[IAEP] Journal prompts (was: Re: [support-gang] FW: [OLPC Bolivia] No logro aprender Sugar / I cannot learn Sugar)

[Walter Bender]

On Wed, Jun 15, 2011 at 5:58 PM, Christoph Derndorfer
<e0425826 at student.tuwien.ac.at> wrote:
> Am 15.06.2011 18:12, schrieb Walter Bender:
>> All of that said, let me repeat an argument I made regarding the Sugar
>> Journal during the EduJam summit last month: we developed the Journal
>> not because we wanted to be incompatible with the rest of the world
>> but because we wanted to address some pedagogical needs. Specifically,
>> we want the children to have a place to reflect upon their work. The
>> Journal is their portfolio. Reflection requires effort and some
>> developers consider the prompts to write in the Journal as an
>> annoyance. But when I ask those same developers if they think adding a
>> commit message to their commits in git, they immediately understand
>> the value. So some of the annoyance of the Journal is because we have
>> not completely solved the UI issues (the good news is that Simon has
>> some patches landing that fix some of these issues) but some of the
>> annoyance is because we want to make the path of least resistance be
>> one where the children are prompted to be reflective-- to write in
>> their "lab notebooks" about what they are doing and why and to make
>> presentations to their teachers, parents, and fellow students about
>> their work. (The latter is facilitated by the new Portfolio activity.)
>>
>> In any case, concrete feedback and criticism is welcome. Thanks.
>
> Hi Walter,
>
> I'm not sure whether this really counts as concrete enough feedback but
> I'll do my best.
>
> The very short version is that while I think that having a tool for
> reflection, annotations, metadata, whatever you want to call it makes
> sense yet I personally don't think that Journal prompts are the answer
> at this point in time.

The prompts have been removed, so they are *not* the answer at this
point. The issue I was trying to raise was that there was a reason for
the Journal prompts and that rather than just eliminate an annoyance,
we should try to come up with a solution that still promotes the
positive aspects.

>
> Regarding the git-example you mentioned here and at eduJAM! I also think
> that the comparison doesn't really work. The main value of git's commit
> messages in my mind are:
>
> (a) as a record of changes so you know which version to revert to when
> things go awry
> (b) as a way to make other developers in collaborative projects aware of
> such changes in general

Well, I don't think these are too far off the mark, but the process of
writing a decent commit message is also a process of reflection. I
wasn't expecting the analogy to be perfect, but I did want to make the
point that as a community we see value in writing in our "lab
notebooks" and I am hoping that likewise we see the value in providing
the children a place to write about their work as well.

>
> As the Journal currently doesn't provide an accessible way to revert to
> older versions of the same entry use-case (a) isn't really available.

Although I have advocated git as a backend for the Journal, as you
correctly note, the Journal is not at present a versioning datastore.
But commit messages are not versioned in the same sense as the data.
There is a running log of messages, much more like a diary (or dare I
say it, lab notebook).

> Since collaboration is also very seldomly used today and even less on
> ongoing projects or activities which span days, weeks, or even months
> and additionally we don't have facilities to merge, fork, etc.
> collaborative Journal entries also (b) isn't available to Sugar users today.

Sounds like you'd like to work on the Sugar Bulletin Board project...
our long-dormant system for persistent collaboration. Never really got
off the ground, but we do have a key on the OLPC keyboard still
dedicated to it. But I am not sure what this has to do with writing
notes in the Journal?

>
> So what we seem to be left with at this point is a system which tries to
> force users to add comments or tags even though there is very little
> incentive or supportive mechanisms to actually do much with that metadata.

One of my favorite examples of using a computer is school is a 1-to-1
program at a middle school in Dorchester, MA. The kids spend five
minutes writing at the end of *every* class, including gym. Even if
they never use "that metadata", the act of reflecting is important.

We do have some tools/supportive mechanisms for using that metadata,
including the Portfolio activity. Have you tried it?

> Now one could argue that just because we don't have the other parts of
> the puzzle yet we should still try to get users acquainted with the
> general notion. Yet here I would say that there's a significant risk of
> conditioning people to simply do whatever it takes to get rid of the prompt.

I guess we disagree about the inherent value of writing notes.

>
> I don't have any data but would assume research into the effects of such
> forced decision making - whether it's Microsoft's classic "Ok or Cancel"
> metaphor or Android's App permission system - exists and probably shows
> that people click whatever gets them back to what they really wanted to
> do as quickly as possible. So from that point of view it seems to make
> more sense to wait until tangible benefits are available to users before
> confronting them with the prompt.

It is moot, since the prompt, which incidentally only occurred on the
first use of an activity, is gone.

>
> One somewhat intermediary approach which seems concrete and actionable
> here even in the short term (and regardless of whether we have mandatory
> prompts or not!) is to use some sort of automatic suggestions for
> Journal entries. A very simplistic example would be to copy the
> behaviour of traditional office solutions and suggest the first few
> words of a Write session as the Journal entry's title.

There is some work going on with a more sophisticating tagging
mechanism. It seems as far as annotations, it will be very much
activity specific. There is nothing to prevent an activity from adding
prompts or comments to the description field.

The plain-text Edit activity already copies itself into the metadata
of the Journal in order to make its text be full-text searchable. I
think some other activities may do similar things. But this is a bit
different from the annotation concept.
>
> A more invasive solution would be to take a note from how you can access
> a Journal entry and its metadata (though only tags at the moment) when
> viewing photos within the Record activity. People like Gary C will
> likely have a more fact based opinion here but to me it seems that
> offering the ability to add information to a Journal entry while you're
> actually working on it is a good option (again regardless of whether you
> force prompts upon closure or not).

This is similar to the approach that Simon is working on.

>
> Last but not least and on a more abstract and philosophical level I
> think that Sugar should create possibilities and both implicitly and
> explicitly encourage certain behaviour without forcing it on users.
> Would we like to see collaboration between users? Yes. Do we make all
> activities show up in the neighborhood the moment they're launched? No.
> Would we like to see children dive into the source code? Yes. Do we make
> them read libabiword man pages before starting Write? No.
>
> More than anything it is about showing Sugar's users the value and power
> of tools like naming, tagging and comments. Including relevant
> information in teacher training and general support materials such as
> the Help activity could probably also go a long way here.

+1. We want to make the path of least resistance have the maximum
impact on the learning.

I gather from your overall comments that you were focused on my rant
at EduJam about removing the "Naming prompt". I teed off on that as a
way to inject into the minds of our developer community the rational
behind it in the first place, not as a defense of a specific
implementation, which was generally flawed. I am not advocating
returning the prompt, but rather I am trying to push the Journal UI in
a direction that I believe has real significance to our learning
mission. The Journal rightly gets a lot of criticism -- it is flawed
-- but it is also there for reasons that are often overlooked.

>
> Cheers,
> Christoph
>
> --
> Christoph Derndorfer
> co-editor, www.olpcnews.com
> e-mail: christoph at olpcnews.com
>

thanks.

-walter

-- 
Walter Bender
Sugar Labs
http://www.sugarlabs.org