On Sat, Dec 22, 2012 at 2:50 PM, Mark Hahn <[email protected]> wrote:

> The confidentiality problem is impossible to fix in weak refs


For the record, I do not think it is impossible to fix, and I never stated
otherwise.

An approach we have talked about that I find promising (though I don't know
if it ever made it into any meeting notes) is to identify a weak ref with
the realm (set of globals) its makeWeakRef came from. Only if the object
the weak ref points to is in the same realm does it point weakly.
Otherwise, the weak ref points strongly. This prevents weak refs from
leaking gc drop info cross-realm. I believe most use cases would be
addressed within this restriction.



> and the
> impression I got from reading all that stuff was that it is considered
> a real problem.  My impression could be wrong.
>
> Does any other part of ECMAScript have that same problem?  If so then
> it could be argued that it doesn't matter because the cat is out of
> the bag.
>
>
> On Sat, Dec 22, 2012 at 2:44 PM, Rick Waldron <[email protected]>
> wrote:
> > confidentiality problem
>
> --
> Job Board: http://jobs.nodejs.org/
> Posting guidelines:
> https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
> You received this message because you are subscribed to the Google
> Groups "nodejs" group.
> To post to this group, send email to [email protected]
> To unsubscribe from this group, send email to
> [email protected]
> For more options, visit this group at
> http://groups.google.com/group/nodejs?hl=en?hl=en
>



-- 
Text by me above is hereby placed in the public domain

  Cheers,
  --MarkM

-- 
Job Board: http://jobs.nodejs.org/
Posting guidelines: 
https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
You received this message because you are subscribed to the Google
Groups "nodejs" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/nodejs?hl=en?hl=en

Reply via email to