System -> Advanced : Miscellaneous -> Gateway Monitoring confusing

awebster

For option "State Killing on Gateway Failure", the Checkbox, default checked, states "Flush all states when a gateway goes down"
The description reads:
The monitoring process will flush all states when a gateway goes down if this box is not checked. Check this box to disable this behavior.

So if the box IS checked (default), are states are flushed, or not when a monitored gateway goes down? This is confusing.
If there are several gateways being monitored, and the states are flushed, does this affect all gateways?

Suggested description:
The monitoring process will flush all states when a gateway goes down if this box is checked. Uncheck this box to disable this behavior.

doktornotor

Perhaps 2.3 would be good time to refactor all those WTF "tick this to disable that" checkboxes to enable/disable select.

State Killing on Gateway Failure - Enabled/Disabled - about 10000% better.

jahonix

@doktornotor:

State Killing on Gateway Failure - Enabled/Disabled - about 10000% better.

Count me in
Usability in a GUI isn't as trivial as some think. And once your mind is stuck to a certain pattern you don't see the forest for the trees anymore.

hda

Yes, in graphic programming use "radio-buttons" (only 1 option valid) i.s.o. a "check-box" (withalotoftext) :)

awebster

I agree, the language should be clear: "check to enable".
There are numerous instances where the text is confusing because of a negation (not) or an exception (unless).

Keeping in mind that pfSense is used all over the world, and while a multilingual GUI would be cool, requires a huge translation effort; to date only Turkish and Portuguese (Brazil) appear as choices. In the interim, simple direct english would be an asset, and it would make the translation effort that much easier for the people tackling the job.

I had the opportunity to read the book Globish by Robert McCrum, describing the work done by Jean-Paul Nerriere of IBM in developing a simple English. I would think that a product aimed at a world market might want to embrace some of those ideas.
More here: https://en.wikipedia.org/wiki/Globish_(Nerriere)

LFCavalcanti

@awebster:

I agree, the language should be clear: "check to enable".
There are numerous instances where the text is confusing because of a negation (not) or an exception (unless).

Yeah, I think part of the problem is that the guys coding don't have the same POV as the users, not blaming them, just a fact.

What I see in pfSense is some logic flaws in the text, the description fits the effect on the code behind the UI, but not what the user understands.

@awebster:

Keeping in mind that pfSense is used all over the world, and while a multilingual GUI would be cool, requires a huge translation effort; to date only Turkish and Portuguese (Brazil) appear as choices. In the interim, simple direct english would be an asset, and it would make the translation effort that much easier for the people tackling the job.

The other part of the problems comes from how the translation is done. You have to install Github, fork the main branch, clone to your local disk, usr xgetext or in my case Poedit to correctly edit, generate the translation and compile it, then commit, do a pull request, sign the CLA. Some people might want to do the translation, but don't want to go through all these steps.

pfSense doesn't have a huge amount of text in it's UI, the PTBR translation I'm revising now will take probably a week to do… alone.