LabVOOP and mild flame wars

[Val Brown]

Right OOP's not in RT -- yet -- and there are at least three different versions that you use (kind of like McCain's multiple houses I guess) and, yes, I understand that LV-OOP will "probably" handle "almost everything" (it's the "probably" and "almost" that bother me) and it will also prepare you for other flavors in the future (back to McCain's houses again), and, yes, it's very clear that OOP is a paradigm that doesn't fit for every use case or every application (and for some it's a particularly awful "force fit" but somehow these issues seem to be downplayed a bit in the discussions I've seen) and, if you haven't seen or heard the "OOPs the best" notes then, hmmm, your hearing aid has been turned down, and yes I will specifically report problematic posts whether from God or mere mortals.

And no I don't ONLY use code that has been officially released from NI or that I write myself. But, when I have used OTHER kinds of LV-based code, it has frequentlyled to some form of long-term problem, limitation, etc.

A good case in point for me is one of the implementations of blowfish encryption that was released for LV a number of years ago. Based on zlib.dll, it used CINs (a great advance at the time) but unfortunately that creates a number of ongoing issues, esp in terms of project management and such. One assumes this is at least one small part of how CINs have become, well de-emphasized shall we say, over the years. However, that implemenation was the only option available at the time, rather robust and full-featured; moreover, it was presented as if it would "probably" address "almost everything" related to encryption that you'll need, etc, etc. And now it hasn't been updated (so far as I can tell) for quite some time and, at some point, I'm going to have to switch off of it completely -- because of the CINs among other reasons. And BTW I was VERY happy with it for quite a long time -- thanks to whoever it was that created and released it! IMO it was yet again another thing that NI should have included in some form within LV.

So yes I know that OOP is a tool and has nothing to do with being a "real" programmer or not. Having done some form of programming since the advent of PCs -- and even before that with PL/1, FORTRAN, etc -- I have no concerns about not being "real" -- nor about not being a programmer -- let alone about being a "real programmer".

And, yes, I do talk with my wallet. I stopped paying for the Mac version right after SPT was released Windows only. Now that Mac is "coming back" isn't it about time to just recompile the code for SPT, etc and release those toolkits to a far wider market??? Or, wait a minute, is this an "OOPs" on my part because maybe I'm forgetting that even C/C++ may not migrate cross platform...

And FWIW I've had customers who wanted Mac versions from the very beginning of my work with LV. That's ten years and counting. I've also had others who've wanted Unix versions as well. No requests for Sun yet though.

And, yes, I know about http://www.info-labview.org/ but stopped browsing it because I really (also) don't want to hear more renditions/versions of "Mac vs Windows". I had enough of those kinds of discussions back in the Pascal vs C days. La plus ca change...