Hi Thomas,

the Recital engine, which has been developed over 24 years now, is the basis for the language engine. It is already used by banks and defense industries on both sides of the "pond." IOW, it already has Enterprise acceptance, at the core of the Enterprises in question.

Any language can call any other language's function through an intermediate call, and objects are respected. This feature was added a couple days after calls for it were made on the forum, which suggests to me integration at a pretty deep level using a meta-language definition common to all languages.

The docs are, as you indicate, largely non-existent, at least in the hands of users. Although they recently added documentation of the UI classes to the Beta9R4 roadmap (in development). I don't think objects can be exposed to those not part of Lianja. Languages can be added quite easily, it appears. That will be an interesting discussion.

I, too, have to believe there are a lot of edge cases that will give trouble. What reassures me is the speed with which things gets fixed: it is measured in days (typically 1 day), not weeks or months. It's an iterative development style that I think produces a better product in less time. It is, of course, dependent on users providing feedback: the forum has been really important in this regard.

All told, both the goal and approach being taken work for me.

Hank

>>Script languages are already highly valued in Enterprises, everywhere -- not by IT, perhaps :),
>
>Yes, I should have qualified: if allowed by IT, they prefer them running on the selected VM:
>Rhino, JRuby, Jython vs. Ironpython, IronRuby and the offerings of MS for Ecmascript.
>
>>but by the workers who make Access apps, excel worksheets that mimic apps, and heaven help us, Word Tables, in order to get their work done. VFP has been a sane alternative in the past; and I think Lianja will be a sane alternative in the future.
>
>The uphill battle of a new runtime in IT is the larger stumbling block . I should have phrased it better ;-)
>Will be interesting to view an adoption pattern of former vfp users by dev count.
>
>On Lianja - I have not checked the last releases, even though now most of the things
>missing last year are in place, if docs are correct.
>How successful was porting one of the simple Promatrix-based apps ?
>
>Is there some documentation describing the interleave of different building blocks of Lianja now ?
>Did they "just" embed a CPython engine, a JS engine, Webview and so on or take a similar "add-on" route ?
>Or create their own parser building something like a common AST delivering a common runtime/variable space ?
>How do they handle the differing rules for variable scope/parameters/return values in mixed mode programming ?
>
>Or is it safer to always use their global app model to make objects visible to other source languages ?
>Will there be difficulty passing such a foreign object as parameter ?
>What happens if you subclass a class/JS-prototype from another language and
>plaster on niceties like Python properties or vfp assign-methods ?
>
>While I applaude the drive to finish the product, having an open BEta without even prelim docs
>on such structure leaves me just trying to run existing code and micro-bechmarks or
>shoot in the dark for edge cases I could otherwise envision at least half of the time.
>
>It seems the best dark horse candidate (of Polarfox, Win/WebRAD, some basics or Delphi/Lazarus): if that is praise.
>
>regards
>
>thomas