Dear Aunty Slashdot,

Like many people, my wife and I have digital cameras that can record short mpeg video files in addition to traditional jpeg stills. Like any still camera, taking photos with the camera held vertically is a perfectly conventional thing to do, if the subject matter being photographed would be better framed that way. Caught up in the moment though, we've also got some video files that were shot this way, and fixing these is proving to be much harder to correct. Does anyone know of a good, relatively painless way to rotate video files so that they're right side up? As video-capable digital cameras become more common, this is a feature that I'd assume an increasing number of people will want.

I'd prefer some kind of freeware approach to this, but so far haven't found anything that seems like it will help. It seems like ImageMagick might be the most promising tool, if I can get the mpeg2vidcodec_v12 plugin working on my Mac (lots of make test errors so far...), but even then will it be as simple as a convert -rotate 90 > video.mov ? So far, I can't even get to that point with the IM toolkit. CinePaint (nee FilmGimp) didn't seem to want to open *.MOV files to begin with, which confused me as I thought that was the whole point of CinePaint. I've also looked into mjpegtools, mpgtx, VirtualDub and TMPGenc, but none of them seems able to do rotate the contents of video files. I was able to open a sideways video file as a series of hundreds of separate still images in Adobe ImageReady, but even with that program's automation tools (and my admittedly shaky grasp of how to use them), rotating them all & stitching it back into one file seems like it'll be annoying. I've also tried Apple's iMovie, but it seems to be geared towards stitching together a collection of video clips rather than manipulating the contents of any given clip in any significant way. I don't have any other commercial software available, and am not that interested in shelling out possibly hundreds for the kind of "pro" software that might work but would be overkill for my usual needs.

As an added bonus, it would be nice to be able to convert individual frames to JPEGs for making thumbnails, or ranges of frames into low-resolution GIF/MNG animations. I have a hunch that the ability to do that may fall out of any solution to the bigger problem, so I'm putting off worrying about this for now, but would like to be able to do it eventually.

Does anyone know of a good way to rotate video files? I realize that the proportions of the converted file will be "wrong", but I don't care -- they're low resolution files meant only for viewing on my computer or maybe a web page, and if I ever want to put the files on a television screen then I can just put up with the vertical letterboxing. So far, the only approach that seems to have any traction at all is to find a way to treat the file as individual frames, rotate one by one, then stitch it back together -- but that seems annoying, particularly if the file also has an audio component that has to be kept track of. Still, for lack of tools to do it any other way, that's the best approach I've been able to come up with. Can anyone suggest something better?

Dear aunty Slashdot,

Does anyone here understand how Mozilla / Firebird's current security module system works? In particular, does anyone know what's up with the "software security device"?

My fiancee's computer -- a WinXP laptop with no user account passwords (it's just two of us using it, and we trust each other) -- keeps throwing these annoying dialog windows demanding that you "Please enter the master password for the Software Security Device." whenever you take Firebird to a web page with a username & password.

The catch though is that no password I can think of as a likely candidate works. A bit of Googling points to a couple of semi-promising solutions, and while all the ones I've found so far talk about Linux, the general description of the issue seems to be spot on. The workaround -- enter the Linux login account -- doesn't seem to apply here: there is no Windows system login for this account, and leaving the password field blank doesn't work either.

Following on from the Mandrake advice, I tried opening up Firebird's dialog window for the security device settings (go to Tools -> Options, then Advanced -> Certificates -> Manage Security Devices [there's a disclaimer that this is subject to move around in future releases]). This brings up a cryptic dialog window with the "Device Manager" (yay! trusted computing IN OUR TIME), with a hierarchy of cryptically labelled "Security Modules and Devices" on the left (e.g. NSS Internal PKCS #11 Module -> Software Security Device), some cryptic "details" and "values" in the middle panel, and a column of cryptic buttons over on the right. (For a crypto system, they've got being cryptic nailed :-/ ).

With those right-side buttons, three seem to do with managing what appears to be the equivalent of OSX's Keychain ("Login", "Change Password", and "Load"), but again if you click on any of those you get asked for the master password -- the lack thereof being the rabbit I'm chasing down this hole. There's also a button labelled "Enable FIPS", but there seems to be no indication of what happens when you click it or what FIPS stands for (if in fact it's an acronym in the first place).

Hilariously, there's also a "Help" button on the bottom of the dialog, but it doesn't seem to be hooked up to anything. Har har har.

----

So, the QUESTION:

Where did this thing come from, and how can one either fix or disable it? If it's like Keychain, and provides some kind of encrypted safekeeping for sensitive form data, I have no problem with doing it "right" and working logged into the subsystem. As it is now though, it's just getting in the way, and I can't figure out how to reliably get it to go away and stay away.

I say "reliably", because on some sites I get the dialog almost every time I follow a link, while on others it's just at the initial login -- I assume that this has to do with how accounts are being managed on the server, but haven't been ablle to pin down what's going on there. One annoyance per site I could deal with, but repeating it all the time like this is really getting on my nerves...

Any help wins an ice cream cone -- TIA :-)

So, much to the delight of every liberal in America, Rush Limbaugh has been caught in a wonderful pair of fuckups this week -- blatant racism, and abuse of a drug that can cause deafness (which it just so happens he came down with a sudden case of a year or two ago -- curious, eh?).

The drug article ran in my new favorite magazine, the National Enquirer. The online version of the Enquirer article doesn't have all the details, but I was thumbing through the paper version in the supermarket yesterday, and it has a copy of an email he sent to his dealer in Florida.

Apparently, Rushie's email address is <rprivate@eibnet.com>.

Again: Rush Limbaugh's private email address is, according to that bastion of journalistic integrity second only to Slashdot, <rprivate@eibnet.com>.

Not that I'm advocating that anyone abuse that nugget of information or anything -- that would be *wrong*, and only commie pinko liberal traitors would want to get this Great Natural Treasure upset -- but if anyone thinks of anything amusing to do with that nugget of information, this commie pink liberal traitor would be your Friend Forever.

According to Macworld UK, Be, Inc. has settled its antitrust suit against Microsoft for $23 million. Microsoft, typically, admits no wrongdoing in the settlement. Readers may recall that Be, maker of the BeOS operating system, brought their suit against the Microsoft back in February 2002. At the time this suit was brought, it was becoming obvious that the US government's antitrust suit against Microsoft was not going to result in any significant punishment for the convicted monopolist. Some observers felt Be's claims that Microsoft's vendor contracts excluded competitors from the market was a stronger case than the browser bundling aspect that the US department of justice pursued, but in the end it seems that Be no longer had the resources to complete the trial. With this case abandoned, the best hopes for a remedy to the Microsoft monopoly now seem to be in the European courts, or with a possible regime change in the USA in 2005.

It looks like the staff at E-Commerce Times have come up with a wonderful new logo for SCO articles, as scaled down to icon size by Google News (and I've stashed a backup of, just in case).

It shouldn't be a Caldera logo anymore anyway. I think a picture of someone shooting themself in the foot is much more apropos.

...now that I think about it, the better thing to do would be to edit the Monty Python foot used for the humor articles, and add firearms as needed. Thus binding in the whole tragi-comic nature of the situation in one fell swoop.

This story is hilarious: Nascar fan gets pissed that a Red Sox game preempted his Nascar races, so he fires off half a million emails to the Boston FOX affiliate in order to make his displeasure clear.

Never underestimate the raw fury & power of a Nascar fan that has, apparently, been taught Perl somewhere along the line... :-)

Here's the article as Sports Illustrated is running it currently:

Pass the spam Fan sentenced for sending FOX half-million e-mails

Posted: Tuesday August 12, 2003 6:48 PM

BOSTON (AP) -- An angry NASCAR fan who flooded FOX Entertainment with e-mails after a Boston Red Sox game pre-empted an auto race has been sentenced to six months of home detention.

Michael Melo, 45, of Billerica, paid $36,000 in restitution before his sentencing and also will have to pay a $2,000 fine.

He pleaded guilty in May to a federal misdemeanor charge of damage to a protected computer system. He admitted he wrote a computer program that fired off more than a half-million e-mail messages to WFXT-TV 25 in Boston after a Red Sox game was aired instead of a NASCAR race in 2001.

The e-mail messages were automatically forwarded to FOX-25's Los Angeles parent company, FOX Entertainment. The network, thinking it was a hacker attack, shut down Internet communications with the affiliate and was forced to spend about $36,000 to clean up the site.

Doesn't this story have some kind of parallel, involving email floods and the Red Sox?

Yes it does:

Boston Perl Monger Plays With the Big Leagues

Posted by Hemos on 1999.07.07 9:53
from the dept.

babbage writes "Chris Nandor tried to help out the Red Sox by stuffing the MLB All Star Ballot. Baseball says they were on to him all along. He says he'll be back next year. " Ah, yes, that wonderful point where sports and computer connect, probably sans changing IPs, but with the mis-use of "hacker".

Of course, that "hacker" went on to become a Slashdot employee, so Mr. Melo may yet have a bright future ahead of him :-)

I have a loosely organized directory tree of photos on my site. I'd like a nicer way to present these than simple Apache directory listings, but I'd rather reuse existing code (even if possibly modifying code, templates & stylesheets) than start from scratch. I've poked around a bit, but nothing I've seen seems to quite meet the rough criteria I have in mind. For people that have photo galleries on your sites, are there good solutions available from somewhere like CPAN or Sourceforge?

Ideal functional criteria that I'm looking for:

• Easy, automatic. If I transfer a new directory of image files from my camera to, say, /web/htdocs/photos/2003/08/09/ (or something more descriptive like 09_roadtrip or 09/roadtrip etc), then this should become visible at an associated url with no additional setup work required, complete with thumbnails, titles, maybe an image resize feature, and other page decoration.
• Listings should allow for visitor comments. If I put up a directory with 100 photos from a relative's wedding, I don't want to be forced to fill in the details for what every image depicts -- but if visitors want to add that that's fine with me.
• Categorization: There should be an optional way to add metadata, so that for example if I or a visitor indicates that a given picture is of a certain person, place, or event, then visitors should be able to browse or search within that category. Ideally, index level pages should be aware of the categories associated with images in that section, and users should be able to follow links back & forth between (say) a directory hierarchy focused organization of links and a category focused organization.
• Something able to run mod_perl1 would be ideal. PHP would be tolerable if the package were very nice (I might as well learn PHP sooner or later...), but given the choice between a "C+" mod_perl application and a "B+" PHP one, I'll probably go with mod_perl. Anything written in Python / Ruby / TCL / Scheme / Bourne / Zope / compiled-C would get a silly look but not necessarily be ruled out. Anything in ASP or JSP is straight out.
• I'm not worried about most internal or external dependencies. That is, having to pull in modules from CPAN is fine with me, as is depending on Imagemagick. If it uses a templating system (Mason, Template Toolkit, or HTML::Template) that's great with me; if layout is done with CSS/XHTML1+, that's even better. I don't mind if the application depends on an external MySQL or PostgreSQL database, or some other storage layer (DBM, XML, text files, etc).

Most of the ones I've looked at seem okay, but not overwhelmingly so. The comment feature, which really appeals to me, doesn't seem to be available in most of the packages I've looked into. The main exception I can think of is a Movable Type based photoblog, which does well with comments but is cumbersome in other ways -- images have to be added one at a time, and as far as I can tell only the site owner can do anything with categories or other metadata.

The best photo sites I can think of are on a couple of different people's personal sites, but it isn't clear to me if the examples I've seen were custom written or if there's some great, uncredited gallery toolkit out there that these people are taking advantage of.

Here's some of what I've looked into so far:

• I've tried setting up a photoblog with Movable Type, and this is okay, but it seems like you have to add images one at a time. Additionally, the tutorials I've seen so far seem to revolve around uploading images through the browser, but I have access to put them there via scp or similar. Whether or not that's a real constraint of MT photoblogs, it still seems like the intended workflow is backwards from my point of view: select an image & offer it up for review & comment, rather than put up an array of images & allow people to comment on any they happen to find interesting. While the MT one I've tried seems okay for the occasional exceptionally good image, organizing everything this way is much more work than I have in mind here.
• I've given Apache::Gallery a try, but from the sample sites linked from http://apachegallery.dk/, none seem to be much fancier than the file browser in Nautilus or recent versions of Windows Explorer: it's still mainly a list of directories & thumbnails, with some metadata, but no categories, no comments, etc. Mason Gallery, as demonstrated at http://bushong.net/pics/ and distributed at http://bushong.net/dave/sw/, seems to have the same default qualities & limitations: prettified directories but no comments, categories, etc.
• Randal Schwartz's site at http://www.stonehenge.com/merlyn/Pictures/ looks nice (if a bit more "Star Trek: The Next Generation" than I would have gone for, but no matter :), but it doesn't seem to have categories or comment mechanism. Paul Mison's "Stem" site at http://husk.org/pics/ looks pretty nice, and seems to have the functionality I want: thumbnails, categories, and comments. However, I can't tell how automatic everything is: the thumbnails are tidily cropped squares from rectangular originals, and most (all?) of the images have the title superimposed somewhere in the image. Both of these neat features seem like they'd have to be at least partly manual to me. In any case these two sites are the property of their owners, and I don't know if either of them intends to share the underlying code.

So, of those that have photo sites, what publically available tools work well?

Thanks for any suggestions :)

Darl McBride's "surprise" at RedHat's suit is almost like a scene out of Dr Strangelove...

[The President calls the RedHat Premier.]

President Darl McBride: [to RedHat] Hello? ... Ah ... I can't hear too well. Do you suppose you could turn the music down just a little? ... Oh-ho, that's much better. ... yeah ... huh ... yes ... Fine, I can hear you now, Dmitri. ... Clear and plain and coming through fine. ... I'm coming through fine, too, eh? ... Good, then ... well, then, as you say, we're both coming through fine. ... Good. ... Well, it's good that you're fine and ... and I'm fine. ... I agree with you, it's great to be fine. ... a-ha-ha-ha-ha ... Now then, Dmitri, you know how we've always talked about the possibility of something going wrong with the Suit. ... The Suit, Dmitri. ... The legal suit! ... Well now, what happened is ... ah ... one of our legal staff, he had a sort of ... well, he went a little funny in the head ... you know ... just a little ... funny. And, ah ... he went and did a silly thing. ... Well, I'll tell you what he did. He ordered his staff ... to attack your industry... Ah... Well, let me finish, Dmitri. ... Let me finish, Dmitri. ... Well listen, how do you think I feel about it?! ... Can you imagine how I feel about it, Dmitri? ... Why do you think I'm calling you? Just to say hello? ... Of course I like to speak to you! ... Of course I like to say hello! ... Not now, but anytime, Dmitri. I'm just calling up to tell you something terrible has happened... It's a friendly call. Of course it's a friendly call. ... Listen, if it wasn't friendly ... you probably wouldn't have even got it. ... They will not reach their courts for at least another year. ... I am ... I am positive, Dmitri. ... Listen, I've been all over this with your legal representative. It is not a trick. ... Well, I'll tell you. We'd like to give your legal staff a complete run-down on the complaints, the allegations, and the insinuations in the lawsuits. ... Yes! I mean i-i-i-if we're unable to recall the suits, then ... I'd say that, ah ... well, ah ... we're just gonna have to help you destroy them, Dmitri. ... I know they're our boys. ... All right, well listen now. Who should we call? ... Who should we call, Dmitri? The ... wha-whe, the People... you, sorry, you faded away there. ... The People's Free Software Foundation. ... Where is that, Dmitri? ... In Boston. ... Right. ... Yes. ... Oh, you'll call them first, will you? ... Uh-huh ... Listen, do you happen to have the phone number on you, Dmitri? ... Whe-ah, what? I see, just ask for Boston information. ... Ah-ah-eh-uhm-hm ... I'm surprised, too, Dmitri. ... I'm very surprised. ... All right, you're more surprised than I am, but I am as surprised as well. ... I am as surprised as you are, Dmitri! Don't say that you're more surprised than I am, because I'm capable of being just as surprised as you are. ... So we're both surprised, all right?! ... All right.

I still think this is brilliant:

In the issues of December 16th 2000 to November 10th 2001, we may have given the impression that George Bush had been legally and duly elected president of the United States. We now understand that this may have been incorrect, and that the election result is still too close to call. The Economist apologises for any inconvenience.

-The Economist

And a good followup line:

The trouble with this view is, er, the facts.

-The Economist, Dec 20, 2001

What a great magazine... :-)

Microsoft is a lot of negative things, but stupid isn't one of them. So, for the sake of argument, let's consider that IE as a freestanding product has been not discontinued, but mothballed. No one seems to be working on it, no new versions are forthcoming, there is no roadmap for future development.

What happens then if Mozilla really does start to gain market share?

How threatened would Microsoft feel if Mozilla's user base hit 10%, 25%, or 50%? How high would the level have to get before they took action? My guess is that the first tactic would be to accelerate the next version of Windows, and provide incentives to make sure that the public upgrades (who says competition is a bad thing?). But if that's not enough, and Mozilla/Gecko use kept rising, how would they respond?

My hunch is that there is some threshold -- and I don't know what it is any more than anyone else does -- above which Microsoft would have no choice but to take IE out of mothballs, and the malarkey about "we can't improve IE without improving the underlying operating system." That's baloney, as should be obvious to anyone that has used any browser that has made a release since IE5/IE6 came out (Mozilla, Phoenix, Safari, Opera, OmniWeb, iCab, CrazyBrowser [which is even IE based!), etc).

So, if the sleeping giant stirs, and independent IE development is reactivated, how long would it take to ramp up work on it? It wouldn't surprise me if a point release (with atrophied features like popup management, maybe tabs) could be out in three to six months, and a full release within six months to a year. At a guess, obviously I don't know how long it would take to allocate people to work on it, get them familiar with the existing codebase, etc, but it wasn't that long ago that Netscape and Microsoft were release major browser upgrades on something like a nine month schedule, and maybe -- just maybe -- some stiff competition from Mozilla (and, to a lesser extent, Safari & Opera) can spur on another round of that.

Mozilla may try to be the tortoise, sneaking past Microsoft's sleeping rabbit, but sometimes the rabbit doesn't oversleep...

Is it me, or is this man (mirrored here, but please be merciful of my limited bandwidth) demonstrating that the best place to use office is, well, your "other" office?

Pretending for a minute that there are no privacy downsides to RFID (obviously they exist, and they may be significant enough to scuttle public acceptance of the technology), why is all the RFID hype targeted at inventory control? That's a topic that I'd assume is only of interest if you happen to work in retail or manufacturing. How about an application that would appeal to actual consumers: point of sale systems?

Around here, all the supermarkets and a lot of other stores are beginning to install self-service checkout machines. They're popular with retail of course, since if they catch on it'll save them a lot on payroll costs. Consumers on the other hand seem a little iffy, mainly because the machines are a royal pain in the ass to use -- they're very picky about how you move items from your basket, across the scanner, and into the shopping bag; things have to be done in a rigid order; mistakes are all too easy & very frustrating. Said one customer in a test conducted by the Boston Globe, "why am I doing all this work?" It's not like it's any faster than the traditional checkout lanes, and to date the stores aren't cutting any deals for people willing to scan & bag their own goods, so this user had a very valid point: what's in it for us?

RFID seems to me like a natural solution to at least part of this problem. If everything in Home Depot or CVS had an RFID tag, then I can picture a situation where a specially made checkout lane would allow me to simply push my cart or basket through the register, then present me with an inventory of what was accounted for. All I'd have to do would be to swipe my debit card & move the goods into a bag to bring home. The only place this seems to really break down is items like supermarket produce, where each individual item is unlikely to be tagged -- do you want to eat a tomato with a microchip in it? But still, it seems like for most products this would work, and in the worst case you'd have to manually enter part of your purchase into the machine, but most could be recorded automatically.

My hunch is that if such a system could be put into use, the increased convenience would be valuable enough to many people that they could accept the privacy tradeoff. My hunch is that improved inventory control will never be enough, by itself, to win over the public -- it's all stick, no carrot.

Am I missing something, or is this potential use of RFID never really mentioned?

I was going to post this in reply to somebody's comment, but it wasn't really relevant there. So, for my own reference (and for that of anyone that reads my journal (hello, Andrew!)...)...

David Boies has:

• represented IBM in its antitrust trial (found innocent)
• represented the DOJ in the MS antitrust case (found guilty, but the defendant got the old wrist slap & is now arguably even more menacing now than they were before)
• defended Napster against the RIAA (found innocent, but ended up folding anyway)
• represented Al Gore in his effort to get an accurate count of Floridian votes (lost, and now El Presidente is dragging the country through the mud)
• agreed to represent SCO in it's Quixotic campaign against IBM and, by proxy, all of Linux, the Open Source world, etc.

Supposedly he had only lost 1 of his previous 45 cases prior to the 2000 presidential "election", but his record in the past few years seems more uneven than it used to be. Moreover, his record in taking what I'd call the "right" side in these cases is generally okay with me, but this SCO thing is just... nasty. Even the "good" lawyers are just in it for the money, I guess...

What's with this? The .dmg disc image file for the new iChat beta generates a .pkg installer for the new version, but when I open it the thing just flickers in the dock before going away without having done anything. If I tail the logs in Console.app, I see

2003-06-23 23:04:11.366 Installer[7012] Unable to load nib file: PagedInstaller.nib, exiting

I don't get it. There's no .nib files inside the .pkg directory, but I'd think that if there should be it would be part of the distribution, and apparently other people have been able to get the thing installed. Is this somewhere else on my computer then? It's not in either of my personal ~/Library or the system-wide /Library or /System/Library directories. Google has no idea what a "pagedinstaller" is, and neither do I.

Has anyone else seen a bug like this?

Workaround found: Based on a discussion at MacOSXHints.com, I found out that /usr/sbin/installer (see man installer) provides a command line interface:

% sudo installer -pkg ~/Desktop/iChatAVBeta.pkg -target /
Password:
installer: Package name is iChat AV Public Beta
installer: Installing onto volume mounted at /.
installer: The install was successful.
%

Not that that resolves the original question, but now [a] iChat is upgraded, and [b] I have a more "natural" (to me) way to install .pkg applications. So I'm mostly happy now...

I've just received a copy of the June 2003 edition of ;login:, Usenix & Sage's journal.

My writeup of the MIT Spam Conference is on page 65 :-)

It looks like the permanent URL of the article will be www.usenix.org/publications/login/2003-06/index.html, but that won't work until the next issue after this one comes out.