[squeak-dev] The Trunk: Morphic-mt.1852.mcz

christoph.thiede at student.hpi.uni-potsdam.de christoph.thiede at student.hpi.uni-potsdam.de
Sun Feb 6 18:39:02 UTC 2022


Hi Marcel, hi all,

what about just keeping the orange triangle when file-out is disabled and there is no accept action? Maybe this would be a clearer indication that nothing has been saved. Or are we relying on a "clean" text morph for any common interaction scenarios? Would this affect Chris' described "checkpoint" workflow or is this workflow no longer available anyway? :-)

Best,
Christoph

---
Sent from Squeak Inbox Talk

On 2022-01-29T18:08:24+01:00, marcel.taeumel at hpi.de wrote:

> Hi Jaromir --
> 
> I don't think that you want to actively manage files during regular Smalltalk programming. Workspaces are neat. They cannot only hold code but also generic notes side-by-side. No need to plan ahead. Just start coding in a Workspace and see where it takes you. Maybe more code. Maybe some notes.
> 
> @all: I updated the #fileOutOnAccept preference to let the user opt-in per Workspace window as Jakob suggested:
> 
> 
> 
> The Smalltalk veterans may either click "No" here or disable #fileOutOnAccept to not get such a prompt in the first place.
> 
> In the long term, maybe, we may want to offer users to have "workspaces" that are not bound to an actual window. And offer a seamless transition from workspace-scripts to classes+methods. So many ideas. :-)
> 
> Best,
> Marcel
> Am 28.01.2022 17:47:46 schrieb Jaromir Matas <mail at jaromir.net>:
> Hi Chris,
>  
> >>  Yes, that's exactly right. Reading Chris's reply I started to suspect I'm using the Workspace for what it wasn't made for: I test various snippets in the Workspace and put down expectations, notes, results etc, to be copied to some document later. Checking standard menus I can't see any text editor - have I missed something important here? (I learn by myself so the clues in the system are crucial for me)
> 
> > Yes, you can use World | open... | file... or simply open Squeak's
> FileManager browser and create and open the file there.  It's just as
> functional as a Workspace, and you can save it and even append to it
> if you want.
> Thanks for pointing me to a text editor from a file, I was not aware of this functionality :)
>  
> Unfortunately there seem to be a couple of problems:
> 1) cancelling the accept menu opens a Debugger; I can't figure out how to fix this
> 2) the text editor differs from the Workspace at least in that it doesn't declare local variables automatically; I like to inspect variables after a do-it and I actually don't know how to work around this at the moment...
> 3) the right-click menu has fewer options than Workspace's and is a bit inconsistent (e.g. the more... option is not the last one); it feels a bit "forgotten"... but having a text editor sounds like a good idea at any rate.
> (Besides that I enjoy having Transcript embedded in the editor/workspace window to see messages from the experiments without always bringing Transcript to the front; but yes, that's my personal preference and I could embed it in a text editor too)
>  
> 
> > Workspaces are intended for *short-term* experiments, but what you're
> doing sounds more like long-term research that deserves something
> better than a Workspace, something already able to persist.  It might
> still be worth considering whether each of the various snippets you're
> testing could be titled with its own method name on the class side of
> a "MyWorkspaceOrganizer" class.  If so, it solves _everything_ we've
> been talking about.  Backups.  Synchronicity.  Image crashes.  Etc.
> It also enables method composition and you could even make versions of
> it..
> E.g. recently I?ve been playing with small things like
>  
> p := [Semaphore new wait] fork.
> Processor yield.
> p suspend.
> p resume
>  
> p := [Semaphore new wait] fork.
> Processor yield.
> p suspend.
> p
>  
> p := [Semaphore new wait] fork.
> Processor yield.
> p
>  
> p := [Semaphore new wait] fork.
> Processor yield.
> p resume
>  
> In this case the Workspace seemed most suitable. But for some real/larger work organizing the experiments in methods sounds like a good idea ? thanks!
> 
> > > > I just realized being able to save Workspaces opens this can of worms of needing a stale-detection system in any case -- otherwise users will face this "doubt" every single time they launch the image. (Chris)
> 
> > > Yes, true, I know this feeling very well :) That's why I like Marcel's idea to append the accepted contents to the saved file rather than overwrite; it's a backup anyway so when a disaster (or doubt) strikes one can check the backups.
> 
> > The above-mentioned file features have an append feature, however, I
> don't think that would help if the image crashed and you didn't save.
> Marcel?s Workspace append has the cool feature it is timestamped so it?s easy to separate the latest accepted version of the Workspace :)
>  
> Thanks again for your tips!
>  
> Best,
> Jaromir
>  
>  
> From: Chris Muller <asqueaker at gmail.com>
> Sent: Friday, January 28, 2022 1:56:46 AM
> To: mail at jaromir.net <mail at jaromir.net>
> Cc: squeak dev <squeak-dev at lists.squeakfoundation.org>; tim Rowledge <tim at rowledge.org>
> Subject: Re: The Trunk: Morphic-mt.1852.mcz
>  
> > Yes, that's exactly right. Reading Chris's reply I started to suspect I'm using the Workspace for what it wasn't made for: I test various snippets in the Workspace and put down expectations, notes, results etc, to be copied to some document later. Checking standard menus I can't see any text editor - have I missed something important here? (I learn by myself so the clues in the system are crucial for me)
> 
> Yes, you can use World | open... | file... or simply open Squeak's
> FileManager browser and create and open the file there.  It's just as
> functional as a Workspace, and you can save it and even append to it
> if you want.
> 
> Workspaces are intended for *short-term* experiments, but what you're
> doing sounds more like long-term research that deserves something
> better than a Workspace, something already able to persist.  It might
> still be worth considering whether each of the various snippets you're
> testing could be titled with its own method name on the class side of
> a "MyWorkspaceOrganizer" class.  If so, it solves _everything_ we've
> been talking about.  Backups.  Synchronicity.  Image crashes.  Etc.
> It also enables method composition and you could even make versions of
> it..
> 
> > > > I just realized being able to save Workspaces opens this can of worms of needing a stale-detection system in any case -- otherwise users will face this "doubt" every single time they launch the image. (Chris)
> >
> > Yes, true, I know this feeling very well :) That's why I like Marcel's idea to append the accepted contents to the saved file rather than overwrite; it's a backup anyway so when a disaster (or doubt) strikes one can check the backups.
> 
> The above-mentioned file features have an append feature, however, I
> don't think that would help if the image crashed and you didn't save.
> 
>  - Chris
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20220129/f1e97843/attachment.html>
> -------------- next part --------------
> A non-text attachment was scrubbed...
> Name: image.png
> Type: image/png
> Size: 22155 bytes
> Desc: not available
> URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20220129/f1e97843/attachment.png>
> 
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20220206/0ff0c35b/attachment.html>


More information about the Squeak-dev mailing list