Hello Rob
On 9 Aug 2006, at 02:01, Robert Shelton wrote:
> On Tue, Aug 08, 2006 at 09:09:29AM +0100, Nick Collins wrote:
> *snip*
>> This just gives a list of all the commands in time, the History,
>> which can then be stored in various document formats, and indeed,
>> played back by reinterpreting the appropriate strings in order at
>> the given times.
>
> Ahh, thanks very much. So where you say that you can render back your
> own performance you will get the audio part but not be able to see all
> of the typeing (live bug fixing) that went into it?
not quite, you get a list in an updating GUI
of each line (or 1st line if part of a block) of code that has been
evaluated,
along with who wrote the code (for networking stuff)
and the time in seconds against each since History started.
If you click on a line item in the GUI,
you get a new text document opened (or an option to overwrite an
existing one)
with the complete code in it's original form - eg a whole block.
If you want the audio back, just reevaluate the code that History has
returned to you.
> I'm thinking that a collection of recorded sessions (edit history)
> where we could watch code being written in real-time would be a very
> useful learning tool. This is similar to people saying that we should
> take videos of our screen, but storing it by programmer edits might
> lead to things like...
I guess history is more efficient as only evaluated code is recorded.
>> (it occurs that whilst it might lead to errors through undeclared or
>> wrongly initialised vars, you could start permuting the playback
>> order and/or timing and manipulate new patches out of existing ones-
>> I could pretend to type a lot quicker than I actually do!)
>
> ...or if you could could code side-by-side with a replay of a previous
> session and then merge the result. That might be fun and an
> interesting way to share material.
This would be possible with a few little extensions to the history
class.
Julian or Alberto may already have played around with this?
Regards
Tom Hall
> Thanks again,
>
> Rob.
> --
> Department of Computer Science and Software Engineering.
> The University of Melbourne, Australia.
Received on Sun Aug 20 2006 - 10:13:57 BST