packages broken by this change are not relying on undocumented behaviour; they're relying on behaviour which is documented as being unreliable.
Either way; broke is broke. No one is suggesting that the change shouldn't be implemented, but is it unreasonable to want to know what the supposed benefits are intended?
In reply to Re^20: Hash order randomization is coming, are you ready?
by BrowserUk
in thread Hash order randomization is coming, are you ready?
by demerphq
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |