Re: [ve-dev] VE for Eclipse version higher than 3.2

[yves.yang@xxxxxxxxxxx]

We haven't noticed the Eclipse crash by VE. Someone else has this problem ?

Regarding the architecture, It seems to me the VM solution is very general
and openness.

Regards
yves
----
>From my point of view there is more than one bug to fix. Very often it
occurse, that VE cannot render a GUI and give for all components the
message "a previous error prevent this component from showing". My
experience is, that if a GUI is more complex, than this failure happens
more often. Some time one can fix that with complete recompile the entire
project or a complete restart of eclipse can help and some times I have
no chance to fix that.


In every cases VE setup runs until "75%" and stops there (for about 10s)
and than VE give up and shows only placeholders for GUI components.


This behaviour can be observed under all operation systems (Windows XP,
Ubuntu, Mac OS X).


Not though often occurse (pehapse one time per week), that VE and Eclipse
crashs without any message and Eclipse are complete removed from memory
(no tasks are running anymore).


I admire the hard work (because the huge complexity) on VE but at the
moment VE is fare away from beeing ready for productive use.


I think the last mentioned error is originated from the usage of remote
VMs. I am not shure if that design is ready to survive and all other GUI
builders don't use that design (Jigloo not, Mantissa not, JBuilder not).


Wy VE developers has increased the complexity.


Jan