Wrong: using a mailing list, Twitter, or identi.ca just to announce your new blog posts.

Right: putting an enclosure tag in your non-podcast RSS feed, so that people who subscribe to it with a podcast client get to hear a weird random audio clip with their daily podcasts.

Tim Bray has posted a thought-provoking What Sun Should Do.

It's good to see that he picks up on two of the reasons why I would still pay more for a Linux server or hosting plan than for Solaris: the Solaris command-line tools are ready for the Museum of Unix, and Solaris has nothing like the time-saving package management available on Linux.

"The Mac and Linux platforms are remarkably similar once you drop into the command-line mode where developers live around deployment time; they have a mostly-common "GNU Userland". Solaris needs to offer a similar set of commands and utilities, so that when you go there for deployment the commands you’re used to using will Just Work."

"Also, developers depend crucially on package management with the implied provision of dependency and version tracking. Solaris must have something that’s generally competitive with APT, which means that we need to get IPS polished and finished pretty damn quick, or adopt the Nexenta approach of combining the Solaris kernel with the Debian userland."

Brian Aker agrees. The first thing you have to do with a new Sun system is put decent tools on it, and installing and updating software on Solaris hasn't changed much since 1992. Maybe that's because almost everyone who has had to do it on Solaris was on the clock, and a lot of the people doing it on Linux were playing with it on their own time. I don't know. (I have suggested that Sun should just build OpenSolaris kernel packages for Ubuntu and CentOS, with ext3 and Linux kernel personality. Add a repository, reboot, and there's a Solaris user, and another person to test the world's free software on your platform. Easier for the user than switching Linux distributions if you do the packages right.)

But what should Sun do? (besides having their, um, "Enterprise Class" Marketing department re-brand MySQL as Sun Java System Enterprise Database Suite, of course.) One more important one. Besides the all web, all the time plan, high-performance computing needs to stay on any Sun to-do list.

Sun holds the number 6 and 29 spots on the November 2008 TOP500 list, and a future Sun that wants to sell to serious web and "cloud" customers will want to grab some more spots above the fold on future lists. Even though HPC customers are the worst IT customers in the world—always beating you up on price, wanting bizarre configurations, running their own weird Linux load instead of the branded OS you want to sell and reporting bugs you can't duplicate—you need them.

You'll probably best be able to sell Management on doing HPC by pointing out the similarities between current HPC environments and future stuff needed for "the cloud" on the software side. Sun Grid Engine is on the Sun Constellation Linux Cluster, the current number 6 machine on the TOP500 list. Bray writes, "I am convinced that we have to go ahead and build some Cloud infrastructure anyhow and operate it and make it pay for itself, so that when the ecosystem does find its shape, we'll understand it and be positioned to sell the Web Suite into it." That's right. And in order to build that kind of understanding, computer companies do HPC, which is sort of like a racing program for a car company. Easy to dismiss as an ego-driven waste, but in an HPC project, you get more control over software at all levels of the stack than in an ordinary data center project. You can put the stuff out there, not bleeding edge stuff but at least not the stale software that ends up on most large IT projects, and see what happens. You can use paying (for small values of "paying", but still) customers instead of paid testers.

But there's a more important reason. When your hardware engineers start sneaking out to smoke crack, your HPC customers will tell you first, because they have the biggest "antenna" to spot flakeouts, and all their stuff has to be exactly right. Scientific journals don't accept Fail Whale pictures. "I've discovered a new subatomic particle! Oh, no, wait. Piece of crap computer." Building a PC-architecture system looks easy. World of Warcraft players do it in their basements—how hard can it be for a company that designs its own processors? But really getting it right is tricky, and HPC customers have been burned by enough not-quite-right x86 machines that they're worth the hassle just as a quality filter.

This is not new advice, just old advice that hasn't expired yet.

"For more than two years, it has been apparent in the IBM Company that we were behind in the large scientific area. This is an area where, since the days of our Harvard machine, we have attempted to lead. Although four or five years ago there was some doubt as to whether or not we should continue to try to lead in this area because of the expense and other considerations, at some point between two and three years ago it became evident that the fallout from the building of such large-scale machines was so great as to justify their continuance at almost any cost." -- T.J. Watson, Jr., May 17, 1965

CIO reports, "The cost of running Lotus Notes is killing a lot of companies. Companies like Virgin Atlantic are still going ahead with Notes-to-Exchange migrations because it's seen as worthwhile."

Wait a minute, though. Both Lotus Notes and Microsoft Exchange have pretty sizable TCO.

Ed Brill, Director, Lotus Software, IBM Software Group, points out that the writer is founder of a firm that "has a focus on delivering migrations of e-Mail systems from Novell GroupWise and Lotus Notes to Microsoft Exchange."

Got to agree with Ed here. "A little full disclosure normally would go a long way." The moral of the story, I guess, is to Google your contributors before the readers do.

The "i dot connect USB-Serial Cable" made by PPA International, which I bought at Fry's, works fine in Linux. (I have kernel 2.6.27.7 on here.)

    idVendor=067b, idProduct=2303

The chip is the Prolific Technology Inc. PL-2303, and the driver has been in the kernel for quite a while. According to the comments it goes back to kernel 2.2. If you're building from source it's

    CONFIG_USB_SERIAL_PL2303=m

(I'm using this with a null modem cable to set up some Coraid ATA over Ethernet boxes.)

"As Linux supports a larger number of different devices 'out of the box' than any other operating system, and it supports these devices on more different processor architectures than any other operating system, this proven type of development model must be doing something right :)" -- Greg Kroah-Hartman

This is where we take our cars (one VW Golf, one Honda Accord) for scheduled service, smog checks, and the occasional random problem: Fred's Wrenchouse. (No web site, that's a link to the comment page on the Car Talk site.) Always fixed right and on time, with no surprises. Some idiot on a motorcycle knocked the rear view mirror off the VW, and Fred found an off-brand mirror for less than the name brand part. Works fine.

Bunch of tabs to close.

Google AdSense Fails on Relevancy, Control, Policy, and Google Says Nothing

Comment spammers: Enough Already, says Danny Sullivan.

Want a piece of the biggest shit sandwich the world has ever seen? Apply at change.gov. (Yes, I already have.)

OCLC is eeevil, but we knew that. (Actually, most of the people who sell IT products and services to libraries should be run out of town on a rail. Library boards say, "ooooo, computers!" and open their checkbooks for all kinds of crap.)

Giles Bowkett: Robot Warriors Will Destroy America. See STEMI Compression: why DIY terrorist robots will be cheap. (And we only have to build one.)

Jeremy Zawodny's font tweak works great with Iceweasel on Debian, and in a white text/black background gnome-terminal.

Give Up and Use Tables! (seriously, use CSS for everything, but to build a big framework thingy to pack stuff into, a table is not the end of the world.)

Protect right-handedness: The California Right-handed Protection Amendment

Matthew Garrett writes down some Good power management practices. More discussion at LWN.

Recession? We can get through it. Justin Mason points out some innovative, state-of-the-art cost-cutting exercises at The Digital Depot.

Fred von Lohmann writes, in Google is Done Paying Silicon Valley's Legal Bills, "Well, Google just put the Valley on notice that the free ride is over, which means more legal burdens for smaller technology companies that previously depended on Google clearing a path for them."

But Google never was an all-purpose legal sugar daddy for anyone.

Google, like most sites that distribute works written by other people, chooses to take advantage of the Safe Harbor provision of the DMCA whenever possible, rather than taking a chance and ignoring the shakiest of the takedown letters. This year, takedowns made the news. The McCain campaign's web video ads have been repeatedly either knocked off YouTube.

We faced the same problem in 2002, when a Scientology lawyer sent an overreaching takedown against Google linking to the anti-Scientology domain xenu.net in search results.

If you look at some of the mainstream media coverage of that Scientology case, it sounds like Google was required to obey the takedown letter. That's not true. Google could have chosen to ignore the letter. In that case, though, the company would have lost its right to use the DMCA's Safe Harbor and would have been at greater risk of an infringement suit from the letter sender.

But Google, like a sensible company, works in its own interest and always has. So it chooses Safe Harbor. On a much larger scale, Google settled a lawsuit from Agence France-Presse over Google News headlines, excerpts, and links. Blogger Mac Slocum wrote, "Terms of the deal weren't announced, but the fact that there's a deal at all misses the point of Google News." The company didn't take a big chance to make a Fair Use precedent. It did what was good for Google.

Prof. Nicholas G. Carr, in The Google Enigma, applies Joel Spolsky's useful "Commoditize the complement" argument to Google. Google isn't a whole new kind of company, and it isn't unusually non-evil. It's a company that just happens to have a lot of complements, and it makes sense to commoditize your complements if you can do it cheaply enough.

"For Google, literally everything that happens on the Internet is a complement to its main business. The more things that people and companies do online, the more ads they see and the more money Google makes," Carr writes. But there are still some areas where old-fashioned lock-in makes more sense than just strewing goodwill everywhere. Google settled with Yahoo instead of fighting a dubious search ad patent.

When Google can can get what looks like a measurable advantage from a deal with a known, large partner, it will make the deal and leave the startups and the EFFistas to make their own precedents.

(Disclaimer: I have participated in a protest against Google over the xenu.net issue and also received major sponsorship funding from Google for FreedomHEC, the Linux device driver unconference. Don't worry, I'll put all this stuff on my Pirate Czar disclosure form when they get back to me about that.)

(Just catching up on links that the Perl script dragged in, and found one of those "my web framework is better than your web framework" blog posts. Total spam URLs in the comments: 519. If I were a black-hat SEO, I'd have something like my "find the stuff that people are making new links to" script, too. Are the black-hat SEOs and I reading the same stuff?)

Apple's ambitious Macintosh project gets a lot of credit for bringing a whole bunch of new ideas to a lot of people in one affordable package, but one of the most important ideas, and one without which it wouldn't have worked, was a book called Macintosh Human Interface Guidelines.

Different keyboard shortcuts in different applications are bad enough. If you used WordPerfect and Lotus 1-2-3 you wasted a bunch of your muscle memory on different commands to do the same thing. But imagine a GUI system with as much inconsistency as the old MS-DOS applications scene, where not only do you need different key combinations to do the same thing, but the mouse button behavior changes as the pointer rolls from appllication to application. (Actually, if you've used pre-KDE/GNOME Unix or Linux desktops you don't have to imagine it.) Apple couldn't have introduced a Macintosh where every application developer did his or her own thing—people's heads would have exploded and their next of kin would have taken the machine back to the store.

The idea of Human Interface Guidelines changes all that. Don't go to disk for something that's in RAM, and don't make the user go to the manual or the menus for something that's already in muscle memory. Like a lot of Macintosh ideas, the rest of the industry has followed, and today something isn't really a "platform" until it has an HIG document of its own. (hey, maybe KDE and GNOME could use their upcoming lovefest to combine their HIG projects.)

But all the HIG stuff out there is more or less based on the original Macintosh version. Which is mostly a very good thing, except for feature that today looks like a mistake. The action to start a new program is the same as the action to open a file.

There might have been a good reason to do this back in the days of "Please insert the disk: Untitled". But today we have safer, more useful ways to install software: package management and click-to-license facilities such as Lindows/Linspire's Click-n-Run Warehouse and Apple's later CNR-like App Store. (Yes, inspiration flows both ways.)

Just clicking on an executable to run it leads to stuff like sneaky attachments and tricks to make executables look like data files. Now we have wack-ass "security vendor" stuff like white listing to cover up what's really an industry-wide interface design problem.

Unfortunately, some platforms are going to have trouble fixing the click-to-run problem. Microsoft has to do all its security design under the vulture eyes of antitrust ambulance-chasers working for the so-called security vendors. Even something as unhelpful as "hey, Rocky, let's pattern-match for known naughty programs!" is a minefield. Imagine if they tried something like an RPM-like installer for signed packages, or Apple's App Store.

If I get the Pirate Czar job in Washington, DC, we'll offer the IT industry a deal: we won't micromanage your product or service design decisions with antitrust law, but in exchange you'll have to live without "pro-trust" policies, such as anticircumvention enforcement against non-infringers and patents that apply to independent clean-room implementations of wire protocols or data file formats. This kind of thing is why we need someone for the Pirate Czar job who isn't stuck deep in the silo of Intellectual Property Law.

Still haven't heard back about the Pirate Czar job. In the meantime, I just want to make it clear: not these pirates (via The Excellent Adventure).

Are these ambitious pirates really privateers? What they are is Global Guerrillas, a new generation of organized violent people. Global guerrillas can organize themselves as a social club, a public service operation, an organized crime gang, a terrorist group, a liberation movement, or even an army or police force, depending on what kinds of connections and resources they can get a hold or, and what opportunities are available. They're constantly changing their style of organization in order to protect themselves and make money.

John Robb, who coined the term, compares global guerrilla networks to open source projects. It's a good comparison. During the first Linux boom, projects spawned companies like an underground fungus popping up mushrooms. When most of the companies got garbage-collected, the projects persisted. Today, as a developer, you might work on the same project for several different companies. When one company implodes, the actual software doesn't become typical Valley abandonware. You just check out another copy from wherever you end up, and keep going. Global guerrillas work the same way. You can't "break the back" of their organizations. William S. Lind wrote, Insurgencies, like octopi, are invertebrate.

Anyway, pirates, privateers, whatever you want to call them. Do you really expect the Liberian navy to come to your rescue when pirates take your Liberian-flagged ship?

When you register your ship with a Flag of convenience, you're not really in a position to ask for help from the USA and other countries that actually enforce safety and labor standards, and, of course, collect taxes, on ships. When it comes to convenience-flagged ships, I'm for restoring the old 24-hour rule: if the pirates have the ship for 24 hours, and the Navy takes it, it's a lawful prize. The US Navy certainly needs to have ships on anti-pirate patrol, but as a US taxpayer myself I'm against free services for deliberate tax dodgers.

There's a related point on protection from infringing "pirates," which I plan to establish as part of my work as Pirate Czar. Just as you have a claim to the Navy's protection when you fly the US flag, follow US maritime law, and pay your taxes, you have a claim on the anti-piracy services of the FBI and the criminal justice system when you respect the principles of copyright law, including your audience's right to First Sale, Fair Use, and time and space shifting.

Some copyright holders try to rewrite the copyright bargain using End User License Agreements and Digital Rights Management systems. That's fine, but understand that when you apply a EULA or DRM system that goes beyond copyright, you're abandoning our flag, and the balance of copyright and speech rights for which, among other things, it stands. When you haul down the Stars and Stripes, and run up the colors of EULAstan (the flag looks gray from a distance but it's really all fine print), any criminal complaints from you are going to the bottom of the pile.

This doesn't mean that on my watch as Pirate Czar we just become the GPL Enforcement Police. Copyrighted works released with the conventional "all rights reserved" or "just like a book" terms are just as entitled to protection by the powers of the criminal justice system as freely or collaboratively licensed works. But if you want US law enforcement to enforce your copyrights, you will need to show respect for US copyright as our law defines it.

(updated 18 Nov: Stormy's Asus machine works fine, but only with an alternative distribution, not with the preloaded Linux install.)

Joe Wein compares the pricing of Asus Eee Box computers preloaded with Linux and Microsoft Windows. "Adjusting for the hard drive cost, the Windows version is now $10 cheaper instead of $30 more expensive than the Linux version, money which we can assume came out of Microsoft's revised OEM pricing."

The desktop Linux market is tiny. But it's driving down prices in a much larger market: the one for desktop Windows XP. Microsoft is making Windows XP available for "ultra low-cost PCs" until June 30, 2010. Would we have seen that without Linux netbooks? No way to tell. And of course the real price is impossible to figure out, because co-op marketing money is the loophole in the first Microsoft consent decree. (Microsoft agreed to charge a uniform price, but refunds an undisclosed amount. Take that, DoJ!)

As long as there's a credible Linux threat at the low end of the market, the PC OEMs can expect to earn easy money from subsidized copies of the current low-end OS from Microsoft. Why? Imagine the alternative. Step one: some simplified desktop load based on Linux gets a toehold in the low-end PC market. Step two: users start returning products such as printers and cameras when they don't work out of the box with their low-end PCs which they don't even know are running Linux. Step three: printer and camera manufacturers start taking the minimal steps needed to make their stuff work with Linux. Step four: users with a little higher budget start settling for Linux. And so on.

Every time Consumer Reports has taken a look at a preloaded Linux box, they've hated it. Even worse than their car reviewers hate Jeeps. And we all know there are still areas that are good for more dorky fun than most people would appreciate. Just getting sounds to play when they're supposed to is a mess. How many volume control doo-dads do you really need?

But the cost of the whole desktop Linux project is tiny compared to the potential gains from subsidies to hardware vendors. So it makes sense for those hardware vendors to keep throwing money at Ubuntu to do integration and tweaking. (And of course, to sponsor Fedora to do all the real work that Ubuntu gets all the credit for because Ubunteros and Ubunteras fly around in spaceships and send out free CDs of nekkid pictures. It's true! I read it on LWN!)

Besides, Linux doesn't need to work all that well, usability-wise, to work on the business level. The Asus system that Stormy Peters bought does not have wireless or camera working correctly, unless you do something familiar to 1980s PC users and Linux hobbyists, but not to today's computer customers: install an OS from scratch. The wireless and camera work with up-to-date Linux distributions that you can download, burn to CD, and install, but not with the software that comes on the machine. Linux on the desktop is a failure, but a profitable failure.

IBM, HP, and Lenovo have all introduced Linux laptops and quietly dropped them. Were those projects really failures, as they looked, or were they good for some hush money via the co-op marketing budget?

Meanwhile, if you want a beefier machine, you can still save quite a bit with Linux. Over at Dell's pre-installed Ubuntu page, an XPS M1330 with a Core 2 Duo T5850 (2.16GHz/667Mhz FSB/2MB cache), 4GB RAM, a 320GB SATA Hard Drive (7200RPM), and Intel graphics goes for $1,049. Pretty much the same configuration, but with a faster FSB, is $1,399 with Microsoft Windows. Do the new designs get qualified with the Microsoft-based load first, then Linux? Makes sense considering how many Dell can sell of each.