BreakPoint
The birth of graphical programming.
You can read his 1966 thesis here. (Claude Shannon was his advisor).
According to his wikipedia entry, he went on to do interesting things. 😉
Wow, so many concepts! Had we good graphical computers then who knows! Of course Jeff Kodosky and Ja,es Truchard would have been college students too when the thesis was written.
Well, unlike LV, he seemed to have a zoom feature 😉 .
As near as I can tell, this is the method he used in the code (yes, I know that it doesn't play well with non-squares, but he did say he used only square integers and I didn't feel like finding a better stop condition):
You can find a basic description of the algorithm at the top of this page - http://mathforum.org/dr.math/faq/faq.sqrt.by.hand.html
Also, sorry about the quality of the image, but I didn't feel like working out how to get several frames and play with them so that their white parts would accumulate and then we could see it nicely.
Having read part of the thesis, I was motivated enough to clear it up a bit. It's still not readable (the text on the left, for instance, is "SQROOT TEST DEFN"), but it's better than it was.
I saw Jeff K's presentation on his evolutionary ideas for LabVIEW at the CLA Summit in France on Monday - one of his concepts for creating block diagram components that represented your electronics circuit was very interesting, but canned about 5 years ago. Made the block diagram look a little like this circuit diagram 
@Hornless.Rhino wrote:
That gives me an idea..... that would get me sued into oblivion.
That's a pretty interesting point... I mean, G is a programming language. I don't understand why it should be considered an infringement of copyright or patents if another application takes the conventions established in LabVIEW and reuses them. I mean, a high level While Loop looks the same in any text language, right?
I suppose the reason I mention this is because over the past year I've been working on a project to create a Java-based G-based programming environment that could be used to generate abstract text code (The example language was Verilog) as part of some individual project we're supposed to work on within university. Originally I was going to do this from within LabVIEW and use VI Scripting to parse the Block Diagram but I ran into some problems when developing a clean architecture that was flexible enough to perform different actions based on a specific GObject instance it was dealing with, so although there was a huge overhead in developing the graphical environment from scratch, I figured it'd lend itself to more specific customisation later on. I ended up learning a great deal too.
The project has been a success but I've talked to some of the field guys from NI and we decided that it'd be unsuitable for me to put it online. Originally I wanted it to be open source because it would serve as a cute attempt to understanding how LabVIEW works; but it's understandable as to why I can't do that. It must look really strange to them considered I worked with them as an intern for a year... I have to confess that the whole time I spent in technical support, I was really longing to see what LabVIEW was doing in the background and never had the chance to. This project truly felt like my only real chance.
The way it worked was so that rather than just Verilog, you could pretty much override the conversion stage at any level and use any text you want with it. Because it had this format, things like the re-entrancy would need to be explicitly coded from within the diagram. There was no Front Panel either. Any newly developed graphical code would have a textual equivalent for any supported language... At the moment, for the few graphical blocks and structures I have, there's a C and Verilog equivalent. It was never my intent to cause any harm to LabVIEW, because I love it. It's just that outside of these forums, any mention of LabVIEW online is quickly followed by complaints about the pricing, and I wanted to be able to give these people a chance at seeing why it's so good. Giving people the chance to play around with G in their own time, not on someone else's.
"I can open up Notepad and write any text code I want in there. Why can't I open up Microsoft Paint and do the same with LabVIEW?".
I read that comment on YouTube somewhere. It really made me laugh.
I firmly believed that the project could have been that foothold for those users who have yet to discover the benefits of graphical programming but don't have the budget. You'd get more people using G, and more people having success with LabVIEW.
People on the forums mention a lot about how they wish it wasn't closed source and I thought it'd be a great example to toy with. What I really started to enjoy was the fact that I could start putting my own little tweaks into the application that weren't present in LabVIEW; it could grown into a little project for users to benchmark their own changes to the IDE. I had things like only being able to make a wire connection between compatible terminals (Which I'd then build on so that it would ghost a potential bridge between initially uncompatible terminals, like a build array between a scalar value and an array input) and tried to get a head start on implementing Altenbach's Synchronizer Bar.
Anyway, it was a nice run. I thought I'd share this little endeavour with you to see what you think. I know that NI are a great company that really embrace the outer communities but I have to admit that I was a little heartbroken when my pet project didn't get the same kind of response, although in hindsight it was plain silly of me to think otherwise.
