DConf 2019: Shepherd's Pie Edition

Adam D. Ruppe destructionator at gmail.com
Sat Dec 22 20:12:29 UTC 2018


On Saturday, 22 December 2018 at 17:13:06 UTC, Mike Parker wrote:
> We like the current format and see no need to change it at this 
> time.

That's all it really comes down to. Y'all like it.

But the time and money COULD be put to far better use. Consider 
this: keep the same schedule, but the talks change from being 50 
minutes of awkward power point to being:

15 minutes - the talk part. Speaker introduces the topic and 
proposes something for everyone to work on.

20 minutes - the conference attendees split into a few work-sized 
groups and do something about the proposed topic. Maybe 
mini-hackathon. Maybe just discuss it. Maybe toy around with the 
stuff. The speaker wanders around groups to help guide them as 
needed and see what they are talking about.

15 minutes - The speaker goes up front again to share what was 
learned from the groups. Open discussion is encouraged. If 
appropriate, an action plan is decided upon. It concludes with a 
feeling of accomplishment.


All topics and talk handouts and powerpoints MUST be made public 
ahead of time for people who want to study it and have some 
thoughts prepared going into it.



For example, let's look at last year's first few talks. 1, memory 
safety. Walter introduces the problem and the new features. The 
groups spend the work session actually trying the feature. Try it 
on your code. Try to break it. Talk about how you hate it in your 
little group, whatever, just get hands-on in-person.

Then, Walter could come back and demonstrate the stuff (basically 
the same as the second half of his old slides) for everyone - 
with the group's comments added. When we get to the last slide 
with "more work to do", the audience knows this - they might have 
even filed some bugs from their experience already! Hackathon 
work lined up.

BetterC talk? Basically ditto.


So what's the advantage here over just lecturing?

* The audience is more engaged. Many people learn more by doing 
than by just watching.

* The work groups can mingle a bit and maybe get to know each 
other and learn from each other. (I'd randomize these a little.)

* Everyone trying it together means they may be able to write bug 
reports, get more insight into DIPs, be prepared to review PRs 
since they have at least SOME experience.

* People's questions will be more refined at the end of the talk.



Pretty small tweak to the current format - and not all talks need 
to be the same, just a new option here - and I think it would 
make a lot more out of the in-person time.


More information about the Digitalmars-d-announce mailing list