On Jun 9, 2013, at 6:59 PM, "phil@highoctane.be" <phil@highoctane.be> wrote:
> As a end user, it gives the impression that the system is shaky.
>
> Now, this is written out of frustration: I spent while documenting a set of classes only to get the thing to a VM crash when posting to monticello and followed by the inability to get the work back without copy/pasting stuff all over again since the changes files wouldn't work (and is full of <historical> entries).
>
The <historical> is *just* that the who changed the class comment when is lost. Nothing else. This is not important, no other language records that information.
> Well, that's what it has to do with each other: the workflow is broken and it demotivated me from documenting classes again.
The information lost due to the <historical> thing is normally not even shown in the browser, and I honestly do not see at all why this is relevant.
The issue with looking at the .changes after a vm crash is completely different, and should be looked at, as should be the other 500 issues on the issue tracker.
> Preventing contribution is not good. Especially given the boatload of undocumented items.
But mixing up issues just makes everyone confused.
� � � � Marcus