Few questions on Legend and adaption

[Marcus Van Hoof]

Hello all,

We are evaluating use of Legend internally and had a few questions. I would appreciate some insights from Legend experts and the community.

1. Apart from GS, is FINOS aware how many other institutions have started using or planning to actively use Legend?

2. Legend seems to be written in another language - Pure. I assume add-ons such as code generators, model importers need to be written in Pure too or can it be done in Java etc ? There seems very little technical documentation on Pure.. (we have some in-house xml-based models which we would like to import).

 
3. How vibrant is the Legend Open Source community? Are organisations other than GS making or planning to make design and code contributions to Legend/Pure?

4. How well does the Legend UI scale when having 100s of model objects? Is it possible to have a multi-project setup with dependencies to other Legend projects?

5. For due deligence, are there other products/tools comparable to Legend (Open Source or Commercial) in the market?

Sorry for so many question. I expect any answers will help other in future too.

Thanks,
Marcus

[Nelke, Beeke-Marie [Engineering]]

Hi Marcus,

 

Thanks for reaching out! We will come back to you shortly.

 

Thanks,

Beeke

 

__________________________________________________

Beeke-Marie Nelke

Tel: (212) 357-5755| E-mail: Beeke-Ma...@ny.email.gs.com

_____________________________________________________________

--
Looking for Legend docs? Visit the Legend documentation site at legend.finos.org.
 
Need an account on the Legend Studio shared instance? Request one at www.finos.org/legend
 
General Legend questions? Email legend-...@finos.org
 
To unsubscribe from this group, email legend+un...@finos.org. View this message at https://groups.google.com/a/finos.org/d/msg/legend/topic-id/message-id
---
To unsubscribe from this group and stop receiving emails from it, send an email to legend+un...@finos.org.

---

Your Personal Data: We may collect and process information about you that may be subject to data protection laws. For more information about how we use and disclose your personal data, how we protect your information, our legal basis to use your information, your rights and who you can contact, please refer to: www.gs.com/privacy-notices

[Nelke, Beeke-Marie [Engineering]]

Hi Marcus,

 

Apologies for the late reply. Please see answers below.

 

1. Apart from GS, is FINOS aware how many other institutions have started using or planning to actively use Legend? [B] This is unfortunately not something we can freely share with the community. There is an open issue where institutions can self-report looking into Legend – I’d recommend to watch this one. There are several major institutions that have installed Legend and are running proof-of-concepts right now. We also have an active community of Legend modeling working groups using the public shared instance of Legend. See more here.

2. Legend seems to be written in another language - Pure. I assume add-ons such as code generators, model importers need to be written in Pure too or can it be done in Java etc ? There seems very little technical documentation on Pure.. (we have some in-house xml-based models which we would like to import). [B] The Legend stack is written in Typescript and Java. The PURE language is used to define the data models incl. functions. You can learn more about it here. However, no knowledge of the PURE syntax is required to create data models and mappings as the UI supports business-friendly drag & drop operations. It is required for more complex operations though. We are working on adding more documentation to our Legend website.

 
3. How vibrant is the Legend Open Source community? Are organisations other than GS making or planning to make design and code contributions to Legend/Pure? [B] We have a Legend community of over 200 engineers and subject matter experts from 40+ financial institutions. We expect that more external contributions will come once Legend is used more widely beyond proof-of-concepts use cases.

4. How well does the Legend UI scale when having 100s of model objects? Is it possible to have a multi-project setup with dependencies to other Legend projects? [B] Internally at GS, Legend is supporting hundreds of model projects. Yes, project dependencies are possible but the functionality is not fully open sourced yet.

5. For due deligence, are there other products/tools comparable to Legend (Open Source or Commercial) in the market? [B] This is probably best answered by someone from FINOS or another FINOS member. There are other data modeling tools but we have not yet evaluated them against Legend.

 

Thanks,

Beeke

__________________________________________________

Beeke-Marie Nelke

Tel: (212) 357-5755| E-mail: Beeke-Ma...@ny.email.gs.com

_____________________________________________________________

 

[Marcus Van Hoof]

Hi Beeke,

Many thanks for taking time to reply, much appreciated. I have made some comments inline.

1. Apart from GS, is FINOS aware how many other institutions have started using or planning to actively use Legend? [B] This is unfortunately not something we can freely share with the community. There is an open issue where institutions can self-report looking into Legend – I’d recommend to watch this one. There are several major institutions that have installed Legend and are running proof-of-concepts right now. We also have an active community of Legend modeling working groups using the public shared instance of Legend. See more here.

[M] As suggested, I had a look at https://github.com/finos/legend/issues/392. There are no replies to date. I guess the danger in counting institutions performing evaluations or PoCs is that not all of them convert to real usages. 

2. Legend seems to be written in another language - Pure. I assume add-ons such as code generators, model importers need to be written in Pure too or can it be done in Java etc ? There seems very little technical documentation on Pure.. (we have some in-house xml-based models which we would like to import). [B] The Legend stack is written in Typescript and Java. The PURE language is used to define the data models incl. functions. You can learn more about it here. However, no knowledge of the PURE syntax is required to create data models and mappings as the UI supports business-friendly drag & drop operations. It is required for more complex operations though. We are working on adding more documentation to our Legend website.

[M] I had a look at what seems to be the main Legend repository. I downloaded and checked - there are 772 *.pure and 2022 *.java files.

Contrary to what is mentioned, these pure files have techincal implementation logic. For example see:

1. https://github.com/finos/legend-pure/blob/master/legend-pure-code-compiled-core/src/main/resources/core/json/jsonSchema.pure 
2. https://github.com/finos/legend-pure/blob/master/legend-pure-code-compiled-core-external-format-xml/src/main/resources/core_external_format_xml/binding/xsdToPure.pure

These 2 files have 1000+ lines of code. What tools/IDEs are used to work with these? 

I was trying to evaluate how we can create an importer of some kind to translate our internal models in XML to Legend, but it appears writing such a translator requires Pure programming knowledge. To reaffirm my understanding, I saw that most of the commits on this repo are changes to Pure files - which indicates a significant part of the new/recent development work happens in Pure. 

I understand the need for Pure syntax use in data models. but whats the motive to write Legend infrastructure-like code in Pure?

3. How vibrant is the Legend Open Source community? Are organisations other than GS making or planning to make design and code contributions to Legend/Pure? [B] We have a Legend community of over 200 engineers and subject matter experts from 40+ financial institutions. We expect that more external contributions will come once Legend is used more widely beyond proof-of-concepts use cases.

[M] I feel its going to be difficult for those outside of GS to make any significant technical contributions to Legend (or adapt and extend internally for bespoke work - my usecase) as those contributions will need to be in Pure, and there is very little documentation on using Pure. Writing  and maintaining 1000+ lines of code in a niche language with no development tools will be difficult to justify. 

4. How well does the Legend UI scale when having 100s of model objects? Is it possible to have a multi-project setup with dependencies to other Legend projects? [B] Internally at GS, Legend is supporting hundreds of model projects. Yes, project dependencies are possible but the functionality is not fully open sourced yet.

[M] I think for FINOS Lengend users, the dependencies will be a much needed feature when starting to deal with more realistic multi-domain projects. Looking forward to more more info on this.

5. For due deligence, are there other products/tools comparable to Legend (Open Source or Commercial) in the market? [B] This is probably best answered by someone from FINOS or another FINOS member. There are other data modeling tools but we have not yet evaluated them against Legend.

[M] Thanks, will await any comments. Given GS's continued investment and oversight of Legend, it must serve its purpose well for GS. For the rest of us, it raises many questions we need to answer internally. 

Thanks,

Marcus