On page connector for LabVIEW like Multisim

[gkartik]

Wiring the blocks feels difficult especially for large program and also the block diagram crowded with wires. I would like to have "On page connector" feature in LabVIEW like that in Multisim. I feel it to be a very important feature, please vote this idea if you feel the same.

[CMal]

For those of us unfamiliar with Multisim, can you explain what an "on page connector" is?

[gkartik]

In multisim, on page connector helps connect two terminals without any wire showing on the schematic, it connects them internally.

A similar connector in LabVIEW could help connect the blocks and make the block diagram tidy by removing wires.

[crossrulz]

This concept is discussed a little in this similar idea: http://forums.ni.com/t5/LabVIEW-Idea-Exchange/Some-wiring-ideas/idi-p/1524884

[AristosQueue (NI)]

gkartikrj:

As a member of LV R&D, I acknowledge that this feature is feasible. It might be a good or a bad thing to add to LabVIEW... it would require usability studies to decide a) how frequently diagrams have wires long enough to benefit from this on-diagram teleportation and b) whether other users can read the resulting block diagrams better or worse once the on-diagram leap is implemented.

As a LV user, I've been generally opposed to this idea (this is not the first time it has come up). It completely defeats the readability of a block diagram as far as I'm concerned. If a VI's wires are getting that long, create some meaningful subVIs for the intervening code. Personally, I have little sympathy for any code that is so badly written that this feature would be useful. Now, that's my personal take on it... I've worked on several LV features over the years that I personally wouldn't ever use or let be used on my team, which is why my R&D answer is more open-minded.

> make the block diagram tidy by removing wires.

I deny that this is possible... you can't make a diagram tidier by removing its wires. Bundling wires together? Sure. Removing redundant routings that are covered by other connections? Sure. Temporarily fading out some types of wires in order to emphasize others? Sure. But not just failing to render the wires. Anything that obscures the connections between nodes harms readability in my experience. This feature strikes me as having great similarity to local variables and functional globals -- both named locations that can store a value and read it elsewhere. Yes, this feature would imply ordering where as the others do not, but just in terms of diagram readability, you've now moved from a wire whose route you can visibly trace to a jumpy thingy that you have to use navigation menus to find its other end.

[tbasso01]

I think that the local variables do exactly this.

[AristosQueue (NI)]

tbasso01: No, they don't, not exactly. Locals have substantial overhead compared to a wire, and the feature of a "teleporting wire" would be equivalent to a wire, just not rendered. Also, local variables do not enforce an ordering of the read/write, so if you try to do this with local variables, you also need a very large sequence structure to force the write to happen before the read. Otherwise you'll sometimes get uninitialized values when the read happens to execute first.

[tbasso01]

AristosQueue:

Yes, I gotcha.

Thank you for your comment.

[Darren]

Any idea that has not received any kudos within a year after posting will be automatically declined.