[squeak-dev] computers too fast these days?

Bob Arning arning315 at comcast.net
Mon Aug 19 18:34:48 UTC 2013


Think about where it's used in the image now. I search for some text and 
it's not found. The screen flashes to let me know. I can't think of 
*anything* I want the UI process to be doing for the next 100ms while I 
wait for the photons to hit my retina. I'm even thinking a bit longer 
delay might even be better. Here are some experiments:

Make a morph.

m _ Morph new openInWorld.

Then try this, with different proposed #flash implementations.

m flash; delete.

Another experiment. How low can you set the delay and still reliably see 
red? Try this in a completely empty project.

m color: Color red.
World displayWorldSafely.
(Delay forMilliseconds: 20) wait.
m color: Color blue.
World displayWorldSafely.

I think the senders of #flash want it to be seen and I don't think the 
UI needs to do anything else while the message is being received by the 
user (the U in UI).

Cheers,
Bob

On 8/19/13 12:58 PM, tim Rowledge wrote:
> Seems to me there are two rather different usages here
> a) an alert within the normal UI, to let you know something odd happened
> b) a diagnostic when debugging where you need a 'bigger' alert.
>
> For b) a 100mS simple delay is probably perfectly fine. I've done debugging where it would be really nice for there to be a good way of 'interrupting the delay' to get in to the problem process; an associated visual widget to click on or something.
>
> For a) a delay that holds up everything is not a good idea. Some form of alarm (like the morphicalarm is supposed to be) that keeps the UI moving is going to be least annoying. I think it's likely that there are better feedback concepts than simply flashing the morph.
>
> tim
> --
> tim Rowledge; tim at rowledge.org; http://www.rowledge.org/tim
> A paperless office has about as likely as a paperless bathroom.
>
>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20130819/5e459c45/attachment.htm


More information about the Squeak-dev mailing list