@Stephan Granted and agreed!
But that does not explain why it seems that nobody has really used this obviously brilliant Orca approach! In my view, this Orca approach seems by far 'better' in many respects than some pseudo-solutions that expect us to program in the browser (= just absurd) and which neglect the gigantic advantages of our Smalltalk IDEs over the JavaScript world. Any Smalltalker who has ever tried to develop a bigger piece of JavsScript code (like I did) must have been disgusted by the stone-age status of the available tools and should welcome the availability to develop client-server solution entirely in Smalltalk. And today, with the availability of WebSockets, there should even be far better ways of having an Orca-based client communicate with a Smalltalk driven server on a message-passing level. It seems that WebSockets have not been used in Orca in 2011 and before, but it should not be problem to add them. Further, I see TIRADE by Göran Krampe http://goran.krampe.se/category/tirade/ as another useful addition. But most importantly, Orca should be the basis for a much better alternative to this (in my view) insane, ugly and very slow Seaside. Orca is the perfect tool to create a modern and entirely browser-based user interface with ONE single source code for browser AND desktop based Smalltalk driven applications. This would overcome our (Smalltalk in general) greatest deficiency and that has always been the user-interface, which is the by far most decisive success factor for every application software today. Look at the *Smalltalk UI status*, which for me is still nothing but a *tragedy*: Desk-top only UI definitions exist in *VA and Dolphin* where Dolphin is at least close to what most users consider and expect as the standard and that is, if we like it or not, Windows. The same is true for VW where the *VisualWorks UI* is internally totally insane, undocumented, old-fashioned in many aspects, not multi-lingual at all (despite their claims), 'polling', it’s simply “kaputt” from the very beginning. *Squeak’s UI* is out of any discussion and *Pharo‘s UI* is somewhat more modern but miles away from what end-users expect and tolerate, not to mention what they would love as an application UI. Having separate UI code for the desk-top and the browser is a sick idea anyway and therefore NO Smalltalk today is really suitable for developing modern, end-user friendly, simply “sexy” user interfaces. I have always been convinced that the total absence of a 'good UI' in Smalltalk for desk-top and browser has been *the major reason for Smalltalk's failure* to attract a large and prefessional (developers of wide-spread standard software) user-base, apart from the greed [Goldberg] and arrogance towards the UI and the absurd licensing conditions and price-wishes of the early managers not only at ParcPlace but also at their successors (some of their licencing is rather slavery). Shouldn't we finally change this sad situation? -- View this message in context: http://forum.world.st/Anybody-using-Orca-Smalltalk-to-JavaScript-tp4960519p4960668.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.
