Hi!

I have no idea whether this issue has been raised before, a quick search I performed revealed nothing. From my own experience, new RouterOS versions seem to bring too many (great) features at the expense of new bugs and the non-elimination of old bugs. Everyone likes new features, but the anxiety involved with repeatedly upgrading multiple routers to get rid of bugs, only to have to downgrade them shortly afterwards due to even more deadly bugs is sometimes too much to bear. You may well disagree, of course, hence the usefulness of this poll.

Obviously, both bug elimination and new features are important, but which one should take priority?

KS

IMHO I don't think that this is the right direction in the long long term. Focusing on fixing bugs is something that is mandatory and unquestionable. MT are already trying this. Additionally, they will never lose an edge by de-prioritizing the development of new features. So it is not a matter of simply choosing b/w fixing bugs or introducing new features.

A better direction (again IMHO) is to split the development into two main branches - "stable" and "feature". Stable releases should only contain bug fixes for previous stable releases. "Feature" releases are intended to present new features to the public and are considered unstable. Nothing new or fancy really. This is well known and MT only need to find their way into improving their development/release process.

Best regards

IMHO I don't think that this is the right direction in the long long term. Focusing on fixing bugs is something that is mandatory and unquestionable. MT are already trying this. Additionally, they will never lose an edge by de-prioritizing the development of new features. So it is not a matter of simply choosing b/w fixing bugs or introducing new features.

No it isn't. Yet resources are scarce – without exception. Whichever way you deploy them, there will always be a tradeoff between stability and novelty. I am all for novelty in a world changing at a lightning speed, but the current direction of MT seems to me plain wrong. Each changelog section contains more features than fixes, with the end result being an erosion of confidence in the RouterOS development process. I am currently managing a network of 21 routers with various 6.x versions and have been mentally fighting the prospect of a costly return to 5.x for months now. The sort of situation dominated by upgrading / downgrading cycles (Yes, v6.11 is out! Let's see if it works fine this time. Oh no, we need to go back to v6.x!) should be avoided. Priorities do matter.

It might also be an alternative, to backport bug-fixes to older versions of RoS, to the last mayor versions, at least, if possible.
As a very dumb MT-user, I only use MTs hotspot features, so I am happy with RoS 4.x, or even 3.x
And I am hesitating to do any upgrade, although there are some glitches from time to time, which MIGHT be fixed in a newer version.
But at least, I can handle these glitches now; in case of a version upgrade, I might start the learning curve again.

So a backport of bug-fixes would be appreciated.