Abstract Wikipedia

134 views
Skip to first unread message

Peter Ljunglöf

unread,
Aug 26, 2020, 6:30:31 AM8/26/20
to Grammatical Framework
Hej GF!
Abstract Wikipedia is a new grand project by the Wikimedia foundation, and it seems that GF could play a part there if someone wants to take the lead.

"It is my honor to introduce Abstract Wikipedia, a new project that has been unanimously approved by the Wikimedia Foundation Board of Trustees. Abstract Wikipedia proposes a new way to generate baseline encyclopedic content in a multilingual fashion, allowing more contributors and more readers to share more knowledge in more languages. It is an approach that aims to make cross-lingual cooperation easier on our projects, increase the sustainability of our movement through expanding access to participation, improve the user experience for readers of all languages, and innovate in free knowledge by connecting some of the strengths of our movement to create something new."
(https://meta.wikimedia.org/wiki/Abstract_Wikipedia/July_2020_announcement)

"The goal of Abstract Wikipedia is to let more people share in more knowledge in more languages. Abstract Wikipedia is an extension of Wikidata. In Abstract Wikipedia, people can create and maintain Wikipedia articles in a language-independent way. A Wikipedia in a language can translate this language-independent article into its language. Code does the translation.
Wikilambda is a new Wikimedia project that allows anyone to create and maintain code. This is useful in many different ways. It provides a catalog of all kind of functions that anyone can call, write, maintain, and use. It also provides code that translates the language-independent article from Abstract Wikipedia into the language of a Wikipedia. This allows everyone to read the article in their language. Wikilambda will use knowledge about words and entities from Wikidata."
(https://meta.wikimedia.org/wiki/Abstract_Wikipedia)

There is a working paper on ArXiV: https://arxiv.org/abs/2004.04733

And most important, GF is mentioned in the "Talk:Abstract Wikipedia" page:

"Denny mentioned Grammatical Framework and I took a look at it. I think it is complex enough to represent most grammatical phenomena, but I don’t see very much that is actually developed. The examples and downloadable samples I see are all small toy systems. It isn’t 100% clear to me that any grammatical framework can actually capture all grammatical phenomena—and with certain kinds of edge cases and variation in dialects, it may be a lost cause—and linguists still argue over the right way to represent phenomena in major languages in major grammatical frameworks. Anyway, it looks like the Grammatical Framework still leaves a lot of grammar development to be done; assuming it’s possible (which I’m willing to assume), it doesn’t seem easy or quick, especially as we get away from major world languages."
(https://meta.wikimedia.org/wiki/Talk:Abstract_Wikipedia#The_"Grammatical_Framework"_grammatical_framework)

Their description of GF is slightly negative ("I don’t see very much that is actually developed" / "The examples and downloadable samples I see are all small toy systems" / "it doesn’t seem easy or quick, especially as we get away from major world languages") - perhaps some GF:er can step in...? :)

best,
Peter Ljunglöf

PS. It was Tobias Kuhn who informed about Abstract Wikipedia on the CNL mailing list:

> Från: Tobias Kuhn <kuhnt...@gmail.com>
> Ämne: [CNL] Abstract Wikipedia
> Datum: 3 juli 2020 06:51:45 CEST
>
> Hi all,
>
> This announcement of Abstract Wikipedia might be interesting to this list: https://meta.wikimedia.org/wiki/Abstract_Wikipedia/July_2020_announcement
>
> It's not directly about controlled natural language, but I think CNLs would be an obvious choice to implement the user-facing part.
>
> Cheers,
> Tobias


------- ------ ----- ---- --- -- - - - - -
peter ljunglöf
peter.l...@gu.se
data- och informationsteknik, och språkbanken
göteborgs universitet och chalmers tekniska högskola
-------------- --------- -------- ------- ------ ----- ---- --- -- - - - - -


Aarne Ranta

unread,
Aug 28, 2020, 2:42:56 AM8/28/20
to gf-...@googlegroups.com
Peter, this looks like something to react to indeed. Do you know what is the best way to contribute to the discussion?

My suggestion would have have a few items:

- that we develop a high-level API for the purpose, as done in many other NLG projects
- that we make a case study on an area or some areas where there is adequate data. For instance from OpenMath
- that we propagate this as a community challenge
- Digital Grammars can sponsor this with some tools, since we have gained experience from some larger-scale NLG projects

Regards

  Aarne.







--

---
You received this message because you are subscribed to the Google Groups "Grammatical Framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gf-dev+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gf-dev/CBF38F53-129A-412C-9E51-15437D6AD7F4%40heatherleaf.se.

Giuliano Lancioni

unread,
Aug 28, 2020, 4:06:13 AM8/28/20
to gf-...@googlegroups.com
Dear Aarne,

I think that the GF WordNet project should absolutely be mentioned in this connection too. It goes far beyond any definition of toy grammar and lexicon (although the lack of WSD tends to return too many meanings, which leads to combinatorial explosion).

Giuliano

Aarne Ranta

unread,
Aug 28, 2020, 5:05:14 AM8/28/20
to gf-...@googlegroups.com
Dear Giuliano,

Yes, gf-wordnet is definitely something to mention in this connection, as a large-scale development.

Regards

  Aarne.

Inari Listenmaa

unread,
Aug 28, 2020, 2:54:08 PM8/28/20
to gf-...@googlegroups.com
I'm interested, and willing to devote my free time for this!

Some of the concerns on the talk page are definitely valid, but worrying about stuff like ergativity or when to use subjunctive tells me that the commenters haven't understood just how abstract an abstract syntax can be. Regarding the mention about only toy examples being available, I understand how an outsider can get that view, but there is a lot of grammatical knowledge encoded in the RGL. It's far from covering all possible things people might want to say in a wikipedia article, but if the focus is in generation, that's not really an issue. An incomplete tool that covers the most common use cases, or covers a single domain, is still very useful.

We also have quite a few low-resource languages in the RGL, though not many with full coverage. Personally, if I knew that some language would come in use, I would be super motivated to write new RGs.

I like the comment by someone in the thread: "It is true that a lot of development would be required for GF, but even more development would be required for any other framework. GF being open source, any development would flow back into the general public, too, so working with GF and its developers is likely the correct solution."

In which I'd like to add that it's true that GF is hard, but with any formalism encoding natural language is hard.

Cheers,
Inari

Roman Suzi

unread,
Aug 29, 2020, 2:02:12 AM8/29/20
to Grammatical Framework
hi!
Definitely interesting project! Whatever system Abstract Wikipedia will choose, that will follow the evolutionary path of GF, so it's better to learn from that regardless of whether they choose GF or not, with or without resource grammars. Especially, initial language diversity. (English is complete disaster to start with aiming for so many languages.)

For proof of concept - having Lua bindings + plus some lexicon handling in Lua is perhaps the easiest way, as the mediawiki already supports Lua scripts. And Lua can probably be "transpiled" from javascript (https://github.com/PaulBernier/castl ), I have not tried though. However, instead of Wikipedia, Wiktionary is perhaps a better playground for starting (some student project). Wiktionaries already use Lua for inflections.

-Roman

Aarne Ranta

unread,
Aug 31, 2020, 2:59:18 AM8/31/20
to gf-...@googlegroups.com
Inari,

Thanks for an excellent analysis and your plan to work on this! I hope you can get followers so that we can set up something convincing.

I also wonder how best to communicate this to Wikimedia? We should try and answer directly to them, right? 

Regards

  Aarne

Aarne Ranta

unread,
Aug 31, 2020, 3:14:27 AM8/31/20
to gf-...@googlegroups.com
Roman,

Thanks for your comments, too! It is a good point that whoever wants to build something like that will follow the path of GF. And it will take long: in particular, developing the RGL has taken 20 calendar years and (at least) 20 person years.

I'm not quite sure what you mean by targeting Wiktionary. I (and certainly others) have occasionally used their inflection tables to build GF lexica, and they seems to have a good quality. But coverage is certainly a problem - maybe we could help populating them with data for more words. Is this what you mean?

I would be in particular interested in generating factual Wikipedia content via NLG. We would need to target some specific enough domain to build up an abstract syntax and API for that, and then scale up rapidly. Tobias and Kaarel made a prototype multilingual Wiki in the MOLTO project, with a cool interface and multilingual ACE as the core grammar. We could perhaps build upon that grammar. 

GF-ACE has also been used in a commercial project by Digital Grammars and Signatu, formalizing the concepts in the GDPR legislation with an abstract syntax and linearizations to 5 languages. A free demo can be found here:


In connection to that, we extended the basic GF-ACE, and that part (but not the GDPR lexicon itself) can be made available open source if there is interest. The lexicon has almost 3000 abstract syntax functions, some of them fairly complex multiwords, which show the abstractness of abstract syntax quite well, illustrating the points that Inari mentioned in her mail.

  Aarne.



Kristian Kankainen

unread,
Aug 31, 2020, 3:23:13 AM8/31/20
to gf-...@googlegroups.com
Hello,

I want to contribute a thought in this thread. Wikipedia is doubtlessly a language-diverse platform and sports many un(der) resourced minority languages. My master's thesis generates automatically the source code for a GF morphology module from inflection tables (using the work of [1]). Since the Wiktionary is a popular place for inflection tables, these could be used for boot-strapping GF resources for those languages. Moreover, but not related to GF nor Abstract Wikipedia, the master's thesis generates also FST code and integrates the language into the Giella platform which provides an automatically derived simple spell-checker for the language contained in the inflection tables.

Coupling or "boot-strapping" the GF development using available data on Wiktionary could be seen as a nice touche and would maybe be seen as a positive inter-coupling of different Wikimedia projects.

I am also willing to contribute as much as I can in the project.

Best regards,
Kristian Kankainen

[1] Markus Forsberg and Mans Hulden. Deriving Morphological Analyzers from Example Inflec-
tions. In LREC, 2016.

Inari Listenmaa

unread,
Aug 31, 2020, 10:41:02 AM8/31/20
to Grammatical Framework
Hi all,

As for how to participate, I asked on IRC and got some answers:

1) Abstract Wikipedia has a mailing list https://lists.wikimedia.org/mailman/listinfo/abstract-wikipedia, participating in the mailing list might be the most obvious way.
2) In addition to or instead of writing to the mailing list, it would be useful if GF people could compile a shared summary of comments to be posted on the talk page à la https://meta.wikimedia.org/wiki/Talk:Abstract_Wikipedia#Some_questions_and_concerns
3) There's also a broader discussion on the risks wrt minorities/smaller languages https://lists.wikimedia.org/pipermail/wikimedia-l/2020-August/095399.html, where the conclusion seems to be "WMF would love to hire a GF expert who's also a native speaker of [insert suitable language]". Or maybe just a suggestion of which "minority languages" / language families could be worth focusing on in the prototype phase.

Regarding point 3, they suggest to involve a non-European and underrepresented language as a first-class development target. I think that sounds promising, and GF community has talent that could be useful for them. Would e.g. our Bantu language community be interested in participating? (Or anyone else reading this who fits the description!) It makes sense to be a native (or fluent) speaker of the language, because here the challenge isn't just to get a correct word order or agreement, it's to actually produce text that sounds idiomatic.

Regarding points 1 & 2, we need some coherent writeup to offer to the Wikimedia community (on the mailing list and/or talk page). The content of such a write-up depends on whether we find a good suggestion for a language and people to develop a prototype in it. This thread has lot of good ideas, so our message shouldn't be too hard to write, just put the points and suggestions from this mail into a more coherent format.

Inari

Aarne Ranta

unread,
Aug 31, 2020, 10:57:19 AM8/31/20
to gf-...@googlegroups.com
Inari,

Thanks for the constructive suggestion! It does make sense to give a concentrated answer to the main points, and your previous mail was a good start to it. Did you already join their mailing list?

As for under-resourced languages, the work on Bantu languages is indeed a good idea. What about for instance the South-African work on health care translation: could that be converted or developed to small articles on important health issues?

  Aarne.

--

---
You received this message because you are subscribed to the Google Groups "Grammatical Framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gf-dev+un...@googlegroups.com.

Aarne Ranta

unread,
Aug 31, 2020, 11:00:06 AM8/31/20
to gf-...@googlegroups.com
Kristian,

Thanks for these pointers! This sounds like an obvious thing to work on. I would love to see Wiktionary resources that are easier to fetch data to GF from, and in return use GF in whatever ways it helps building such Wiktionaries.

  Aarne,

Laurette Marais

unread,
Aug 31, 2020, 11:06:15 AM8/31/20
to Grammatical Framework
Hi everyone,

A quick response from South Africa. :) The other Laurette and I, along with a fluent (non-native speaker) Zulu linguist, are working towards a Zulu RG that will hopefully be relatively comprehensive within the next 3-6 months. (WIP code available here for anyone interested in having a look: https://github.com/LauretteM/gf-rgl-zul. This indeed was used in the health care app Aarne mentioned.). I'm glad to say that a significant chunk of our time will be available to work on the RG and any related work for the next two years.

So we have a project team that contains a fluent speaker, and we will be involving native speakers as part of our project as the need arises.

Laurette Snr will probably have more to say about current efforts to develop Wikipedia for the South African languages, but as far as the Zulu RG goes, we would definitely be keen to get involved here. Two PhD students are currently working on Xhosa and Northern Sotho, and I'm sure they will be interested in getting involved as well.

Regards,
Laurette M


Laurette Marais

unread,
Aug 31, 2020, 11:57:50 AM8/31/20
to Grammatical Framework
A follow up note with regards to the health application and its content: the use case was extremely specific (binary questions during a maternal health consultation), and the needs of the domain and use case very much drove the development of the RG in its current form. It is not immediately clear to me that it could be re-used without significant development to the RG, but that is luckily exactly what we are doing. At least various concepts in the domain have already been expressed in Zulu using a GF grammar that uses the existing RG, and so I agree that this could serve as a starting point.

I attach a conference paper we presented at Fusion 2020, which gives an overview of the application and the role of GF in translating.

Fusion2020_Marais_et_al_final.pdf

Aarne Ranta

unread,
Aug 31, 2020, 1:52:55 PM8/31/20
to gf-...@googlegroups.com
Laurette,

Thanks for your quick and thorough response! It may be that you already have enough for nontrivial NLG tasks: one does not need a complete RGL for that. 

This is getting very exciting so it is a pity I have to leave the forum for a while because of vacation away from computers and internet. But I will be back and join it again next week.

Aarne

Inari Listenmaa

unread,
Aug 31, 2020, 2:07:59 PM8/31/20
to gf-...@googlegroups.com
Sounds great! Congrats on having time available for RG development ^_^

Would you (some subset of youPl_Pron in ZA) be interested in taking the role of communication? There's no hurry to get a message out right now, we can also wait for more points of view in this thread. I just want to make sure that someone feels like its their task to take contact. :-P I could also do it in case you don't want to do it, but it feels more natural to have direct contact from the most relevant people.

I signed up to the Abstract Wikipedia mailing list. I'm planning to just lurk for now and comment when I see something I could contribute to.

Inari

Laurette Marais

unread,
Aug 31, 2020, 2:30:36 PM8/31/20
to Grammatical Framework
Thanks, haha, we're very pleased about it ourselves.

We are very keen to be involved, but I think communication from the GF side would be better handled by someone like you, who has wide-ranging experience in developing RGs for various languages.

This really is very exciting!

Inari Listenmaa

unread,
Aug 31, 2020, 2:38:53 PM8/31/20
to gf-...@googlegroups.com
Alright, I'll write something! Probably not before the weekend, so people have time to contribute more.

Inari 

Inari Listenmaa

unread,
Sep 7, 2020, 4:49:06 AM9/7/20
to gf-...@googlegroups.com
I have added a summary + some of my own suggestions here https://meta.wikimedia.org/wiki/Talk:Abstract_Wikipedia#Response_from_the_Grammatical_Framework_community. To quote myself and Emil:

Of course, for this to be a viable solution, more people than just me need to join in. I believe that if the GF people know that their grammars will be used, the motivation to write them is much higher. To kickstart the resource and API creation, we could make Abstract Wikipedia as a special theme of the next GF summer school, whenever that is organised (live or virtually).

I second what inariksit says about the interest from the GF community - if GF were to be used by AW, it would give a great extra motivation for writing resource grammars, and it would also benefit the GF community by giving the opportunity to test and find remaining bugs in the grammars for smaller languages.

If you would like to contribute, please add your support to the talk page. You can create a user account easily, and when you edit the page, sign your comment by writing ~~~~ (four tildes in a row), it will transform into a signature and a time stamp.

Inari

Aarne Ranta

unread,
Sep 7, 2020, 12:38:29 PM9/7/20
to gf-...@googlegroups.com
Inari, all, thanks for the great job and a thorough and convincing answer! I will be happy to confirm my support in Wikimedia as well.

  Aarne.

--

---
You received this message because you are subscribed to the Google Groups "Grammatical Framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gf-dev+un...@googlegroups.com.

Alexandre Rademaker

unread,
Sep 8, 2020, 10:58:47 PM9/8/20
to gf-...@googlegroups.com

check out other grammar formalisms, like HPSG, you'll see similar coverage to GF, but no unified API for different languages.

Hi Inari,

This is not 100% true. Do you know the http://matrix.ling.washington.edu/index.html from Emily Bender?

Alexandre 
Sent from my iPhone

On 7 Sep 2020, at 05:49, Inari Listenmaa <inari.l...@gmail.com> wrote:

I have added a summary + some of my own suggestions here https://meta.wikimedia.org/wiki/Talk:Abstract_Wikipedia#Response_from_the_Grammatical_Framework_community. To quote myself and Emil:

Steve Braich

unread,
Sep 9, 2020, 3:21:17 AM9/9/20
to gf-...@googlegroups.com
Hello, 
Sorry if I'm late to join the conversation here.  But here is my two cents:

I really think that Inari Listenmaa research paper "Outsource Morphology in GF: a case study for Hungarian" is a perfect blueprint for the future of GF.  

I replicated her work for Serbian for a morphology class. 
Leveraging Morphological Resources in RBMT: The Case for Serbian

Yes it's a toy mini-resource that I made.  But Inari's vision of leveraging existing morphological resources just seemed like the obvious direction GF should go especially after I took a look at the Russian module.  It was impossible to use the Russian resource as a template for Serbian.  

However, if the morphology were streamlined, then I could have leveraged and modified the pure syntax from Russian into Serbian.

Streamline the morphology and then focus on syntax.  OK, so maybe for morphologically poor languages, this won't have such an impact, but for me it just seems as a no-brainer. 

Hope I am not oversimplifying this, but Inari's work really inspired me.

Thanks, 
Steve


--

---
You received this message because you are subscribed to the Google Groups "Grammatical Framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gf-dev+un...@googlegroups.com.

Inari Listenmaa

unread,
Sep 9, 2020, 3:30:20 AM9/9/20
to gf-...@googlegroups.com
Hi Alexandre,

Thanks for the correction! I'll update my statement on the AW talk page.

Inari

--

---
You received this message because you are subscribed to the Google Groups "Grammatical Framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gf-dev+un...@googlegroups.com.

Inari Listenmaa

unread,
Sep 9, 2020, 3:42:01 AM9/9/20
to gf-...@googlegroups.com
Hi Steve,

I agree that this would be a scalable way to develop new languages. If we target languages that have existing morphological resources, then we could do a whole bunch of related languages at one go. It would also improve performance for morphologically complex languages. On the other hand, I've been working on Malay and Indonesian the past weeks (code not in gf-rgl yet), and the morphology is rather simple, it would just add an unnecessary layer of complication.

Note that my paper on Hungarian is also just a toy mini resource grammar! :-P If I remember correctly, Aarne made a larger-scale version for Finnish some years ago. 

Inari

Aarne Ranta

unread,
Sep 9, 2020, 6:34:52 AM9/9/20
to gf-...@googlegroups.com
Hello Alexandre,

Thanks for pointing this out. This reminds me that I should write a little summary of what would be involved in reproducing the RGL, and what our starting point was back in 2001. So here it is:

The main inspirations of GF-RGL were

- XFST, Xerox Finite State morphologies for several languages
- CLE, Core Language Engine, an SRI-Cambridge-Telia etc project for building syntax modules for some languages to be used in applications

The main lesson learned was

- make it open source and involve a community. CLE in practically disappeared because nobody had the rights to continue with it, and XFST was increasingly replaced by open-source variants

This brings us to

- the LinGO matrix, using HPSG, an open-source project

also an inspiration for the RGL, just a bit later, but still alive and active. The difference we wanted to make was

- think about non-linguist programmers as the majority of users

This led us to 

- design GF and its module system in a way similar to programming languages, rather than grammar formalisms
- separate the linguist's view (the internals of the RGL) from the application programmer's view (the RGL API)

The closest GF counterpart of the LinGO matrix is thus the internal abstract syntax of the RGL. But when looking at the LinGO/DELPH-IN documentation back in 2003 and still today, I cannot see anything corresponding to the API. It is more of a linguists' project than of programmers'. And I think it would be quite a job to develop it into an API direction similar to GF. Not only is the starting point less friendly to that (with GF's formal distinction between abstract and concrete syntax), but even in the GF world, it took several years to bring the module system and the compiler into a state that smoothly supports the division of labour between linguists and application programmers in the way we do.

This said, HPSG has reached longer in their linguistic coverage in many languages, in particular in the English RGL: GF has nothing like that, and again it would take years of work to build it.

Of course, the nicest thing would be to share resources in a formalism independent way. This looks quite feasible in the case of morphological lexica, and is an ongoing practice already. But when it comes to syntax, I am less sure. Syntax code in GF and HPSG and other higher-level (above context-free) formalisms is essentially like code in different programming languages. There the practice is that each language has to build their standard libraries from scratch (think about for instance collections and generics in Java, C++, Haskell,...) An alternative is to enable foreign function interfaces (like from Python to C), but I cannot see very concretely right now how this would look for instance between GF and HPSG - and how much there would really be to gain. But of course we have mutual communication, for instance by co-organizing GEAF workshops (Grammar Engineering Across Frameworks), and see each other as allies rather than enemies.

ParGram (in LFG) could also be mentioned, but it used to be a proprietary system that was more difficult to learn from.

Regards

  Aarne.





























--

Roussanka Loukanova

unread,
Sep 9, 2020, 6:44:26 PM9/9/20
to gf-...@googlegroups.com
Hi Aarne, Hi Everybody,

Actually, I tried and extracted some common syntax from several grammar formalisms. I ended up with something that I call Generalised Constraint Based Lexical Grammar (GCBLG), by also generalising some common linguistic characteristics of languages (a little bit). Among the grammar formalisms, I had in mind GPSG, HPSG, LFG, Categorial Grammars (CG), but also generalizations across syntactic categories (which is also available in CG operators).

I used a typed version of Feature-Value Language that  is to some degree a generalization of M. Johnson (1991) Logic and feature structures.

When I was learning a little bit of GF, perhaps that was around 2007-2009, as I quickly recall the time now, what was caming to my mind, spontaneously on the spot, was seeing in GF abstract syntax what I had in GCBLG.

One thing is for sure, what I have as GCBLG is not HPSG, despite superficial resemblances of feature-value descriptions and some principles using a notion of a head component.

Up to now, as I recently looked into GF abstract syntax, from a fresh perspective, I think that there is a good, actually great, background for developing shared resources. I am convinced that is realistic work, at least, at first by developing a theoretical framework for shared computational interfaces between lexicon-syntax-semantics, at abstract levels, by taking the GF lesson. Something in such lines would be very valuable. Indeed there are difficulties, as Aarne describes them.

An important characteristic of HPSG that makes sharing resources difficult is extracting and expanding all available linguistic information from the components, including the concrete language syntax. Lexemes, which are not expanded into full words to be included in phases, down to certain levels remain abstract. This makes it possible to share resources of such lexicals.

I think that what Aarne describes as a quite feasible possibility for sharing resources, in the case of morphological lexica, is actually feasible for syntax of more complex expressions. But, indeed, that would be with quite serious work, may be by more than one person, as I have been trying to figure out  on my own :-)

Best Regards,
Roussanka

Roman Suzi

unread,
Sep 12, 2020, 2:29:24 AM9/12/20
to Grammatical Framework
Hi Steve,
I've re-implemented Russian RG totally this Summer. While I started from Czech, and there were many similarities in the noun and even numeral systems, I believe it's not possible to have enough synergy between Northern and Southern Slavic languages for a common functor. Maybe, Ukrainian and Belarusian can benefit from Russian RG or even share a functor, but hardly others. (I am not a professional linguist, so please take this with a grain of salt). Nonetheless, you can take a look at the new grammar https://github.com/rnd0101/gf-rgl/tree/new-rus-rg to see if it's any better for Serbian than before. It will be merged to the RGL at some moment. I do not know how common participles and transgressives are used in Serbian, I feel like currently GF does not have good enough support for those (well, it may be too fundamental problem). In English and many other languages participles and converbs can be included in verbs' paradigm. In Slavic languages (I may be mistaken in generalizing) "converbs" can't be inside the verb paradigm, because they are too much like adjectives and it will inflate the resulting grammars beyond reasonable memory limits.

With best regards,
Roman

Inari Listenmaa

unread,
Oct 11, 2020, 9:12:59 AM10/11/20
to Grammatical Framework

Hi,


@Inariksit:@Skarpsill: - thank you for your message, and thank you for reaching out. I am very happy to see the interest and the willingness to cooperate from the Grammatical Framework community. In developing the project, I have read the GF book at least three times (I wish I was exaggerating), and have taken inspiration in how GF has solved a problem many times when I got stuck. In fact, the whole idea that Abstract Wikipedia can be built on top of a functional library being collected in the wiki of functions can be traced back to GF being built as a functional language itself.
I would love for us to find ways to cooperate. I think it would be a missed opportunity not to use learn or even directly use the RGLs.
I keep this answer short, and just want to check a few concrete points:
    • when Aarne mentioned to "develop a high-level API for the purpose, as done in many other NLG projects", what kind of API is he thinking of? An API to GF, or an abstract grammar for encyclopaedic knowledge?
    • whereas math would be a great early domain, given that you already have experience in the medical domain, and several people have raised the importance of medical knowledge for closing gaps in Wikipedia, could that be an interesting early focus domain?
    • regarding our timeline, we plan to work on the wiki of functions in 2021, and start working on the NLG functionalities in late 2021 and throughout 2022. Given that timeline, what engagement would make sense from your side?
I plan to come back to this and answer a few more points, but I have been sitting on this for too long already. Thank you for reaching out, I am very excited! --DVrandecic (WMF) (talk) 00:44, 26 September 2020 (UTC)

I think personally that the timeline of NLG in late 2021 and throughout 2022 sounds reasonable. How are the South African team feeling about that timeline?

Regarding Denny's question: Aarne, what kind of an API were you thinking about?

Note that anyone may answer to the wiki talk page, I'm not an official representative of the GF community. If you're interested in the developments, you can subscribe to get notifications whenever someone writes on the talk page.

Inari
Reply all
Reply to author
Forward
0 new messages