The whole purpose of using J for your multithreading is that you won't
have to worry about low-level race conditions. (Once we get it right).
You will have race conditions at a higher level. If two threads are
relying on the same changing global name, you will have to use
yet-undefined functions to manage that.
If you don't use any shared writable globals - e. g. if you pass each
thread everything it needs as arguments - you will have no race
conditions at all.
Henry Rich
On 4/13/2022 1:26 PM, 'Pascal Jasmin' via Beta wrote:
The main point I was trying to make was:
And then god/J gave us t. ... and some of us rejoiced because now our
imperative code could add the fun of race conditions... not.
I'd rather be discouraged from code races, rather than have debugging effort
for it.
On Wednesday, April 13, 2022, 12:47:28 p.m. EDT, Henry Rich
<[email protected]> wrote:
Yes, I made a typo. Your sentence will run fine in the next beta. I'll
wait a day or so before asking Eric to release again.
Till then if you confine your tasks to using private names you will have
better luck.
Henry Rich
On 4/13/2022 12:28 PM, 'Pascal Jasmin' via Beta wrote:
(3 : '-:`+:@.G y'"0 t {:@>@[ 3 : 'G =: -. G'"0 t) i.1111
--
This email has been checked for viruses by AVG.
https://www.avg.com
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
