled apache, it can download open file (by fileID and token) and put in the temporary location known by ViewerJS, and render it. It also supports spreadsheets and presentations in .ppt but the original file should be temporary converted to OpenDocument format inside of Docker container and cached for some time. ViewerJS can be placed as widget in any website, see more example here.

Of course, there is something to think about restricted files available only for depositor or his group with right permissions but in general it can be first great feature to open up content stored in Dataverse.

Best,
Slava
DataverseEU/DANS

kevin...@berkeley.edu

unread,
Apr 20, 2018, 8:20:16 PM4/20/18
to Dataverse Users Community
Hi Philip,

Thanks for the help. We're working on a prototype now. Please keep us in the loop if there are any updates with the external tools that may be of interest to us.

Thanks,
Kevin Yang and Sean Dooher 
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-community+unsub...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-community+unsub...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

Philip Durbin

unread,
May 25, 2018, 7:43:17 AM5/25/18
to dataverse...@googlegroups.com
Last night Aaron Culich posted a screenshot of the prototype (attached) and opened an issue called "Binderverse: integrating Binderhub with Dataverse (using docker+kubernetes)": https://github.com/IQSS/dataverse/issues/4714

Thanks!

Phil

To unsubscribe from this group and stop receiving emails from it, send an email todataverse-community+unsubscri...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-community+unsubscri...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsubscribe...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsubscribe...@googlegroups.com.

To post to this group, send email to dataverse...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-community+unsub...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.
binderhub-dataverse-prototype.png

Philip Durbin

unread,
Sep 18, 2019, 6:31:37 AM9/18/19
to dataverse...@googlegroups.com
Backend support for Dataverse in repo2docker was merged a few hours ago! Thank you to Kacper Kowalik from the Whole Tale project for his hard work on this!! (Whole Tale uses repo2docker under the hood.) The next logical step for Binder specifically is to add Dataverse to the drop down in the UI and I linked to the issue I created from this update I just made: https://github.com/IQSS/dataverse/issues/4714#issuecomment-532623346

Please stay tuned! And please feel free to dive into the discussion or help with code if you'd like!

Thanks!

Phil



To unsubscribe from this group and stop receiving emails from it, send an email todataverse-commu...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email todataverse-commu...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-commu...@googlegroups.com.
To post to this group, send email to dataverse...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-commu...@googlegroups.com.

To post to this group, send email to dataverse...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataverse-commu...@googlegroups.com.

To post to this group, send email to dataverse...@googlegroups.com.

Philip Durbin

unread,
Nov 21, 2019, 11:35:23 AM11/21/19
to dataverse...@googlegroups.com
Don't look now but it's already possible to use Dataverse dataset DOIs with https://mybinder.org **IF** you pick Zenodo from the drop down.

In about half an hour Ana Trisovic and I are going to hop on the JupyterHub/Binder Community Call to talk about https://github.com/jupyterhub/binderhub/pull/969 which adds Dataverse to the Binder UI.

All are welcome to join!


Thanks!

Phil
Reply all
Reply to author
Forward
0 new messages