MozillaZine

PSM 2.0 expected by May

Thursday January 25th, 2001

Bob Lord explains in n.p.m.crypto:
"We've just posted the first draft of several PSM 2.0 (codename PIP) documents. In addition to the Roadmap which we posted a few days ago, you'll now find the first cut at the workplan (a version of what we internally call the 'PRD'), a tasklist, and a link to the UI mockups.

Start here: http://www.mozilla.org/projects/security/pki/psm/

These documents will evolve over the next few weeks as we learn more and as Mozilla developers identify the areas where they'd like to help (platform support, for example).

We're trying to release PIP by the end of May. We'll soon post more information about building PIP, as well as the target date for switching the nightly builds over."


#1 Every bit counts?

by jelwell

Thursday January 25th, 2001 9:50 PM

Reply to this message

This is to cut down the overhead that is required on some OS's when creating new processes. This will help keep the memory requirements a tad lower. But I doubt most desktop users will even notice this minute cut in memory requirements.

#3 Re: Every bit counts?

by sleepy

Friday January 26th, 2001 9:02 AM

Reply to this message

If I'm reading "in-process" correctly, it probably means that we don't need to wait for PSM to load the first time we enter a secure connection. This is a significant speed up because even on my Pentium II, current PSM takes about 5 seconds to load. I think "in-process" is the way it should be because all previous browsers can make secure connections instantly.

#2 Specifics?

by Martyr

Friday January 26th, 2001 6:02 AM

Reply to this message

Even 8 Mb makes a difference on older systems...and if ground floor is 150 Mhz, then that will be noticable my at least some.

#4 Why less meory requirement?

by spaetz <Sebastian@SSpaeth.de>

Friday January 26th, 2001 3:59 PM

Reply to this message

I'm reading in-process PSM would lower mem requirements a tad. But how so? Just now, PSM is only loaded if needed during a session, which might not be always. While an in process PSM would be loaded upon start of mozilla, increasing start up time, right? Or will it still be dynamically loaded? Sorry, if the answer is in the PSM docs, I haven't read them yet.

#5 Re: Why less meory requirement?

by basic <_basic@yahoo.com>

Friday January 26th, 2001 5:02 PM

Reply to this message

It will be dynamically loaded with less overhead if I read the docs correctly.

#6 taking bets

by Kovu <Kovu401@netscape.net>

Friday January 26th, 2001 6:50 PM

Reply to this message

I'm taking bets on whether this comes out right before Mozilla 1.0.

#7 KEY ESCROW??

by Dayspring

Sunday January 28th, 2001 6:27 AM

Reply to this message

WTF?<BR>

<I>PSM 2.0 Prioritized Feature List Retain all existing PSM 1.4 functionality including (but not limited to) these features:<BR> ... Encryption key escrow<BR> ... </i><BR><BR>

I\\\'m not very confortable with this. I haven\\\'t even seen an option to turn it off.

#8 Re: KEY ESCROW??

by rblord <lord@netscape.com>

Sunday January 28th, 2001 1:30 PM

Reply to this message

Some Certificate Authority (CA) products (such as the iPlanet Certificate Management System) have a key recovery feature. The CA administrator can configure the CA to escrow a user's encryption key during the enrollment process. When this happens, the user gets a warning dialog and has the option to continue or to cancel.

There is never a time when the CA gets your encryption key without your consent. We'd never let that happen.

I'd point to the exact text, but haven't completed those UI mockups yet. :-)

-Bob Lord