any activity?

[narayaan]

Is this still an active project?

[ojoven]

It doesn't look like so. Their API link is broken. I'm interested in this project, too.

[hanaldo]

Thanks for your interests, we are currently struggling with the progress, but our demo server is still running at here, feel free to check it out.

[bitscuit]

any plans for creating a contributing community around this? Might help with the progress

[Marc-Bogonovich]

Hi Bitscuit, a community would be fantastic. The code is open, so we could work on it as a group.

[bitscuit]

The repo would need a license before anyone besides the owners can really start contributing. Instructions on how to build the project along with contributing guidelines would also help.

[Marc-Bogonovich]

These decisions would be good step at this point. Let me try to explain why we haven't worried about these two issues before.

Previously, if I can encapsulate this in a few sentences.. Our strategy was to find one institution that really supported open source, and would build this technology with us, inside their classroom, our services and work all done at cost. This nearly materialized on a few occasions, but did not pan out (we even had one grant funded, and then the funds failed to materialize). So our thought was to produce an MVP, in a small community, in the field/in vivo (where the system would be used with students), and at that moment invite an even wider community to do whatever they wanted with the code or build it further with us.

In the past year, I've been thinking about what to do next. And if there is a more traditional approach to open source development possible, then we can't turn our back on that.

So as @bitscuit suggests, we should clarify which open source license we adopt, and develop some guidelines for contribution. Any input on these issues would be appreciated.

[bitscuit]

I'm not an expert, so feel free to take my suggestions with a grain of salt. There is a list of popular open source licenses, and which one you want to adopt would depend on what you want to do with Openwords.

For contribution guidelines, you could take a look at how some other well-known communities do this. One example of a very formal/well-developed community would be Kubernetes, which has a governing body. Of course, you don't need to do everything these communities are doing. Applying everything in the example I gave seems over the top considering the size of this project in its current state. The guidelines could be simple to start with, e.g.

• steps on building the project
• steps on opening PRs and merging code (maybe require that all tests pass)
• who the maintainers are for contacting when there are issues

As time goes on and the project matures, you could add to the guidelines as necessary

[jniles]

Hi all, has there been any evolution on the license discussion?

[Marc-Bogonovich]

Hi @jniles, sorry for the delayed response. I'm going to have some in-person conversations with people in the Hong Kong Open Source organization to get their input on this question. They are having their conference this Friday. Let me get back to this thread after those conversations, and we'll have a few initial proposals.

[marie-joechahine]

Hello, anything new on this? Seems like an interesting project

[bubbster95]

Hey, just dropping in 3 years later... after covid... I assume this project took a break. I would love to make some contributions to build my portfolio a little. Is this project still alive?

[Marc-Bogonovich]

Hey, just dropping in 3 years later... after covid... I assume this project took a break. I would love to make some contributions to build my portfolio a little. Is this project still alive?

Hi Bubbster95, The project isn't active. But I appreciate your interest in building and contributing. I visited your website, very fun! I'd be happy to chat with you, see if there's something you could do that would help with your portfolio. These days I'm tutoring privately, and have considered distilling something from Openwords specifically for my students. But haven't figured out which component. Marc

[bubbster95]

Sorry to hear that. I'm still just shopping around to try and find a repo to contribute to. The issue is so many projects are disinteresting or dead. What kind of data do you have stored for openwords? As far as languages and their translations. I'm interested to see what we could do with it.

On Sun, Sep 10, 2023, 9:32 PM Marc-Bogonovich @.***> wrote:

Hey, just dropping in 3 years later... after covid... I assume this project took a break. I would love to make some contributions to build my portfolio a little. Is this project still alive?

Hi Bubbster95, The project isn't active. But I appreciate your interest in building and contributing. I visited your website, very fun! I'd be happy to chat with you, see if there's something you could do that would help with your portfolio. These days I'm tutoring privately, and have considered distilling something from Openwords specifically for my students. But haven't figured out which component. Marc

— Reply to this email directly, view it on GitHub https://github.com/Openwords/OpenwordsWebapp/issues/7#issuecomment-1713024101, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOBQB36Y5VPDU5QBMTZUXJTXZZSZZANCNFSM4KVHE4PA . You are receiving this because you commented.Message ID: @.***>

[Marc-Bogonovich]

I think the main thing Openwords has is 1) the concept of the Openwords Markup Language (for expressing language learning problems/exercises) and that format's relationship with a webapp that reads that data.. and 2) We have one implementation of the app that reads that data. Regarding linguistic data: The linguistic data that one might want are massively available with a little scraping, or with a little prompting toward AI. I've actually gone ahead and tested different AIs in their ability to generate language problems in our format, and that doesn't seem to be a problem. It is well within their capability.

In terms of the two things I listed above, I say the concept of the Openwords Markup Language (OWML) because I'm no longer sure that our particular implementation for expressing language learning problems was the best one. Really perhaps we should have designed the OWML in an existing text-based format rather than building our own, or at least made ours extensible with other formats like Moodle's GIFT format (you should check that out, it is pretty cool).

I have considered rebuilding Openwords this time in a different programming language. Either Javascript or Python. I'm much more familiar with the latter but also can use the former. We got very close to a full implementation. But the application isn't terribly complex either.

And I still think this project is pressing, and still solves UX pain points for both users and communities. The development of AI LLMs.. rather than replacing the value for a system like Openwords, makes the development of this project much more feasible.

These are some of my thoughts. If you want to chat verbally, I'm in Hong Kong these days, which means I'm 12 hours off of Eastern Standard time. Marc

[bubbster95]

Okay interesting, I originally thought theain thrust was that you had a huge block of learning data but now that I think about it the code makes more sense for what you described.

My intention was always to allow real time user interaction. Multi-player learning mode. It's always been a disappointment how duolingo handles it. I love their vast curriculums and the Ai they use to format the questions but that is all way out of my reach. I'm a little too busy to build an entire project like I used to.

Ideally contributing to something bigger in my free time is what I am interested in. I'll just have to keep looking around.

On Mon, Sep 11, 2023, 1:04 AM Marc-Bogonovich @.***> wrote:

I think the main thing Openwords has is 1) the concept of the Openwords Markup Language (for expressing language learning problems/exercises) and that format's relationship with a webapp that reads that data.. and 2) We have one implementation of the app that reads that data. Regarding linguistic data: The linguistic data that one might want are massively available with a little scraping, or with a little prompting toward AI. I've actually gone ahead and tested different AIs in their ability to generate language problems in our format, and that doesn't seem to be a problem. It is well within their capability.

In terms of the two things I listed above, I say the concept of the Openwords Markup Language (OWML) because I'm no longer sure that our particular implementation for expressing language learning problems was the best one. Really perhaps we should have designed the OWML in an existing text-based format rather than building our own, or at least made ours extensible with other formats like Moodle's GIFT format (you should check that out, it is pretty cool).

I have considered rebuilding Openwords this time in a different programming language. Either Javascript or Python. I'm much more familiar with the latter but also can use the former. We got very close to a full implementation. But the application isn't terribly complex either.

And I still think this project is pressing, and still solves UX pain points for both users and communities. The development of AI LLMs.. rather than replacing the value for a system like Openwords, makes the development of this project much more feasible.

These are some of my thoughts. If you want to chat verbally, I'm in Hong Kong these days, which means I'm 12 hours off of Eastern Standard time. Marc

— Reply to this email directly, view it on GitHub https://github.com/Openwords/OpenwordsWebapp/issues/7#issuecomment-1713170840, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOBQB32XW2RUVACIZPXCV5LXZ2LW5ANCNFSM4KVHE4PA . You are receiving this because you commented.Message ID: @.***>