I just sent this letter to New Scientist magazine; no idea if they’ll print it or not, so I’m posting it here too, as several people have recently expressed interest in this idea:

James Love’s proposal for cash prizes to replace today’s patent monopolies (“Fair prices, fair profit”, page 24, Nov. 10-16 New Scientist Special Issue) would be a welcome improvement on the current patent system. But there may be an even better solution: sliding registration fees with a public buyout option.

Under this system, the patent applicant names a monetary value for the patent at the time of registration. They may pick any value at all — ten dollars or ten billion — but the registration fee will be a percentage of that value, so there is an incentive not to declare too high. Each year upon renewing the registration, the patent owner has a chance to adjust the declared value up or down, and the renewal fee adjusts accordingly.

Now comes the key: since the declared value of the patent is a matter of public record, any party can liberate the patent into the “public domain” (to use the language of copyright law, whence this proposal originally came) by paying the patent holder that amount, in a mandatory transaction. The registered owner must accept, and, having chosen the value in the first place, cannot claim to have received less than the patent’s worth.

This system preserves all the market dynamics that defenders of the current patent system rhapsodize about: there is still an open market in patents, and a patent can sell for more or less than its registered value (since a purchaser may be interested in retaining the monopoly, rather than in liberating). At the same time, the public always has a clear path by which to liberate a given patent, and at a speed that matches the urgency of the public’s need.

By doing away with the need for a national board to decide who gets cash reimbursements, and depending instead on free market dynamics, the proportional registration system may be more acceptable to those who worry about the political implications of having governments decide what rewards to give to what drug companies.

-Karl Fogel

I’m not convinced that any patent system at all is necessary, by the way, but the issues with patents are a bit trickier than with copyrights.

Patents are partly a means of preventing people from keeping secrets: if someone invents a new artificial heart valve, we want them to publish about it in great detail, and granting them a temporary monopoly as a reward for doing so is one way to ensure this happens. On the other hand, medicine and medical devices are almost always the example the patent industry uses when it wants to scare us into imagining a world devoid of innovation, so it’s appropriate to note that there’s a whole separate secrecy-prevention mechanism in place for that category of inventions: the medical approval process (in the U.S. this is run by the FDA, for example). You can’t get your drug or heart valve approved anyway without revealing the technical specs, so the anti-secrecy argument is rather weak in the very case of the poster-child industry for the pro-patent lobby, as it happens.

While I’d be very happy to see the proportional registration system adopted for either patents or copyrights, the real purpose of the proposal is to show that even if you accept the argument that monopoly-based market dynamics are necessary, there’s still a better way to do it than the way we do it now. I’d really like to see on what grounds the pro-monopoly lobby would argue against the proposal… They often talk about how we must “balance” the needs of the creators against the needs of the public (a false choice if ever there were one), but if balance is the desired goal, what could be more balanced than a system where the public gets a buyout option based on the owner-determined market value?

If you haven’t read The Feud in this Tuesday’s New York Times “Science Times” section, do have a look. It’s worth it just for the unintentionally self-damning quotes from the two heart surgeons involved, Dr. Denton A. Cooley (now 87) and Dr. Michael E. DeBakey (now 99), who have apparently been engaged in a 38-year-long feud over the circumstances surrounding the first implantation of a fully artificial heart in a human.

However, my favorite part of the article had nothing to do with the feud:

Dr. Cooley recalled that a lawyer had once asked him during a trial if he considered himself the best heart surgeon in the world.

“Yes,” he replied.

“Don’t you think that’s being rather immodest?” the lawyer asked.

“Perhaps,” Dr. Cooley responded. “But remember I’m under oath.”

The New York Times probably didn’t fact-check that, since they’re just transcribing a quote: for the purposes of the piece, the important thing is that Cooley told the story, not whether it’s true. But court documents are public records, and it would be nice if someone were to track this one down. If it’s real, then it’s a verifiable instance of an anecdote I first read years ago in The Little, Brown Book of Anecdotes, edited by Clifton Fadiman (a wonderful book that is, oddly enough, neither little nor brown — it’s just published by Little, Brown & Company):

ROWLAND, Henry Augustus (1848-1901), US physicist, professor of physics at Johns Hopkins University (1875-1901). He laid the foundation for modern spectroscopy.

1. Professor Rowland was summoned as an expert witness at a trial. During cross-examination a lawyer demanded, “What are your qualifications as an expert witness in this case?”

“I am the greatest living expert on the subject under discussion,” replied the professor quietly.

Later a friend, well acquainted with the professor’s modest and retiring disposition, observed that he had been amazed to hear him praise himself in this way; it was completely out of character. Rowland asked, “Well, what did you expect me to do? I was under oath.”

(This anecdote is also told of others.)

I have to admit, my instinct is that Cooley just appropriated this old chestnut for himself, and that it never actually happened (to him). After all, what expert wouldn’t fantasize about finding themselves under oath for such a question? But I’d be pleased to discover that I’m wrong and that it really took place.

Last July, at OSCON in Portland, Oregon, I put a whiteboard in the main conference hallway (with help from the indefatigable Vee McMillen), and wrote

Call For Software: Tools We Wish We Had

across the top, dividing the rest of the board into a grid of blank cells. Our hope was to get an ad hoc brainstorm on what tools open source developers feel the world is missing — anything at all, not necessarily just development tools.

By the end of the conference, the board looked like this:

OSCON CFS whiteboard, at the end of the conference.

(One reason I waited so long to post this image was that I’d wanted to transcribe the board first, but of course never found the free hour or two… Then I got sane and realized that if it were posted, either I could transcribe it or anyone else could. Duh. At Eric Hanchrow’s suggestion, I set up a wiki and a bunch of people pitched in to finish the transcription (thanks everyone!). Then at some point after that that whole wiki installation got completely spam-bombed and had to be taken offline. Then years later, in late 2015, I recovered the entry from the still-offline wiki’s database, converted it to HTML, and posted it here:

  • HIERARCHICAL CHECKLIST MANAGER: X-PLATFORM, PYTHON IF POSSIBLE [someone wrote “This is an OPML editor” on the side later]

  • LDAP auth that matches Distos & Samba [addition: commercially from Centrify]

  • GUI builder for XUL [addition: +2]

  • Ultra light-weight memcache style queue, with cross language client libraries [addition: twitter wrote one, ask them for it]

  • dtrace for Linux or something similar
    [addition 1: (LIKE SYSTEMTAP? — jdub)]
    [addition 2: (NO, LIKE dtrace — VLAD)]
    [addition 3:(like systemtap, but not crap)]

  • Unified blog/wiki/forums mailing list publishing system [addition 1 (partly illegible): it’s called su??two ] [addition 2: trac] [addition 3: drupal.org]

  • Online readonly running example machines
    ps, cat, etc just no writes
    look at a well setup “fiesty” for example [addition: (See FreeOSZoo live)]

    [addition 1 : read the swe books from ms press. yes, really!] [addition 2: all of them?]

  • Greasemonkey plugin for untrusted (conference) wifis that warns about sending passwords in clear. [addition: If you don’t mind being warned for all forms, in Firefox set: security.warn_submit_insecure to ‘true’ in about:config]

  • A URL syntax that lets me deep link to specific text in a page without an anchor there. E.g.: http://…/foo.html@”string” would jump to “string” in foo.html (& highlight it?). [first note says: “Do as Firefox plugin, other browsers will copy…”] [second note: “It’s called XPointer”] [third note: seems to say “jqFrag”, maybe?]

  • Shared calendaring that doesn’t suck! [note at bottom: “Don’t count on it. This is a deceptively hard problem.” Follow up says “Chandler06 solves this”] [around the original proposal, a chorus of four “+1″s has been written] [there’s a pointer to a domain, but it’s hard to read: something like “bun1.org” or “burl.org” or “bum1.org” or … ? None of the domains I tried resolved to anything interesting. If anyone knows what this is, please clarify.]

  • Open sensor / actuator code [line break] start with good camera capture [line break] need to move robotics along

  • Mailing list software that lets you subscribe to individual threads instead of just to the list. [many follow up notes: “grokbase.com!”; “It’s called USENET” (respondent asks “??? How?”); “Or Google Alerts” (respondent says “not reliable”, further respondent says “But I want to _post_ too… real subscription”); “gmane news gateway”]

  • Free / Open Source Music Typesetting Software that non-technical people can use (e.g., like Finale or Sibelius) [later addition: “Lilypond?”] [still later response: “Last time I tried Lilypond, it was not ready for non-geek users. Has that changed?”] [further response: “No :-(“]

  • A tool to surf Band and Myspace sites (w/o RSS 🙁 ) for tour date info. Bonus points for filtering on city!

  • CLIENT AND SERVER INTEROPERABLE WITH M$ EXCHANGE TO RECLAIM ENTERPRISE SHARED CALENDARING AND MESSAGING [entire entry has been cleverly surrounded with an HTML anchor tag whose href points back to row 2, column 4 “Shared Calendaring that doesn’t suck”]

  • Document management workflow systems — XML based [someone has added “JBEM?” after this]

  • Blog reflector: A blog entry is just a post with comments — a thread. Let that thread be displayed at as many sites (in as many themes) as people want, & let it ”accept comments” from all those display sites. When I write a blog entry, I don’t want to be forced to choose where to run it, nor risk dividing comments among multiple sites. ”One Thread, Many Doors” should be the way the world is. (Can RSS do this?) [someone has scrawled a “+1” in the upper left corner of this entry]

  • Image processing sw to clean up camera pics of whiteboards (how much will you pay me to write it?) [transcriber’s note: hah hah, very funny]

  • configfile + parsefile(OAwiki?); load html file with links for keywords; “in situ config explainer” coar{_AT_}apache.org

  • TESLA power transmission wireless 801c

  • Stable kernel DDI for Linux

  • ADDRESS BOOK IN MOZILLA/Thunderbird [someone has written “Tools⇒Addressbook”, but then someone else crossed that out in green ink and drew an arrow pointing over to row 5, column 4, which simply says “that doesn’t suck”, presumably a generic destination qualifier for many of the entries on this whiteboard!]

  • Good open source schematic capture/layout tool

  • OSS CAD PROGRAMS [begin bulleted list] CAD, PCB, OSS Spice [end bulleted list] … that’s better than gEDA. [someone has written “+1” for this whole entry]

  • A proper MPEG GUI editor [someone has written “+1” for this whole entry]

  • Universal compress/decompress tool that doesn’t make me remember “xjvf” and stuff. Call it “expand-dammit”.
    [$50 bounty added by Jacob jacob@…something-unreadable… (suspect Jacob Kaplan-Moss, though)] [Later note says: “Archive::Extract — just wrap a CLI around it”] [Last note: “Contact gerv{_AT_}gerv.net for a suitable script. I claim the $50 :-)” ]

  • Test case mgmt/tracking, repository, coverage reports [line break] SVN commit ⇒ bug ⇒ feature ⇒ requirements tracking. [someone has written “Maybe Trac? Or ReleaseMonkey?”] [someone else responded “Maybe Teotopia for the Mozilla project?”]

  • Something to make you zealots less f-ing religious about Linux. [someone has added in different ink “and Git!”]

  • [drawn heart symbol]

  • [the first word is clearly “virtual”, the second seems to be a pair of words, but I can’t quite tell what they are. “man slaves”? “main staves”? “man sraves”?]

  • [an asterisk followed by the phrase “that doesn’t suck”; this is a form of whiteboard compression, whereby a common phrase from other cells has been abstracted to this cell and referred to from elsewhere. Except that many of the other cells still say it too.]

  • Webmail that doesn’t suck and isn’t tied to a specific provider. [people have added: “Zimbra?” and what looks like “Bongo” or “Bovgo”]

  • A MOBILE PHONE APP THAT DOESN’T SUCK. [addition in green ink: “Yes!”]

  • OCR software that can take a digital image of all the shelves in a bookstore or library and read the titles/authors off the spines. Then you not only have a catalog of the books, you have a locator and theft-detector as well.

If you’re considering starting an open source project, there are a lot of good ideas on that board; have a look. Even if none of them quite fits the bill, one might push your thinking in a new direction.

Of course, earlier in the conference, the board wasn’t quite as… shall we say… constructive:

OSCON CFS whiteboard, at the beginning of the conference.


What finally motivated me to make this post was receiving the following mail from Greg Wilson, reprinted here with his permission:

From: Greg Wilson
Subject: student projects
To: Karl Fogel
Date: Sun, 4 Nov 2007 13:52:12 -0500 (EST)

I'm running a combined grad/undergrad course on software 
consulting next term (Jan-May'08), and need to find projects
for 25 to 35 bright, hard-working programmers, each of whom
will spend about 120 hours on it. I want students to work in
pairs or triples (so that they have someone local to bounce
ideas off); I also want the projects to be open source (so
that students can talk about/show off their work) and to have
real customers (people outside the CS department); it's a
bonus if those customers are in Toronto for face-to-face
meetings, but not essential.  If you have something, please
let me know.


So: anyone need some eager student programmers for an open source project?

I received the oddest spearphishing attack the other day. At least, I’m pretty sure that’s what it was, though can’t be 100% positive. Here’s the correspondence, with the name changed slightly to protect the innocent (if she is innocent, which I highly doubt).

From: "Marina Mitropoulos (ABCTours)" <express@abctours.gr>
To: <kfogel@red-bean.com>
Subject:  urgent!
Date: Thu, 25 Oct 2007 13:43:07 +0300

Dear Mr Fogel,

I would like your help in a very serious matter.

Recently, my boss received an unknown senders email referring to me
and accusing me for many terrible things that are not true and my job
place is in jeopardy right now.

I need your help to find this persons Id or even the password of his
email. I want to find out who’s this person that is trying to ruin my

The email that this message came from is: gianluigi.farina@alice.it

Will you please help me?

Im waiting for your kind reply,

Thank u


Now, I didn’t read that original mail when it first arrived. When email from an unknown sender has the subject line “urgent!”, I don’t even consciously process it as spam anymore — a couple of neurons somewhere in my brainstem take care of hitting the Delete key, while I go on to read the next subject line in my inbox.

But then came another message from her:

From: "Marina Mitropoulos (ABCTours)" <express@abctours.gr>
To: <kfogel@red-bean.com>
Subject: KNOTSPAM  Hi, Karl, long time no see...
Date: Thu, 25 Oct 2007 15:57:12 +0300

    Dear Mr Fogel,

    [...the rest is the same as the original...]

Whoa. Only a human could have sent that, because it had the “KNOTSPAM” marker signifying that the sender has read my web page explaining how to send me email that won’t be mistaken for spam. So she was real, and she was trying to talk to me in particular.

(Note the “Hi, Karl, long time no see…” in the new subject line, by the way. This was a lie: as she later admitted, she was a complete stranger.)

Now somewhat intrigued, I replied:

From: Karl Fogel <kfogel@red-bean.com>
To: "Marina Mitropoulos \(ABCTours\)" <express@abctours.gr>
Subject: Re: KNOTSPAM  Hi, Karl, long time no see...
Date: Fri, 26 Oct 2007 13:44:36 -0700

Do we know each other?  What makes you think I can help with this?

When I received your first mail, I assumed it was spam.  But then I
received your second mail, with the "KNOTSPAM" marking in the subject
line, which means that you read my web page and figured out how to
send me email.  That was a surprise.  But I don't recognize your name,
and I have no record of ever having exchanged email with you before.
If we are acquainted, I apologize -- I do not have a good memory for

Who are you?


She replied a day or so later:

From: "Marina Mitropoulos (ABCTours)" <express@abctours.gr>
To: "Karl Fogel" <kfogel@red-bean.com>
Subject: Re: KNOTSPAM  Hi, Karl, long time no see...
Date: Mon, 29 Oct 2007 10:20:58 +0200

Dear Karl,

No, we do not know each other, I work for a travel agency in Greece,
my name is Marina Mitropoulos and as I wrote you in my previous email
I am in a jeopardy to lose my job because of some idiot that is trying
to make me look very bad at my boss's eyes.

Unfortunately, this person did a good job by sending all he/she wanted
through a free account email from a foreigner provider.

I don’t know to whom to turn to find out the truth, I’m not
interesting in to read his/her emails, I only want to find out from
where (country/area) this account was opened and if there's any real
name given, or if nothing of the above if I can at least read some of
other emails and try to understand to whom it could belong to by the
way of writing..(I hope you understand what I’m trying to say)

I understand that this could sound a bit unorthodox to you, but If you
can help me I would very much appreciate it.

I have tried even with a private investigator but here in Greece
things are not so easy to find out a thing like that as it is in
united states…

I have searched the internet for trying to find anything I can on my
own, but I’m clueless with these things and I only end up paying some
stupid site for promising me to find it and at the end they couldn’t
even find my work emails details… anyway…

I deeply apologize if I caused you any kind trouble, it was not my
intention,  I only need help if you can please...

If you wish to contact me here's my phone nr +306974301136

I thank you again even for reading this email and for responding to it…

I m waiting your response,

Thank u again.

Marina Mitropoulos

Hmmm, a phone number! Why, I remember when I couldn’t pry that out of a woman at a bar for all the charm in the world… and now they’re throwing them at me by email. I love the Internet! Just kidding. Here’s how I responded:

From: Karl Fogel <kfogel@red-bean.com>
To: "Marina Mitropoulos \(ABCTours\)" <express@abctours.gr>
Subject: Re: KNOTSPAM  Hi, Karl, long time no see...
Date: Mon, 29 Oct 2007 10:25:37 -0700

I can't help with your problem.

But I'm fascinated that you picked a complete stranger at random on
the Internet to ask for help.  That seems very, very odd.  I don't
think it's likely to solve your problem.  Most strangers would be
suspicious that maybe you have some other motivation.  For example, if
someone's telling lies about you on the Internet, then it's just as
easily possible that you are telling lies about yourself: from an
outsider's point of view, neither one is more likely than the other!

If she’s a spearphisher, she’s going out of her way to keep her cover. She stayed in character to reply:

From: "Marina Mitropoulos (ABCTours)" <express@abctours.gr>
To: "Karl Fogel" <kfogel@red-bean.com>
Subject: Re: KNOTSPAM  Hi, Karl, long time no see...
Date: Tue, 30 Oct 2007 10:01:33 +0200

I see your point and I totally understand... and in this case there is
no point for me to try convince you otherwise.

Just informational, I dont have any motivation to get into this email
and snoop around for fun, I need it to find out who is trying to hurt
me… and this action to “picked up someone random” on the internet , it
shows how really desperate I am to find out the truth.

I thought you could do something to help, but I see now how silly it
might seems to you all this and Im very sorry I disturbed you but I
thank u deeply for even replying to my mail.


I have no idea what to think now (other than I love the Internet!, of course). Searching for her real name gets exactly one hit, on a web page at the same ostensible Greek travel agency as her email address. On the off chance that she’s telling the truth, I’ve changed her name, and that of the travel agency. But not her number: if you want to get in touch with her, go for it, and good luck!

I like this photograph because the light and image somehow seem to capture the sound of the bluegrass jam at the Atlas Cafe in San Francisco last Thursday. The photograph was taken by Wilfredo Sánchez Vega, and the fellow at the center — the one most in focus, the one for whom the light rays seem to part, the one with, yes, the banjo — is none other than Ben Collins-Sussman, Subversion developer extraordinaire and well-known speaker on topics open-source. He also played a darned good solo that night!

Atlas Cafe bluegrass jam

Lee Bollinger has gotten a lot of criticism for his extremely harsh introduction of Mahmoud Ahmadinejad, the president of Iran, before a Columbia University audience on September 24th. Most of the criticism has been that it was a rude way to treat a guest. I’m not sure I buy that one: a politician coming to give a speech is not quite the same thing as a guest in one’s home, and anyway Bollinger telegraphed quite clearly in advance that he would not be sparing in his remarks; Ahmadinejad could have cancelled if he were worried about how he would be treated. (For what it’s worth, Ahmadinejad’s own speech contained passages so nonsensical as to be indistinguishable from the sort of insane mumblings that usually cause passers-by to cross to the other side of the street. No, really — check out the transcript. That dude’s a looney!)

Bollinger started his remarks with a detailed explanation of his motives, so there would be no mistaking what he was trying to do. Good for him: he tried to do the right thing. I actually think most of what he said was justified, if a little over-the-top and needlessly ponderous in tone. He made Will Franken very happy, anyway.

Nonetheless, I think Bollinger muffed it. He elevated Ahmadinejad’s importance by attributing Iran’s policies to its president. But everyone knows Ahmadinejad isn’t running the show over there: the mullahs are. When Bollinger said to Ahmadinejad “you exhibit all the signs of a petty and cruel dictator”, he was exactly wrong. Far from being a dictator, Ahmadinejad is barely a president. He was elected from a filtered set of acceptable candidates (anyone objectionable to the religious council was not allowed on the ballot), and his every move is supervised by those who really run the show. Ahmadinejad is a figurehead; it’s doubtful he could have gotten elected in a truly open field. In a way, it hardly matters that the President of Iran is an incoherent conspiracy theorist with at best a shaky grasp of science and history — even were he a savant, he would effect little good, so perhaps it’s better that he’s an obvious joke.

What I wish Bollinger had asked was “Mr. President, why did you ally yourself with the perversion of democracy by standing for office in an election in which many legitimate candidates were not allowed to run? Do you believe in democracy, and if so, will you publicly call on the religious leaders of Iran to allow truly open and fair elections next time?”

Instead, Bollinger treated Ahmadinejad like he matters, which is exactly what Ahmadinejad wanted. It will probably give him a boost in the polls back home, too. Sigh.

Surprisingly, the verb “wikipudiate” does not seem to exist yet. At least, a Google search turns up no hits. I think I’ll add it to the glossiary:

Wikipudiate: (v) To refute someone by referring them to a Wikipedia article. Ex: “I’ve been wikipudiated!” Alternatively, to deny an assertion of fact on the grounds of its provenance in Wikipedia. Ex: “You’ve been wikipudiated!” (wikipudiation, n)

This Friday I attended a productive and unexpectedly fascinating workshop at UC Berkeley, aimed at developing a model policy for archives that host images — often disturbing and graphic ones — from war zones and other areas of violent conflict. In particular, the Iraq war is generating a lot of such footage (it’s the first war in which many of the participants carry portable video cameras), and archivists understandably feel conflicting responsibilities about how to handle the deluge of material. On the one hand, they don’t want to be in the position of helping terrorists disseminate recruiting material, nor do they want to further hurt the feelings of family members of those whose deaths are shown in some of these videos and still photos. On the other hand, they’re archivists, after all: their mission is to preserve and make available all the information they can, not to judge what might or might not be valuable to future historians.

The meeting was organized by Jeff Ubois, who has worked with the Internet Archive and other repositories of digital images. The Internet Archive has to deal with these kinds of questions every day, and wanted to stop making their decisions ad hoc and in isolation from other archivists in similar situations.

Peter Brantley of the Digital Library Federation has already written a detailed blog entry about the workshop, I won’t repeat here his excellent description of the discussion and its conclusion. But I would like to talk a little about why the issue is so complex, and how it is that sixteen people can spend seven hours in a room debating just this one topic, in the end producing just a 67-word draft policy that basically says “We’ll take feedback from users, we’ll try to do the right thing, and we’ll try to err on the side of more access rather than less”. I think it was a good outcome, too: this was one of those rare instances when one can say that the process was more important than the result and really mean it.

I started out the meeting thinking “Well, obviously if a family member or loved one asks an archive to take down footage of their child being tortured or blown to bits, that’s an easy call: take it down. How important can it be, after all, that it’s worth adding insult to those already injured so much?” Annalee Newitz of Techsploitation, while understanding that sentiment, responded that the historical record has a value too. That footage might be the only footage of some important event (say, the equivalent of a My Lai massacre, though Annalee didn’t use that example), and it’s simply not good enough to have a policy that takes into account how much someone will be hurt, without also taking into account the intrinsic worth of archival materials to future historians and researchers.

That really got me thinking. What is the intrinsic worth of these kind of materials?

A person who grows up in a war zone — say, a child growing up in Baghdad today — may see things on a weekly basis that others never encounter in their entire lives. The child in a war zone does not “see” them on a computer screen, but in life: in the street, on the way to school or to a friend’s house. But even an image on a computer screen is much closer to real experience than a description would be. One thing this meeting made very clear (partly via the small but effective selection of images sent around to prepare participants for what they would be dealing with) is that a verbal description of an image simply cannot have the same impact as the image itself. I’ve heard gory descriptions of public beheadings before, but what will stick in my mind forever is certain video footage I saw exactly once. It’s not merely that it’s more graphic than a prose description; it’s a different level of sensory experience altogether.

I think this implies that there may be no way to understand what that child in Baghdad is going through except through seeing what’s actually happening there. Thus there may be no way to understand the adult that child will become, except through these images.

A decade or two from now, we may be wondering why it is that Iraqis have chosen (or at any rate acquiesced to) a new tyrant of more or less the same type as Saddam Hussein, who was himself more or less the same type as Joseph Stalin and many others throughout history. If that day comes, the footage we have from Iraq today will go a long way toward helping us understand how it happened. Even today, a number of people in Iraq are starting to say that as bad as it was under Saddam, it was better than now. I realize it may not be a majority — and I hope it never becomes one — but that even a statistically significant portion of people would say that should be a clue that something has gone badly wrong. And I’m not sure how a historian would get inside their heads, now or in the future, without access to images of what those people were exposed to. Interviews with foreign correspondents and “embedded journalists” won’t cut it. Even interviews with natives won’t do the job. You have to come as close as you can to the real experience, and for that, these archives are the best we have. We censor them at the risk of our own understanding.