Development: Any new versions of OpenRefine to come?

[Phil Anonymous]

Is there a developer's list, or some way to get a feel if/when there will be an updated version of OpenRefine released?  The current release candidate is from 2015.

I absolutely love OpenRefine, and am about 1/2 way through the book.  I have seen a lot of open source projects die, though, so before investing too much time, I wanted to get a feel for the future of OpenRefine.

Phi

[Thad Guidry]

The future of OpenRefine is rather bleak currently for a few reasons (but could change at any moment given a motivation by another developer or two coming into the picture again, or another new developer or two).

1. The 4 original developers including myself are busy with life (Here's one - http://davidhuynh.net/)

2. OpenRefine fits a very niche use case in the world of Software, so doesn't attract attention much or get a Hollywood star on the walk of fame.

3. There are other similar projects to OpenRefine like Talend Data Preparation and Trifacta Wrangler and MS PowerPivot, each of which has a slightly different reason for being than OpenRefine.

OpenRefine is a broader brush as a power tool for cleaning messy data, and well...that's about it...there is some reconciling support but after Freebase died, this feature needs further development love and integration with Wikidata.

I would say, continue on your journey with learning more about OpenRefine.

Besides the Book, go through our semi-official free course also : https://bigdatauniversity.com/courses/introduction-to-openrefine/

Because OpenRefine is a different kind of power tool with light scripting / programming available at command, it is a very good tool to have in your arsenal.  I would know because I am a Data Architect and use other commercial tools all the time, but do use OpenRefine sometimes daily for exactly its use case.  No other tool has customizable Facets, and the nice GREL syntax that makes OpenRefine easier to approach for scripting. (Thanks David Huynh for GREL and preview dialog !)

And who knows...like others you might enjoy OpenRefine so much, you'll eventually learn Java and Javascript just to contribute to fixing small bugs/issues, create new extensions, and working on feature requests in our project.

All the best,

Thad

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

[Patrick Maroney]

re: "OpenRefine fits a very niche use case in the world of Software, so doesn't attract attention much or get a Hollywood star on the walk of fame."

 

OpenRefine deserves a Hollywood star on the Analysts walk of fame. 

re: Because OpenRefine is a different kind of power tool with light scripting / programming available at command, it is a very good tool to have in your arsenal.

For those who have not discovered just how easy it is to drop your own custom Python scripts in to do your own Transformations, et al:  I highly encourage you to give it 5 minutes to test/understand.  I didn't realize how easy it was for over a year of using it daily.

 re:  "...you'll eventually learn Java and Javascript just to contribute to fixing small bugs/issues, create new extensions, and working on feature requests in our project."

For me, I'm still trying to fully learn Python after 20+ years, but it would be great to get some Jave/Javascript SMEs engaged for a little OpenRefine TLC!!!

re:  "The 4 original developers including myself are busy with life"

Enjoy, you've earned it!!!!   I know your efforts with OpenRefine (and the supporting community) have given many of us back days/weeks/months of our lives.  I know you have for me.   Just count the countless hours wasted staring at the spinning ball in Excel when trying to iterate  through a data set.

 

[Owen Stephens]

Thanks for this Patrick - I agree with everything you say here.

To go back to the original questions - paraphrased with my responses:

1) Is there a developer's list?

Yes - https://groups.google.com/forum/#!forum/openrefine-dev

2) Is there a some way to get a feel if/when there will be an updated version of OpenRefine released?

Not really - while I'm inclined to be slightly more upbeat that Thad in his reply, there is no doubt that the amount of resource available to OpenRefine is minimal and getting a release (or a new release candidate) out is difficult and the timing unpredictable. There is, of course, no guarantee there will ever be a final 2.6 release (although I feel it will get there eventually). However, on the plus side we are still seeing a low level number of commits and PRs in the code base - so while it may not be changing quickly, it is still alive as a project - not least because of the active user community as can be seen on this list

3) Is it worth investing time in learning OpenRefine?

Absolutely! Even if there is never another release of OR I'd still recommend investing time in it - if you work with data I'd say it is a key tool in your toolkit. When doing a 1 hour intro to OR I've had people say to me "this would have saved me hours" - so I believe that for people doing this kind of work, even if OR was only going to be viable for another 12 months, it would still be worth learning. However since OR 2.5 (released December 2011) still works 5 years later without any major issues I think it is a reasonably safe assumption that a version of OpenRefine will continue to work for several years even if all development stopped completely today.

So - I'd recommend investing now and starting to use OR as soon as you can (assuming it is relevant to your work!).

And finally to echo Patrick - thanks to those who have worked, and those who continue to work on OpenRefine - I use it almost everyday, and I find it not only helpful, but a pleasure to use :)

Owen