InterSystems Official
· Jul 9, 2018

Atelier and Studio

Atelier has reached a threshold of feature completeness. Atelier 1.3 which is currently in flight will mainly focus on supporting the latest release of Eclipse Photon. After release, Atelier and Studio will continue to receive critical software corrections.

Discussion (19)8
Log in or sign up to continue

Well, now a few questions.

Does it mean that any new requested features will be declined to implement?

Any possibility to implement new features by own?

How about the future of Atelier API?

If InterSystems not going to extend Atelier anymore, maybe it's time to open sources Atelier or at least Atelier API.

Currently, it sounds like Atelier will get the same future as Studio, it means no future. Or am I wrong?

@Andreas Dieckow Does this mean it is the end of Atelier? and if so, why?

This post from Bill McCormick (https://community.intersystems.com/post/why-atelier-and-what-about-studio) mentioned Atelier being on the strategic roadmap for Intersystems as well as being the IDE of choice in the future. Please confirm which IDE will Intersystems be using going forward? one that will be continuously developed and enhanced. Will it be Studio or Atelier or something else?

We are in the process of migrating over from Studio to Atelier as it is not possible to manage without source control any more (unfortunately source control hooks in Studio simply don't cut it!). If Atelier is not going to be around in the future then it will be good to know now so we don't waste our time and energy learning it.

Please clarify?

Thanks John. Appreciate your team's effort in getting us to this stage. We will upgrade and start testing it out. 

LUT and HL7 need to be on the roadmap at some point. Our code is currently fragmented and not entirely committed to GIT. Changes to these files need to be manually tracked which defeats the purpose of having a source control system.

Thanks again. 

In the first place I'll have to say that I never understood the choice for Eclipse as the base for an ISC editor/ide.

So I'm not that sad, for this decision.

But what will we get then ?

I think ISC is big enough to have their own IDE (do we need an IDE ?)

It could be home-grown (a little bit too late) or Atom based, which is a editor not an IDE.

If  ISC abandons BPL (which I think is really a very bad thing), it would make a transition much easier.

Thanks for all your questions and comments.    A bit more of an explanation:

 

    We’re reviewing our plans beyond Atelier 1.3 in order to radically improve the developer experience with InterSystems IRIS Data Platform. 

    We welcome your ideas and feedback, and invite private communication to ensure that our plans are sound.  

 

    In the meantime:

·        If you are using Studio and happy with it, keep using it.   We are maintaining it indefinitely and are happy that there are so many active developers and that the community has expert advice as well as add-on tools that work with Studio.

·        If you are using Atelier and happy with it, keep using it, and look forward to version 1.3.   We are also maintaining it indefinitely.  We've reached a milestone and have many users that are productive and effective with Atelier, including combining it with other Eclipse plug-ins.   

·        If you are starting work with a new development group, use Atelier.   Eclipse is familiar to millions of developers, and Atelier is more than sufficient to get the job done.  Consult with us on these projects as you get going, because there are some areas such as Ensemble productions where developers may want to use Studio or other tools along with Atelier.

I make every effort to take a passive role here on the community but I have to make a statement here. What you are saying is that regardless you are leaving people with a broken experience. If Atelier does not get the added editors needed to support HealthShare development fully with integrated source then its appeal is greatly diminished as you have large pieces of work that miss out on integrated source control. But you can't go back to Studio either as it leaves you with no supported plugins for ANY source control system at all and means you have to make that implementation yourself via Studio Hooks. Neither of these situations is viable unless your intent is to leave people with no forward path for managing code and developing on ISC technology using any modern professional development approach.

In what way you where involved in the choice for Eclipse I don't know.

But in my (not so) HO, that was a very bad choice.  Eclipse is a jack-of-all-trades that serves no-one.

If ISC would have made a choice for Git as a SC a long time ago (conflicting probably with long-time parties such as GJS) , Studio could have been vialble. Only on Windows. But I think the developer wars on Win/Lin/Mac are long time settled on VM's and containers.

Building on Studio further, opening up API's (Atelier), supporting Language Server Protocol , ISC could have been less dependent on others.