Frank's Emails	Strip links 2
		Strip links 3 Strip links 4
Hi John, >> I don't know if you like this idea nor if it resembles your >> webneuron, webaxon and webstore model (which I don't understand): >> >> Lets strip the links from the webneurons. >> >> Why? the links form the flow-control of a 'program', my feeling >> is that this has to be coded seperatly. >> >> What do you think of this? > I can understand your unease with the links controlling the logic flow, > because this is such a different approach from the structured programming > we have all learnt. I have no problem with links controlling flow, i have a problem with including them within a weblet. > Your idea means we will still have this sort of thing:- > WHILE x<10 > IF y>4 OR p=t > gosub random(x,y,p) > ELSE [etc] Now this is just normal programming, I want to seperate the flow controle from operators. > Having programmed in this way for many years in many languages I > intuitively feel that there has to be a better way. Off cause there is a better way (there always will be) > The brain does not operate in this fashion, it is a method devised > without referencing the way the brain works. I metioned before I don't want to model a brain. (when flow controle is stripped you still can model a brain, not only this you can model many brains!) > We have to carefully think what properties a more correct programming > method should have. I'll start: - all properties of OOP - plus: a stronger seperation of data/flow controle/operators - plus: self-similarity > I can't see how you can disagree that "mappability" must be one of them. > Structured programming is not mappable, weblets are. I think structured or oop programs are mappable too, but what I want is not drawing maps (the computer should do this) but draw routes (=flow controle) on the maps. > I really don't know what to say to convince you except what I have > already said in previous Emails. I don't think you will be convinced > except by trial and error on a live weblet system. By the way I have > never had a fully live system yet, in any form. I think the rough idea should be clear and then it gets polished by trail and error > If you look at the way the website works, with mirrrored back links for > every forward link, it is easy to find your way about, and still will be > even when the website gets larger and larger. This is because it is > mappable. One more point about mappability: The question, 'is something mappable?' depends on the position of the observer and the information he/she has, in other words is relative. our perspective it's not mappable (maybe it will in the future when technology improves) from the perspective of a (intellegent) bacteria the brain is mappable if it lives long enough. X-UIDL: 836173871.001 > Now if we actually PROGRAM using the same linking principles then I think > you may agree that it will still be very easy to find your way around and > hence to EDIT, TRACE, RUN and DEBUG programs - the four big ones. If we > can produce a "zoomable" graphical plan of the whole labyrinth then > things could be even easier. I don't know how to do this yet. for this zoomability we need self similarity. > If you still don't agree then please try to address the points I > mention above, as I would like to discuss your ideas on each one > extensively (especially mappability). I hope my idea's are clearer to you know. Stripping links from hups doesn't mean you lose the ability to map, it means the map information is completely seperated from the operators. Secondly it is possible to draw a lot of maps using the same hups. > If eventually you still don't agree then I will do a deal with you. Trust > me, and try link logic flow first, and I will suffer to learn Linux for > weblet development (I think I have more to lose). > A further point to add is that we could BOTH have our ways and a winner > may emerge. I hope you see that my idea means we have both our ways. > i.e. in the webneuron script we can build in as much logic > flow as we like. We can include WHILE, FOR, GOSUB, - whatever you like. > For a particular application you may then just have one massive webneuron > with no links, just loads of tangled script if that's what you like.. I don't like the idea of applications. > It may help my case to send you a pictorial of webneuron type structures > implemented in assembler. I designed most of this several years ago, but > it is remarkably simple. Only a few structures, functions, small kernel, > and a look-up table are required to get the ball rolling. Even if we > don't use assembler, the assmebler implementation may be good for > theorising about how the system should be built. I will try and send it > soon but I am making some adjustments to fit in URL's to the look-up > table. It might help me understanding your view better. > By the way Frank, thank you very much for continuing with me. I believe > our chances of success are increased by mutual support. The discussion with you helpt to clear up my own idea's on programming. And at the moment I see possiblilities to join concepts. As I metioned before my idea's about programming is a more biological way. If you are interested in the logic of seperating flow controle from operators, the way program starts to look like (actually a more powerful way than) DNA coding. Please say so, I'll try to explain it. Regards, Frank ____________________________________________________________ Drs. F.P. Schuurmans CYBER PUBLISHERS frank@bio.vu.nl Amsterdam, The Netherlands