irclog.perlgeek.de/perl6book/today | source: github.com/perl6/book/
Set by moderator on 26 April 2011.
13:25 sjn joined
sjn moritz: ping 14:42
15:27 p6eval left 15:28 p6eval joined 15:44 p6eval left 15:46 p6eval joined 16:08 p6eval left 16:10 p6eval joined
moritz sjn: pong 16:16
sjn o/
sjn and krunen have played a little around with an outline 16:17
and tried to figure out what needs to be in place to make a perl6 book with aspects
moritz: you up for spending a moment on this? I'd like a quick sanity-check :) 16:18
moritz sjn: yes, though my tuits are non-continuous due to $daughter 16:19
sjn ook :) 16:20
moritz: here's where we've jotted down a couple of our (unstructured) notes... 16:21
github.com/sjn/book/tree/outline-rework 16:22
where perhaps github.com/sjn/book/blob/outline-r...-goals.pod is the most useful one
moritz takes a look 16:24
sjn the question I'm trying to answer with those, is "what qualities of a text do I have to be inflexible about in order to be flexible about how the content is told?"
so, my hypothesis is to try to decide an outline that both covers the needs of a reference and the needs of an introductory book 16:25
and then stick to that, while "fleshing out" the narrative and specifics according to what's needed or practical 16:26
now, you should know that this is very much an experiment that I'm proposing here. :) 16:27
trying to figure out a good way to allow for those nice and juicy aspects, while still keeping the writing task manageable 16:28
16:28 p6eval left
moritz can some parts of the outline be simply missing in one of the aspects of the book? 16:29
for example for the current intended audience of the book, I wouldn't want to write about "Chapter 1.2 - What is a program?" or "Chapter 1.4 - Breaking a big problem into smaller problems" 16:30
16:30 p6eval joined
sjn moritz: that's one of the points that krunen and I have talked the most about 16:30
his view is that there's value in removing some sections/chapters, and consequently renumbering the superceeding chapters 16:31
I disagree with him
I think it's worth keeping the sequence intact 16:33
moritz is pretty sure he agrees with krunen
so will you have empty chapters in some aspectsß
s/ß/?/
sjn especially for the situations where there are several readers that try to keep a equivalnet pace of progression
(e.g. in a classroom)
I'm proposing "super-short" chapters 16:34
just mentioning the absolute minimum, so the reader can decide if they want to switch to an aspect with more information
moritz how does that look like, for a printed book?
sjn and still be able to do the rigth thing when the teacher says "no, look at the first task in chapter 12" 16:35
moritz: probably (in the index) like this:
chapter 1 (intro)...... 1 16:36
chapter 2 (more stuf).. 1
chapter 3 (even moar).. 1
etc.
until there's actual content
moritz I think the main impedance mismatch between your and my vision is that I never thought of "Using Perl 6" to be used in a class room 16:37
sjn yep
moritz and it's not the vision i cater for, because it's too long in the future that there will be any Perl 6 classes
sjn that's why I think it's especially interesting to try make this work :)
a reference is still needed 16:38
moritz but that's why I'm not keen on compromises that make it easier to use in the class room (ie same chapter numbers for same things among different aspects)
sjn well, that compromise is a rather small one for you :)
moritz doesn't feel that way 16:39
sjn and it opens up quite a few interesting options later
atm, we're basically talking about what sequence of events to write about
now, with any book, it helps to choose a sequence where you can build on previous events in that sequence 16:40
that's true for any linear book
I'm proposing to make that sequiens in that linear story into something "special" that can be reused in several situations 16:41
sequence*
"sequence" as in "outline"
now, to actually make this work, I thinbk it's _very_ important that the outline doesn't say _too_ much about the content 16:42
just maybe a few keywords about what one can expect to learn there
very general, and high-level 16:43
it's like deciding up front the names of multi's that should be available in a module 16:44
and then let different people start writing those multi's
...with a little guidance from a narrative ("we expect the multi to return $foo") 16:45
...and leaving the rest up to the writers to flesh out. 16:46
does this make sense?
moritz yes, but it doesn't change my opinion about the compromises 16:47
sjn :-/ 16:48
would you mind telling me a bit about what you're thinking? 16:49
moritz writing a book is a lot of work
and I don't know if we ever get further than a single aspect
or maybe two
16:49 sjn_ joined
sjn_ <- server reboot, sjn's logging off 16:50
moritz so I don't want to make comprises on the first one we want to print
just for the possibility of integration with future aspects
sjn_ well, a big point for me is that I'd like to make it easier to involve new people in the writing process 16:51
moritz or put another way, I don't believe strongly enough in the multiple aspects to accept a drag down for the first aspect
sjn_ so there's that
also, I'm fully aware that adding aspects later is possible
but my guess is that it might require a lot of rewriting 16:52
and my guess is that over time, (as people find the text useful, and perhaps make use of it in new circumstances) there rewriting my become much more costly 16:53
especially when we're starting to handle overlapping aspects
so for me, deciding a broad-but-shallow outline is a way of reducing the amount of later work 16:54
maybe it's a bad strategy (premature optimization)
but together with that, I know the immense value a good narrative has for any book 16:55
moritz not bad, just not my approach at doing things
sjn_ even a reference
moritz my approach is more bottom-up
sjn_ yep
and I'm striving for a top-down approach 16:56
which makes this whole thing especially useful :)
s/thing/discussion/
moritz: do you have any suggestions of how we can make this work? 16:58
moritz sjn_: sorry, have to take care of my little daughter, will be back in an hour or so
sjn_ ook, I'll hang around here :) 16:59
17:02 sjn left 17:14 sjn_ is now known as sjn
moritz back 17:31
one option is to drive up6 to the point where it can be printed
and then refactor it enough to allow other aspects 17:32
another is to fork the project (we can still interchange contributions between the forks) 17:33
a second mismatch is that the our current approach is example driven 17:34
and your proposal is outline driven
(we do care about structure, but finding really good examples is so hard that we compromise with the outline a bit 17:35
dalek ok/first-example-more-robust: e9930d3 | moritz++ | src/basics.pod:
[basics] fix duplicate invocant explanation found by PerlJam++
 17:38
ok: 0281365 | moritz++ | src/basics.pod:
[basics] made first example more robust wrt whitespace. Text still needs updating
ok: 346ed6a | moritz++ | src/basics.pod:
[basics] start to updated description of the new first example
ok: e83af18 | moritz++ | src/basics.pod:
[basics] further wording changes after the s/split/words/ refactor
ok: e9930d3 | moritz++ | src/basics.pod:
[basics] fix duplicate invocant explanation found by PerlJam++
ok: 3d702f6 | moritz++ | src/basics.pod:
Merge branch 'first-example-more-robust'
sjn is back 17:47
moritz: my porposal being outline-driven is only partly true
krunen and I spent some time asking questions on "what makes a good example" and "what makes a good task" 17:48
and an important point we found, was to make sure that the examples and tasks in the text were part of a larger, and continuously _useful_ story 17:49
that idea we talked about was that the "story" started with the very basic task of creating a tool for asking questions and getting answers 17:50
our "story" here (as far as we got) was about different ways to extend that tool into something bigger and more useful 17:51
e.g. storing the answers for later review
e.g. giving grades to the answers as the review
e.g. trending the review, so one can get an idea about retention/results 17:52
ultimately, our idea was that the student made a tool for doing online exams
with the option of using question databases, central storage, or maybe even asking the student to formulate good questions herself 17:53
plenty of options there, both to get creative, and to cover all kinds of useful ideas within the programming language 17:54
(including related topics, like revision control, cooperation, source code clarity, etc.) 17:55
moritz it's a nice theme 17:56
... and far away from what we have done so far
sjn agrees it's a nice theme 17:57
but my point is that the examples _are_ the story 17:58
or at least _part_ of the story
so "example driven" is in essence the same as "outline driven", it's really about the story 17:59
well
maybe not
you can make examples that don't convey a larger story
and in a reference, that's perfectly fair
I still think a book becomes better if you make the reader ask "oh no, what'll happen next?" now and then :) 18:00
burt to do that, you actually need a story to tell
but* 18:01
so the real question here is if the UP6 reference text would become better if it's examples were part of a bigger story 18:34
sjn is thinking "maybe not for the minimal examples"
18:53 fhelmberger joined 21:26 fhelmberger left

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!