Opinions on DConf talks

Joakim joakim at airpost.net
Tue Jun 25 08:18:44 PDT 2013


Now that the last talk from DConf is up, I thought it might be a 
good time to review them, as someone who didn't attend.

 From the standpoint of a D conference, the worst talk was 
Walter's.  It was clearly aimed at a non-D audience, so there was 
nothing new there for a D audience: I'm guessing it's a familiar 
talk he gives to outside groups.  However, as a recording on the 
internet, it might be the best talk for newbies, as an 
introduction to the language, as every other talk goes deeper 
into the language.

I wish Walter had done a real technical talk on his experiences 
developing D and dmd, as he is a font of knowledge on many 
technical topics, rather than a basic overview of D.  Maybe next 
year, perhaps he can talk about porting dmd to D. :)

The rest of the first day's talks were interesting looks at 
various technical issues.  Ben loves testing a bit too much, ;) I 
suspect that a distributed approach like Robert's is the future.  
Didn't realize there was so much to copying and moving till I 
watched Ali's talk, still not sure I grasp all of it.  Good to 
see a different approach to GC by Leandro.

I was underwhelmed by Manu's talk: too much low-level technical 
detail about the integration effort between C++ and D, not enough 
discussion of the benefits of using D.  The Q&A panel with Walter 
and Andrei should have been an hour, or until questions petered 
out, and held every day of DConf. :)

Vladimir's talk was the highlight of the second day for me, a 
great mix of technical material and exploring the current state 
of related D libraries.  Adam's talk was enjoyable, a nice look 
at D through C# eyes.  Iain and Rainer's talks were interesting; 
started watching but haven't finished Martin and Maxime's talks 
yet, not that interested in shared libraries or JITs but I'll 
finish them later.

Don's talk was the best of the conference: great mix of technical 
material, pragmatic considerations, and humor.  The title was 
horrible though, wasn't expecting much from a talk on 
"metaprogramming," was pleasantly surprised when the topic was 
barely talked about.  Nice overview of SIMD by Manu and LLVM/LDC 
by Nadlinger.  Simcha's talk was well done but I wonder if I'd 
ever use any of those higher-level patterns; good to hear what 
Stefane's team is up to with static analysis.

Andrei, as usual, was very good, though there was too much 
structure and boilerplate for me, however little there was 
compared to his usual talk.  The bits about scaling to a million 
users through "professionalism" were weird for a volunteer effort 
though.

All in all, a great effort, looking forward to the next one.


More information about the Digitalmars-d mailing list