Laccio Framework

< .. snip .. > Hi Alessandro: I downloaded your first sample and will take a look at it. I'm sure that I will have some questions to ask, and I will post them to this ML. BTW: I'm referring to a very proprietory organization - so I'm not at liberty to say *who* and must leave out some details, but here's their scenario: Many - perhaps hundreds - of small rebol scripts running and being managed project-wise via TCP/IP. Including message-passing. Different from my way of doing things, but very rebol-ish in fact, I think. I manage large projects by "loading" or "importing" modules via 'do, whereas they are using TCP/IP. I don't know if they use rugby or if rugby would be helpful. tim

-- --Alessandro

Hello, there is no restriction using Laccio Framework and tcp/ip or other protocols. Infact, Laccio is a single ".R" file, that you can directly download and import using something like:

Furthermore, the components could be created using a single ".R" file or more files, that you can download and use in your environment. About message passing, it is not used to exchange information, since Rebol already supports several protocols and method to efficently solve the scope. WHen I talk about "message", I'm talking about events. I named them messages, since a message is a signal that a method can send, and if your procedure is registered to receive such event, it will grab it. And consider that, usually when you use the typical event dispatcher, you can associate only a method to a single event. Example: if you create a button, you can associate only a function to its click; instead, if you use Laccio, you can associate as many "listeners" as you want to a single event (message). I downloaded your first sample and will take a look at it.

You are welcome! Please don't hesitate to make me any question! On 8/22/07, Tim Johnson <tim-johnsons-web.com> wrote:

-- --Alessandro

-- --Alessandro

-- //Alessandro http://sguish.wordpress.com http://laccio.wordpress.com

Hi Alessandro.. examples examples examples >> oh yes please.. very nice - like Flash Actionscript without all the ECMA syntax junk ;-) Perhaps you can put presentation with examples into an EasyVID active-book format [one of Rebols undedeveloped jewels imo] This is just what I need for my next project -- which uses both flash and rebol. Ah how sweet to have a reasonably consistent implementation of interface. btw, do you know MaxMSP or PD PureData ?? prgrammable patchbay metaphor software for musicians and others... MaxMSP is the parent system - color, very sexy, sophisticated, powerful. [$] PD Pure Data is the opensource free child of Max, spawned by one of the original authors [Miller Puckette] Anyway, among many great ideas they have neat syntax for ketting patches and things to communicate send and receive -- its so sweet and simple -- no 'lucid' like the best parts of REBOL. check it out.. http://en.wikipedia.org/wiki/Pure_Data see the main docs and search for "send" -- generally read intro and read down from here http://crca.ucsd.edu/~msp/Pd_documentation/x2.htm#s1.3 thanks Jason

-- //Alessandro http://sguish.wordpress.com http://laccio.wordpress.com

Yes.. 1) I believe PD PureData offers a superb interface metaphor/paradigm for your components and for Rebol3. 2) PD has an open source API. That means *someones* [not me alas] could write a Rebol interface to PD. Then all sorts of cool modeling could happen very quickly, including hopefully engaging a larger very creative community of programmer-musicians-experimental innovators. One could also join the rebol parts via PD to lots of others... yay! Ruby Python and other languages have been used to write PD patches. In general this is because people want to do something in PD and use the language they know and love. However, it can also work the other way around -- I'd call it "host system and interface symbiosis" ... The Wiki for PD API is http://www.puredata.org/docs/developer/PdAPI and one example using that, which I just found googling is http://www.le-son666.com/software/pdj/ PDJ enables you to write java code to interact with pure-data objects. another ... mxdublin a Java/Python framework that can be used to sequence generic events within pure-data or max. http://le-son666.com/software/mxdublin/ uses Jython = Python implementation written in Java << ouch wow cool 3) The design philosophy is very familiar--- PD REBOL Laccio fertile ground I think. 4) I also imagine/hope that a PD-inspired interface could be written in 100% rebol -- actually R3 [rebol3] probably best. That would mean a light, fast, 100% rebol graphic interactive user interface toolkit could support inter-component inter-system communications and instrument building. beautifully so :-) -- many advantages including opening access to non-rebol programmers USE of such tools. 5) You might find this is perfect -- or too much, but inspiring none the less I hope. 6) A delightful design feature of PD is that PD files ["patches" they are called] are just simple TEXT files - interpreted by the PD environment as dynamic interface. You have to install and play with it to understand what I am talking about really. It's a rich universe and thriving community. But here's an example from the interactive user Manual when you install the software. 'patch' = a file called 04.messages.pd If you open that in PD it looks like this screen shot [but it works] If you open in a text editor it look just like this: #N canvas 0 0 595 446 12; #X floatatom 225 110 0 0 0 0 - - -; #X floatatom 184 109 0 0 0 0 - - -; #X msg 184 56 5 6; #X floatatom 132 108 0 0 0 0 - - -; #X floatatom 64 105 0 0 0 0 - - -; #X text 30 21 Most Pd messages are just numbers or short lists of numbers: ; #X msg 64 55 5; #X obj 64 80 + 9; #X obj 132 83 +; #X obj 184 84 unpack; #X msg 288 55 5; #X obj 288 107 print; #X obj 288 81 pack 34 78; #X msg 132 55 5 6; #X floatatom 195 328 0 0 0 0 - - -; #X obj 195 303 +; #X msg 195 254 1.2 3.4; #X msg 205 277 5 6; #X text 36 206 Unlike Max \, in Pd all numbers are floating point. Numbers whose values happen to be integers are displayed without decimal points.; #X text 31 363 For more on messages \, get help on any message box by right-clicking.; #X text 329 409 updated for Pd release 0.33; #X text 34 149 If you send a list to an object with more than one inlet \, the items in the list are spread out over the inlets \, as seen in the 5+6 example above.; #X connect 2 0 9 0; #X connect 6 0 7 0; #X connect 7 0 4 0; #X connect 8 0 3 0; #X connect 9 0 1 0; #X connect 9 1 0 0; #X connect 10 0 12 0; #X connect 12 0 11 0; #X connect 13 0 8 0; #X connect 15 0 14 0; #X connect 16 0 15 0; #X connect 17 0 15 0; hope this makes sense regards Jason

-- //Alessandro http://sguish.wordpress.com http://laccio.wordpress.com

... and here is a url to show what I am talking about: http://tranzilla.net/instrumentalites/puredata/examples/ Jason

-- //Alessandro http://sguish.wordpress.com http://laccio.wordpress.com

lovely.. A fun place you might find some help, and meet some nice people is dorkbots /people doing strange things with electricity / http://dorkbot.org/ Many of them use PD I see.there is group in Milano http://dorkbot.org/dorkbotmil/ Where are you [geographically] ? stay tuned Jason

-- //Alessandro http://sguish.wordpress.com http://laccio.wordpress.com

Dorkbots is fun regular group event which started at Columbia University Computer Music Department [Douglass Repetto and others]. They do all sorts of experimental interfacing of devices, ideas, software, hardware, often musical but may be visual or other. They are characteristically non-evangelical about solutions -- try everything , love new ideas, techniques, fusions etc. PD is used by that department and most other computer music departments in universities all around the planet. My idea about using PD and Rebol/Laccio is that one could leave aside all the specifically musical features. Just focus on the question: - How to interface openly programmable instruments and components in a good way? - As it happens, music imposes a great problem domain (especially since MIDI) and its fun :-) Programmable music toolkits must address lots of things -- message passing, complex timing, event handling, interfacing virtual devices, filtering and transformations etc etc. PD scales pretty well. One work in the same interface at a high [sophisticated] level, but also dive into that to reveal all the sub component patches and modules which combine to make a whole. The main paradigm is a 'patch bay' message architecture with graphical interface. PD objects typically all have at least one input and one output. You drag the mouse to connect these. Keep adding and combining to create a useful 'patch'. Patches and objects can talk to each other uses a nifty 'send' 'receive' object. Very similar to rebol send. I look at most things as instruments -- piano, computer, telephone, knife, drill, camera, programming language, hands, ideas.... Rebol is a beautiful language to build internet instruments and components. Problem is finding versatile way to connect them up and to share the results with a wide range of people so that they can do useful, interesting things with them. By design, rebol is toolkit suitable for laboratory type experimentation. But it lacks a cool interface metaphor to build on its inner strengths. It lacks easy interactivity. It's a lot of work to build and maintain a huge set of functioning widgets in any language. So why not take good advantage of existing user interface toolkits? But not any will do. You need ones which really have a compatible design philosophy [Carl are you listening ?] PD is a great candidate. oh and here's some youtube demos of PD in action http://video.google.com/videosearch?q=PD+pure+data MaxMSP is also, but its expensive [PD is free] http://www.cycling74.com/ However, I did test out a Flash interface to Max which worked great as proof of concept http://www.nullmedium.de/dev/flashserver/ I'll let you catch up this. keep me posted Jason

A rebol gotcha - rebol ignores case on internal symbols but is trumped by the OS file system... tj