Anomaly on Wiki4D GuiLibraries page

[John Reimer]

Hello jcc7,


> Other issues (such as whether DWT should be the standard, whether it
> should be noted so prominently on the wiki page, and whether there's
> any brainwashing occurring) are still up for debate.
> 


Our adoption of the original dwt newsgroup and project space was a move of 
convenience and simplification... and perhaps may have been perceived as 
somewhat opportunistic.   That's something I can't completely deny.  I've 
contributed a fair bit of time to management and organization (and source 
contribution) of several projects on dsource for the last 5 or so years. 
 Cleaning up the stagnating old dwt, and freshening the project page with 
the new seemed like a very natural move. 


Also, please understand that Walter was very encouraging of both the old 
and new DWT projects even though he had hoped to see a Phobos version for 
the new.  He appears to be well aware of the situation as far as I know and 
has communicated with Frank on a few occasions.  As far as I know, he hasn't 
let on either way about it's "standardness".


It is possible that the situation may be awkward for him because of the risk 
of discouraging something that showcases D (since it remains possibly the 
largest D project yet), despite its use of Tango.  I don't like the idea 
of putting him in such an awkward situation, so I very much hope he doesn't 
feel manipulated into supporting it as a "standard" gui.  I think it can 
showcase D quite well without that title.


However, I think that the complexity of the situation is inevitable because 
it originates from the Tango/Phobos issue.  For us, a Phobos port just isn't 
so simple because we've made use of many Tango conveniences that Phobos just 
doesn't provide.  Also we didn't want to work at maintaining two core libraries 
in DWT, a project that, in itself, requires a fair bit of work to update, 
debug, and maintain on multiple platforms.  Consider that it supports two 
platforms with possibility of a third (in development); and there are only 
two main maintainers and several contributers (a third maintainer/developer 
works on the mac port).  If we ever get it working on ldc, we end up not 
only having to support multiple platforms, but also multiple compilers with 
there subtle differences.  Things get ugly almost exponentially this way, 
so we're not looking for ways to make the process more difficult.


-JJR