I want to elaborate - with the help of some technical information Arne Langsetmo provided - on my previous comment, in which I assume that the proclaimed "foreign-to-foreign" intelligence "gap" is not one of telephone communication surveillance, but rather one of email surveillance in connection with U.S.-based InternetServiceProvider servers.

[Aside to casual_observer: I concur completely with EFF's statement and your point about the absurdity and futility of debating issues in the dark. But because we aren't legislators, the more they try to keep us in the dark about the actions of our government, the more I feel compelled to try to force the facts out into the open. So though we can't be sure of our theories, we can challenge the secrecy, while continuing to point out the impossibility of conducting a full and fair public debate while so handicapped, as you rightfully do.]

At Balkinization, Arne helpfully critiqued Marty Lederman's (year-old-now) supposition about which TSP-related basket warrant application(s) were disapproved by FISA Court judges in early 2007. Lederman's otherwise-excellent legal analysis specifically removes - I think mistakenly - email surveillance as the candidate for the FISC-identified problem, because it's not a controversial fix in Congress.

I stated, based on information revealed at a March ABA forum, that though not a problem politically, technically fixing the email "gap" seems likely to be challenging. [This is based on the premise that literal "foreign-to-foreign" telephone calls intercepted on U.S. soil are not now and never have been covered by FISA, as revealed at the ABA forum, and as referenced and accepted by Lederman.] My partial transcript of that forum includes this exchange, which seems to indicate that there is definitely some sort of email-related "dilemma," per Ken Wainstein of the DOJ:

David Kris: "So in an effort just to sort of move this one, tiny millimeter forward, the question for Ken is:

.

You heard Jim [Baker] throw down the gauntlet. He said outright [earlier in the same ABA forum, paraphrasing Baker]: 'Foreign to foreign wire and radio communications are outside of the scope of FISA, have never been within the scope of FISA, regardless of whether the surveillance occurs in the U.S. or elsewhere.' Is that true, or is that false, or you can't answer the question?"

Kenneth Wainstein, Assistant Attorney General, National Security Division, DOJ: "I agree with him [Jim Baker] that foreign to foreign was not a part of original FISA."

Kris: "And, has it ever been within the reach of FISA - just let's, limiting to wire and radio comms now, not...?"

Wainstein: "Well, the concern is, and I think he [Jim Baker] alluded to this, that, especially with email, at the time of interception, you don't know where the recipient's going to be. So carving the world of surveillance up between foreign-to-foreign and everything else is good in certain areas of surveillance, but for instance in email, it doesn't get you where you need to be, because, at the time of surveillance you're not going to know if it's foreign-to-foreign or foreign-to-domestic. And that's our, that's the dilemma."

http://letters.salon.com/opinion/greenwald/2008/03/07/surveillance/permalink/b0e57886590600cb4102bbf2bc1a4c13.html

I think that clearly acknowledges an inherent technical problem and FISA trigger for U.S.-based email surveillance (both real-time and stored) that Lederman's analysis over-simplifies. Lederman does finger a "foreign-to-unknown" communication issue (which the government obscures as 'foreign-to-foreign calls'), by hypothesizing that some fraction of (apparently) calls (as opposed to emails) can't be pre-identified as foreign-to-foreign, and that this technical problem may be where the FISC dispute lies:

Perhaps, for example, certain types of surveillance of wire "switches" in the U.S. do not give the NSA the capability, at the outset, to tell if a particular communication involving a "target" in a foreign location has a terminus in the U.S. -- in which case at least some unknown number of the communications intercepted at the domestic switch would be "electronic surveillance" requiring a FISA court order.

To which Arne posted this response over at Balkinization:

For phone communication, it is possible with high (close to 100%) accuracy to determine the geographical locus of the participants.

.

And even for IP communications, snooped in real-time, you can tell with high certainty the location of the communicating entities as well (just note the location-specific ads that pop up on your screen when you're on the road, for instance). This must be the case, because routing wouldn't work otherwise....

http://balkin.blogspot.com/2008/06/what-fisa-debate-is-not-about.html

Which seems to pretty definitively rule out a technical problem involving U.S.-based surveillance of "foreign-to-unknown" phone calls as the cause of 2007's FISA Court reimposition of "thousands" of individualized, probable cause warrant applications for whatever the "intelligence gap" problem was (pre-PAA) and will be (post-PAA authorizations).

Whereas Jim Baker [the former head of the DOJ office responsible for preparing FISA warrant applications] makes clear during the linked ABA forum that there is a problem of "foreign-to-unknown" email interception (that results from the indirect, intermediary-hosted nature of email transmission).

[As Arne notes, the technical methodology of transmitting [real-time only?] email is more akin to transmitting radio broadcasts [per 1801(f)(3)], than it is to conducting telephone calls (the unidirectional nature of email, in particular). Of course the nature of the content of (stored and real-time) email is very much more akin to the content of private phone conversations than to public radio broadcasts, when taking the 'spirit' of FISA, rather than the technical realities of communication transmission, into consideration.]

In short, a great deal of foreign to U.S.-server-aka-"unknown" (until retrieved) email seems to be covered by "old" FISA's (pre-email) probable cause warrant requirements. Foreign phone calls don't seem to have that problem.

Note, also, that the experienced Jim Baker (who praises and commends the effectiveness and utility of "old" FISA) considers the domestic surveillance in the Cheney/Rockefeller FISA bill "much broader" than needed for even 2006's so-called Terrorist Surveillance Program.