Submitted by Armando on 2010/03/21 15:49
 
IQ is great, we all know it here. But I sometimes wish I could use it more as a user and less as a beta tester (2.5 years).
 
Suggestion : could at least one  stable IQ version 1- without any showstoppers  (all the serious bugs squashed -- bugs with no practical workarounds) 2- and no new features added... be offered to beta testers ?
 
(bugs with workarounds taking minimum 15-20 min a day are absolutely not practical... E.g. : row calculations problems) 
 
This sounds "fair" to me -- if I may say so. Comments anyone ?
 
 
Pierre : thanks for considering this important matter.
 
 
================
 
(optional explanations, justifications, etc.)
 
I now have so much stuff in my IQ base that it makes me nervous to never know whether my DB will be fully usable or not the next day (without having to use super time consuming workarounds). (Yes, most urgent bugs are fixed rapidly but this is not always the case.)
 
If every time I encountered important bugs I could've reversed to the previous fully working version, it wouldn't have been a problem... But, in all fairness, it's been impossible for the last few (many) months : previous versions often had other serious bugs I was not ready to face once more (it was/is often like choosing between loosing my left hand, my right hand, an eye or my a few toes)... So I continuously upgraded, and continuously faced new bugs. The result was/is an real noticeable (to put it mildly) decrease in productivity. Yes, I know that this is part of the beta game -- IQ is beta and so it's buggy, etc. But... I've been beta testing for more than 2 years, and I now feel that I could maybe use a version with all "current" features (not any new ones) fully usable -- I don't mind cosmetic bugs and other small ones, and missing features here and there... As long as I'm not spending hours a week trying to work around these.
 
 
Yes... as I said, I know bugs are a normal aspect of software development. But I find that users (especially long time users who probably invested 100s of hours in debugging) should now be able choose whether they want to take further risks or get some work done for a while.

Comments

That sounds very fair.
 
 I guess we'd use our really important file in the stable version, and either a not-so-important file, else a copy of the important file, for use in the latest version for the beta testing.
 
I'm hoping that when 0.9.25 gets here, that progress towards 1.0 and stability will be much quicker.
It would be nice to get a few more clues as to when all calendar stuff is going to come together!
 
 
 

I hear your suggestion and do my best to fix bugs bugs as soon as they're reported.
 
I'll fix all reported critical bugs and then split into 2 branches: stable and dev.