LabVIEW
LabVIEW Pros vs. Cons?
The problem is with your generalizing assumptions and circular logic:
@Synaesthete wrote:
... My conclusion is that with LabVIEW, you are allowing non-software engineers to build large-scale applications such that they are not following best practices, which represents a significant inefficiency.
- Whos is actually "you" in "... you are allowing ..."? I don't know anyone matching that pattern.
- You are stating an assumption (that the ones building large scale LabVIEW applications are non-software engineers and are not following best practices) as fact. Using that unproven fact as a premise to later prove that LabVIEW is not suitable for "true" software engineers is rather circular.
- Your entire diatribe is full of such fallacies.
Ben, I completely agree. I definitely don't believe there's a necessity to be a formally trained 'type' of engineer in order to practice in that field. An engineer is an engineer. Also, no reason you must do something in an established and approved manner in order to be efficient and effective. There is, however, a difference between academically-approved and industry-approved. MVC architecture has the approval of both.
Nevertheless, in any given project there are "electrical engineering" tasks and "software engineering" tasks. The EE and the software engineer 'could' be the same person, although for larger projects you'd want them to be different people focusing on their individual task, and applying the best practices for their practiced discipline (regardless of their degree, if they even have a degree, which also isn't always necessary IMO). Too much 'bleed-over' can have negative consequences, so if your EE discipline starts bleeding over in to the software engineering aspects of the project, you will run in to problems (this definitely happens with LabVIEW... a lot... G is not a wiring schematic! 
I'm sure everyone in this thread understands this sentiment).
So if you are working with state-of-the-art software engineering practices, the MVC architecture is very likely to be the best technique for any medium- to large-scale application.
Sorry if my posts have been at all rambling, offensive, or appear misdirected, misinformed, a nuisance, etc. I'm taking a free-form approach to this discussion and tossing all my thoughts to the wind, testing the waters. Your responses have been invaluable. In 6 months I'd rather return with a well-reasoned argument that doesn't offend. In 6 months, time permitting, I'd really rather present a tool for practicing what I preach and avoid the convoluted flame/troll-like threads altogether!
Your post so far could have had some useful information if they were not presented in such a tendencious way. From browsing through this thread I still have the idea that you mainly complain about the fact that LabVIEW allows non-programmers to write horrible programs. And that you think LabVIEW needs to have a "true" OOP system and just about anyone should be forced to use the according frameworks that this allows.
LabVIEW's strength however has been and still is that non-programmers actually can setup systems and do experiments on their own, without having to explain to a programmer first what they were wanting to do, and if halfway the experiment requirments change, having to beg this programmer to reachitecture his "perfectly" architectured application to adapt to those new requirements that may have been impossible to be foreseen.
Also the ability for non-programmers to work with LabVIEW does not prevent a programmer to write a well architectured application at all. You can use C and C++ and just about any OOP oriented environment to write at least as horrible programs then the worst ones I have seen done in LabVIEW in my 20 year LabVIEW carriere. To not do this, requires not only formal training but also discipline from the programmer and that holds true for LabVIEW as well as any other programming language out there.
Some people think easier in pictures and others prefer the textual representation. None of this is wrong or bad but simply a personal preference. For me coming from an EE ingeneering direction, LabVIEW looked immediately like the most logical aproach to computer programming when I saw it 20 years ago. Was it perfect? Of course not, just as Pascal in all its incarnations including Modula, or C as I was learning it in fact after having started with LabVIEW. I liked of these the clear and strict syntax of Pascal and especially Modula, the logical overview I could get much easier in LabVIEW, and the flexibility of C at the price of a sometimes almost incomprehensable syntax that can sometimes look like an armadillo was rolling over the keyboard. C++ in fact only made the syntax more involved and even less comprehensible without a lot of study.
Complaining about LabVIEW allowing non-programmers to write software in the first place sounds simply like trying to protect your job security as "real" programmer. It's not helpful at all and can really only backfire if it can have any effect at all. And MVC, Actor Framework and all that may have indeed a deserved place in mid-sized to large scale applications, it's by no means a guarantee that such an application will work nor a requirement to get such an application to work well.
@Synaesthete wrote:
Well with all these Knights of NI, that's kind of what I'm suggesting could happen.
Hey!
