To view parent comment, click here.
To read all comments associated with this story, please click here.
Roadmap != deadline
Wow, I completely missed this comment earlier.
This is *exactly* how I feel. What Haiku needs is a "well defined" goal, regardless of dates.
Right now everyone is just plugging away to complete R1 without knowing what's really left to complete. There's just a loose "it will be a replacement for BeOS R5" goal that is repeated over and over.
There are some milestones setup in Trac, but I think they're not complete. Trac also is not a very good way to publish your roadmap.
What needs to be done is define the areas and or milestones required for an R1 alpha/beta/final and publish them. Without that, everyone will keep asking the questions over and over.
Also with milestones, newcomers know what the priorities are, and where they can help first.
Finally, as soon as you feel you can set deadlines confidently, then they should be set (with some room for error). This increases confidence in the team. It increases confidence in themselves. Setting and achieving goals (even if they're smaller ones) is what drives people forward.





Member since:
2005-10-17
Roadmap != deadline
The reluctance to have a roadmap may have its roots in the early (unrealistic) predictions that Haiku would have a beta in one year or two (which obviously never materialized). This is understandable, but there is no need to go from one extreme to the other.
A roadmap does not have to be a deadline set in stone, but rather a set of milestones (for alpha, beta, R1 and so on) with flexible time frames (instead of dates; for example, 2008 Q1) subject to change depending on how development progress. This would not only show those watching the project what to expect down the line, but it would also avoid having to answer ad nauseam the "when will R1 be released?" question.