I don't think you need NASA to solve that...
Just to clarify:
Us can't move forward without evolving tools and resources.
My criticism concerns the methods used to force uncomplete/undebugged/untested/unnatural things.
Like rushing the BE model to market before the software is ready and tested...
Do you see what I mean?
I really dislike forced obsolescence. Believe me. But sometimes you have to recognize when asking for something is no longer realistic, and then move on.
See my previous post...
I know.
So, as you can see, I am not against WB4,
but rather against the methods used both to develop an unsightly, non-ergonomic GUI and to force us to use the new version,
which isn't yet ready in terms of feature parity and usability.
As the saying goes, the better is the enemy of the good. But until the better shows up some day, MikroTik decided to get rid of the good first. As said: all my hope goes into WB4 development. Give feedback, help making WB4 great.
I guess we need a specialist in Riga. 
Then there are things like this that make everything clear and precise: they don't test what they compile.
You are exaggerating and this is totally unnecessary and untrue... Go and do something productive instead..
So, what you wrote yourself is untrue and unnecessary? Are you reporting things that are fake and incorrect?
Just give me the source code.
Perfection is economically almost impossible to achieve. Costs too much.
There is pareto principle. But WB4 is not even touching 80%.
I could state, with almost absolute certainty:
80% of RouterBOARDs worldwide have SSH enabled, and only 20% either have the firewall correctly configured or have updated RouterOS.
80% of RouterBOARDs are still running a version of RouterOS other than v7, while 20% are running v7.
No. I'm mentally blind
Was there a Winbox 2? A Winbox 1?
What was it like to let go of those guys?
So that BMW (and others who think the same shitty way) can create products like:
- Pay a monthly fee to make the heated seats work.
- Pay a monthly fee to lower the suspension below a certain level.
i remember wb2 , wb1 not so much
As far as i remember, the wb2-wb3 transition related to the way the backend fetched (and trusted) the DLL descriptors from the managed device.
there was no change in the design language, resource footprint.. generally in keeping with Torvald's principle (We do not break userspace)
it was mostly transparent to the users
Yeah... exaggerated and untrue...
@MatiasMK88 You see, this proves that they don't test things...
Giving it a thumbs-down (I'm refering about this case, not the others) doesn't change the fact that they don't test their own peripherals enough...