| Re: [wtp-incubator-dev] 0.5M8 Schedule |
DOUG SATCHWELL wrote:
I'm not to sure what the criteria is, but I think there is still some key areas in the way of debugging, and just some extensibility that needs to be in place yet before we graduate. I leave specific criteria to David Williams to decide.I'd vote for that idea Jesper. I think we should now seriously start thinking about what we will need to do in order to eventually graduate.I know we have quite a lot of work to do in the areas of user docs, javadocs and unit tests (I admit I've been particularly bad at all three), but what I don't know is how this will be measured in terms of graduation? For instance, is their some automated tool Eclipse uses to find javadoc coverage before a project can graduate? Its easy to put javadocs in, but meaningful ones are more difficult!
When I compare our project with another, say xml.ui, I see that our project settings are currently tighter than need be. For example, in xml.ui there is no need to javadoc every public method. But I think we should at least document every public method of every class that is not in the 'internal' namespace?I disagree on the internal namespace javadocs. The main reason is from my own experience in trying to understand what some of the current WTP internal code does at times, because of the lack of javadoc to help give a hint. So, while I understand that other WTP projects don't have as tight a restriction on what gets javadoc, I think it helps people in the long run to have them. If you take a look at JDT, their code is very well documented and fairly easy to understand. Of course they have refactored much of the code while they have developed it as well.