Linux vs. Bullshit

Linux doesn't lie, any more than gravity lies, or geology lies, or atmosphere lies. Like those other natural things, Linux has no guile, no agenda beyond supporting the entirety of use-space. In rough words, there's no bullshit about it, and that's one reason it gets used. Let me explain.

Prior to discovering Linux (or having it discover me) I was peripherally involved in high-level UNIX debates, while helping Sun Microsystems promote its SPARC microprocessor architecture. At that time (late '80s, early '90s), UNIX was a horse race, and Sun's steed was Sun OS, which was then at V4, as I recall. I attended many meetings at Sun, where representatives of various commercial factions worked toward reconciling Sun OS with AT&T's System V R4 (called SVR4 then). Somehow this led to Solaris, but that was after I had moved on.

What stuck with me from those meetings was that every company involved had a self-interested stake in how UNIX evolved. I suppose that's one reason Linux fascinated me from the moment I first learned about it. As Andrew Morton explained to me in 2005 "On the kernel team we are concerned about the long-term viability and integrity of the code base. We're reluctant to put stuff in for specific reasons where a commercial company might do that."

In other words, no bullshit.

I mean bullshit seriously, as does the philosopher Harry Frankfurt in his landmark book On Bullshit (Princeton University Press, 2005). Back when I first read the book, I page-flagged this passage:

Wittgenstein once said that the following bit of verse by Longfellow could serve him as a motto:

In the elder days of art
Builders wrought with greatest care
Each minute and unseen part,
For the Gods are everywhere.

That's the same point as "Linus's Law" (coined by Eric S. Raymond and named for Linus): "given enough eyeballs, all bugs are shallow". Frankfurt goes on to observe:

Is the bullshitter by his very nature a mindless slob? Is his product necessarily messy or unrefined? The word shit does, to be sure, suggest this. Excrement is not designed or crafted at all; it is merely emitted, or dumped. It may have a more or less coherent shape, or it may not, but it is in any case not wrought.

The notion of carefully wrought bullshit involves, then, a certain inner strain. Thoughtful attention to detail requires discipline and objectivity. It entails accepting standards and limitations that forbid the indulgence of impulse or whim. It is this selflessness that, in connection with bullshit, strikes us as inapposite. But in fact, it is not out of the question at all. The realms of advertising and public relations, and the nowadays closely related realm of politics, are replete with instances of bullshit so unmitigated that they can serve among the most indisputable and classic paradigms of the concept.

It should then be no surprise that Linux folk have a heightened tendency to resist advertising, at least in their browsers. Confirmation of this comes via "Ad Blocking, Measured", a May 2012 research report by ClarityRay, a company in the business of thwarting ad blocking. In it they find an "overall rate of ad-blocked impressions in the US and Europe" of 9.26%. Among six content domains, the ad-blocking rate was lowest in Business & Finance at 6.11% and highest in Tech at 17.79%. Among browsers, the lowest ad-blocking rate was earned by Explorer at 3.86% and the highest by Firefox at 17.81%. Among operating systems, the ad-blocking rate was lowest with iOS at 1.33% and highest with Linux and Ubuntu at 29.04%.

So it also should be no surprise that Microsoft is now turning on Do Not Track by default in Explorer, and Mozilla (parent of Firefox) is battling the DAA (Digital Advertising Alliance) and the IAB Interactuve Advertising Bureau) over proposed blocking of third-party cookies. (Apple's Safari already does that.) On the whole, browser makers are on the users' side. (One exception is Google's Chrome, which has no plans at this time for blocking third-party cookies. Google is the biggest player in the on-line advertising business.)

To be fair, both the DAA and the IAB would like advertising to be as wrought as possible, and for consumers to appreciate the good intentions and effects of their business. I know that because I've talked to them about it. Those organizations see themselves, correctly, as advocates for good behavior in a business rife with the opposite. They also know consumer appetite for advertising, such as it is, tends to be highest for that which is most wrought—for example, Super Bowl ads. Among less expensively wrought forms of advertising, there are breeds that also enjoy a degree of demand by consumers. You'll find these in Linux Journal and Vogue. Subscribers and advertisers both pay for those magazines, and there is a kind of symbiosis in the middle. In specialized publications like these, ads tend to enhance rather than to diminish editorial content.

So this topic is close to home for us here at Linux Journal. It's also close to home for me personally, since I labored in the advertising business for much of my adult life. The "inner strain" of which Frankfurt speaks is one I experienced in that business, and still feel today as I try to make sense of what it has become in the on-line world.

Advertising in the pre-digital days was an obvious business. You knew how an ad got to be where it was, and what it was doing there. Because of that, advertising also carried what economists call a signal. That is, a message of sufficiency. In "The Waste in Advertising is the Part that Works" (Journal of Advertising Research, December 2004), Tim Ambler and E. Ann Hollier say advertising's base-level purpose—aside from its specific message—is akin to a male peacock's fanned-out tail. It speaks of the company's substance, and the fact that it can afford to advertise.

With today's "adtech" business, the provenance of many ads is unknown. If you see an ad for badger-hide gloves when you're reading the Daily Feh, and then again when you whimsically look for vacations on Pluto, it's not clear why that ad is there, or why it's following you around. That's because the ad might be targeted by way of some combination of the following, only the first two of which existed in the old pre-Internet advertising world:

  • Advertisers
  • Agencies
  • Agency Trading Desks
  • SSPs (Supply Side Platforms)
  • DSPs (Demand Side Platforms)
  • RTB (Real Time Bidding)
  • Exchanges
  • Creative Optimization
  • Retargeting
  • Verification/Privacy
  • Media Planning and Attribution
  • Tag Management
  • Measurement and Analytics
  • Data Suppliers
  • DMPs (Data Management Platforms) and Data Aggregators
  • Ad Networks—
    • Horizontal
    • Vertical/Custom
    • Targeted Networks/AMPs
    • Performance
    • Mobile
  • Media Management Systems
  • Ad Operations
  • Ad Servers
  • Publisher Tools
  • Web Analytics
  • Gamification
  • Real Time Message/Offers
  • Sharing Data/Social Tools

Most of that list comes from Luma Partners' "Lumascapes": amazing graphical representations of marketing businesses, each sandwiching dozens of companies and categories between first source (such as marketer or advertiser) and the consumer. Many arrows run between collections of companies in one specialty or another. In some cases, the arrows point recursively back toward the source. Randall Rothenberg, IAB's CEO, tells me this whole collection is a "black box" to the consumer and a problem he and others at the IAB want to fix.

Another view of that box comes to us by way of IBM and the research firm Aberdeen, which together diagram "The Big Datastillery". Copy at the top describes it as "Best-in-Class Strategies to Accelerate the Return on Digital Data" and "a revolutionary new appliance to condense terabyte scale torrents of customer, transactional, campaign, clickstream and social media data down to meaningful and actionable insights that boost response rates, conversions and customer value".

Below that is a maze of pipes pouring stuff into a hopper of "Best-in-Class companies" that are "2.8 times more likely than Laggards to incorporate unstructured data into analytical models". The pipes are called:

  • Customer Sentiment
  • E-mail Metrics
  • CRM
  • Clickstream Data
  • PPC (Pay Per Click)
  • SEO Data
  • Social Media
  • Marketing History
  • Ad Impressions
  • Transactional Data

Coming out of the hopper are boxes and tanks, connected to more piping. These are accompanied by blocks of text explaining what's going on in that part of the "datastillery". One says "Ability to generate customer behavioral profile based on real-time analytics". Another says "Ability to optimize marketing offers/Web experience based on buyer's social profile". Another says BIC (Best in Class) outfits "merge customer data from CRM with inline behavioral data to optimize digital experience".

Customers are represented (I'm not kidding) as empty beakers moving down a conveyor belt at the bottom of this whole thing. Into the beakers pipes called "customer interaction optimization" and "marketing optimization" excrete orange and green flows of ones and zeroes. Gas farted upward by customers metabolizing goop fed by the first two pipes is collected by a third pipe called "campaign metrics" and carried to the top of the datastillery, where in liquid form it gets poured back into the hopper. Text over a departing beaker says "137% higher average marketing response rate for Best-in-Class (6.2%) vs. All Others (2.6%)". (The 137% is expressed in type many times larger than the actual response rates.) The reciprocal numbers for those rates are 93.8% and 97.4%—meaning that nearly all the beakers are not responsive, even to Best-in-Class marketing.

In fact, advertising and marketing have always been good at bullshitting themselves. Consider, for example, the old saying (often attributed to John Wanamaker, who is not known to have actually said it) "I know half my advertising is wasted, I just don't know which half." The correct answer is that most of it is wasted, and the industry has known that for the duration. They just don't want to talk about it. And, on the whole, neither do we. Frankfurt explains:

In fact, people do tend to be more tolerant of bullshit than of lies, perhaps because we are less inclined to take the former as a personal affront. We may seek to distance ourselves from bullshit, but we are more likely to turn away from it with an impatient or irritated shrug than with the sense of violation or outrage that lies often inspire.

Lately, however, people are feeling somewhat more violated, especially by tracking, thanks to Edward Snowden's revelations of how the NSA is spying on everybody. We now know that the Feds and marketing mills are both harvesting massive amounts of personal data without revealing to us what they know, and that the two are actually in cahoots, at least some of the time. This is especially vexing, because the feds should be the ones protecting us from bad actors, rather than bad actors themselves.

But let's set that stuff aside and just look at the bullshit aspects of the whole thing. How much good is all this data collection and manipulation actually doing for its perpetrators? And how much are they also bullshitting themselves?

When I was doing research for The Intention Economy, the most important input I got came from Doug Rauch, the retired president of Trader Joe's. One big reason for Trader Joe's success, he told me, is (translating to the vernacular) that it minimizes marketing bullshit. It has no loyalty program, no coupons, no discounts and none of the expenses any of those involve, including the cost of running a big data mill. Doug's job as president of the company, he said, was to shop along with customers. Talking in person, in stores, with customers, was his main form of research. One result, he told me, is that there is hardly a product on the shelves at Trader Joe's that isn't influenced directly by customers talking to workers in the store. Trader Joe's also doesn't go to retailing tradeshows, Doug told me, because too much of what goes on at those things is all about manipulating the customer. These manipulations are highly complex and therefore come at high costs to the stores as well. By avoiding this kind of thing, Trader Joe's spares itself the cognitive overhead required to rationalize complicating the living shit out of everything, which is what marketing tends to do—and does now, more than ever, with Big Data. Thanks to Big Data and the perceived need to run big complex marketing mills, the Chief Marketing Officer (CMO)—a title that didn't exist twenty years ago—runs an overhead-fattening operation that can dwarf the old IT budget.

Meanwhile, only a tiny percentage of the output of these mills is actually useful to us. The rest is cognitive overhead, and we don't like it, even if we appreciate all the free-as-in-beer stuff it pays for.

What we need are fixes on the Linux model: stuff that's useful.

This issue of Linux Journal will hit the streets on the seventh anniversary of the month I made it my my mission in life to encourage developers to create tools that make individuals both independent of companies yet better equipped to engage with them. That work is happening, and I continue to have faith that it will succeed, and improve the way business works for everybody. That faith comes from watching how Linux and other free and open-source developments, by simply being useful, defeat bullshit.

No bull image via


Doc Searls is the Editor in Chief of Linux Journal


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Browsers OR Market

Misha's picture

This is a general business/market tendency that the businesses shifts according to the market inclination. At present, market is fully based on the users. So, why the businesses and technology is also shifting towards the same.

The Bullshit Wastes Money

Dana F. Blankenhorn's picture

Ever since the Web was spun, marketers have been spending billions of dollars to prove that intrinsic marketing can be better than extrinsic marketing. That is, that ads aimed at you are better than ads aimed based on what you're reading.

Even with 137% better performance, that's not the case. It's 137% better than run of market, not than aiming based on content.

An ad in Linux Journal will draw a response because just about everyone reading the copy next to it is interested in Linux. The ad becomes a service, supplementing the copy, pointing toward next steps. It's relevant, in other words.

That's not necessarily so when an ad pops up on a random site telling me I need "blergh." It may be popping up because I just bought "blergh," or because I clicked on a link to "blergh." But if I'm not thinking of "blergh" the ad is still worthless. It doesn't provide service.

In other words, the pre-web techniques still work best. I believe that's not just true in customer relationships, but everywhere.

All hail extrinsic marketing

Doc Searls's picture

Thanks, Dana. These are excellent points that should go in our advertising sales kit. :-)

They are also in line with what Don Marti (a former Editor-in-Chief here) has been saying for some time here: .

Doc Searls is the Editor in Chief of Linux Journal

LInux has its share of BS

RO's picture

When I read the "diff -u" column in LJ, I am frequently reminded that the claim for Linux that it puts new life in old computers is often BS. Many times, "diff -u" chronicles yet more dropping of support for that older hardware, which I experience personally as I try to keep using my older PC's, and find that more and more new distro's won't fully support what I have as one driver after another is "deprecated" out of existence.

The most annoying BS nowadays for me is the PAE support required even in 32-bit versions - really? What need is there for this 64-bit addressing workaround when a PC does not even support that much RAM, which is typical of many slightly older notebooks with Pentium-M/Centrino CPU's that can use only DDR modules, and often only 1 or 2, thus 2 GB being their max like my Dell D800. It has the max 2GB and a 2.1 Ghz CPU (and a gorgeous 1920x1200 resolution screen, which is noticeably nicer than the movie-mode 1920x1080 vertically challanged screens considered hi-res now) - plenty of power for anything I would want to do with a notebook class PC (for personal use), but to upgrade from Mint 9 (Ubuntu 10.04-based) since the updates are getting few and far between despite having 2 years of "Long Term Support" remaining, I have to check carefully that PAE is not required for the distro.

Ubuntu 12.04 only has certain of the lower end DE's like XFCE that come that way, and I was hoping for a whirl with KDE, but no go. The presence or lack of that requirement is often omitted from distro descriptions I figure because they mostly assume only "modern" machines with PAE-capable CPU's will be used. So much for extending the usefulness of older PC's - BS!

BTW, your article DOES misspell, as noted by "Wouter", Wittgenstein ("Wittegenstein") and inapposite ("inaposite") in the quotes from Frankfurt (as of my current view), so maybe "sic" would be apposite with them in those quotes ;-}

An idea, and thanks

Doc Searls's picture

So maybe we should have a "calling BS" section here at LJ. I mean this seriously. Why not? Might be fun as well as useful.

As for the spellings, thanks. They are now corrected (I hope). I have an explanation for the fail, but it's lame. Better just to get debugged and move on.

Doc Searls is the Editor in Chief of Linux Journal

Calling BS LJ Section: Give it a Shot

RO's picture

It could be instructive, although more egregious cases like some of Ubuntu's weirdness (Unity, and the Amazon "tracking") do get called out. Still, a focus on that topic should hightlight the lower profile stuff - sort of like a less combative and scatological "Linux Hater's Blog" (I have not checked on that for quite some time - the guy had many valid points to ponder about Linux/FSW/OSS failings, but his language was absolutely too far over the top - the message got lost in the gutter).

Linux Hater's Blog

Doc Searls's picture

Thanks for bringing it up. Good idea for a column here at LJ. Meanwhile, a link to Linux Hater's Blog, which has been idle for a few months.

Doc Searls is the Editor in Chief of Linux Journal

It should then be no surprise

Supputator's picture

It should then be no surprise that Linux folk have a heightened tendency to resist advertising, at least in their browsers.

I'm willing to bet that it's not only because Linux users have a "heightened tendency to resist advertising" as you put it, but also because the average internet user does not know it is possible to block advertising.

There are older generations that are very much against advertising. But when they use a computer to do something online, they have no way to understand that it's possible to remove such ads. They assume ads are like tv commercials or newspaper advertising, close impossible to avoid.

Therefore I think Linux users are simply more technologically adept, so this is why they block ads more frequently.

Average users

Doc Searls's picture

At this point average users, on the whole, don't block ads. Yet ad and tracking blockers are among the most popular browser add-ons and extensions. I know plenty of non-technical people using them. According to the ClarityRay report I cite above, nearly a quarter of all the Web advertising in Austria is blocked by users.

Obviously technical folks are ahead in the practice, but I'm not sure the rest of the world is that far behind. Especially after recent revelations about surveillance.

Doc Searls is the Editor in Chief of Linux Journal


anon coward's picture

Ubuntu/canonical has become an embarassment to the linux community.


Anonymous's picture

I'm not sure if you're attributing a fundamental *cause* to the "no bullshit" aspect of the Linux community, but it seems fairly clear to me. There is one person (as opposed to 5 companies) with a self-interested stake in how Linux evolves. Right now, that is Mr. Torvalds, but it seems like the way the organization is set up, it will always be a single person. Many other projects have a single "The Buck Stops Here" leader, so we even have a handy name for it: "BDFL".

Teams with a BDFL don't have bullshit. Teams that try to rule by consensus, or negotiations, or whatever, end up with bullshit. (Extreme example: Congress.) There's no bullshit in Linux because Linus can say "Yes" or "No", and that's it.

You can almost tell which products have such a leader by how good they are. For example, among techies, the creator of GMail is fairly well-known, but the creator(s?) of Google Groups is not. GMail is much more loved by users than Groups is.

Every piece of software I've used that I've enjoyed using, it turns out was the brainchild of one person. It's their baby. Every piece of software I've used that I've hated was something "designed by committee".

Trader Joe's sounds like a good example of this. It's neat that they didn't have a complex marketing department to screw it all up, but the core reason they didn't have that is because it was run by one guy who shopped there, too. His name was on the sign, so he had great incentive to keep it great -- like Linux, come to think of it!

You can do things by

Anonymous's picture

You can do things by consensus but the model isn't very well known and widely used. Consensus does happen within the BDFL model too! The problem is that building software and a lot of great things is like making a movie or writing a book. You need a few people who have the full vision and who can direct it. But there's a danger in that the vision may fail if it isn't sufficiently communicated with everyone else.

The "bullshit" part

personne's picture

The more I thought about it, the less I like it. The entire piece is dazzling at first, and I still really like what you're saying about marketing, but it gets to the heart of the problem. There is rarely true correctness. You can have arbitrary technical correctness, and "works for me," but forcing everyone to come around to that is a distortion. Then you have the attitudes about other's work and perspectives, that it's "bullshit." Which is very often really rude and bullying. A work is not poo, and you're stunting the conversation with the comparison. A work is a person or group trying to achieve something. Maybe it's poorly thought out or implemented, but it's still meaningful of a perspective that probably has considerable merit. I know I've hacked on a lot of projects and they could be considered poo, yet ten years later they exist and sometimes I was a part of the formulation. Today more and more it's easier to achieve things. But there will always be a contingent that likes to think of itself as superior, that everyone else was unimportant, that if it can yell the loudest with the most powerful method, then it's correct.

It would be great if you could decouple the machismo part about "bullshit" with the things you're saying about marketing, and where you think that might be heading.


Doc Searls's picture

Hmm. I don't think I'm calling for correctness, or seeing it in Linux. I am trying to say there is a model in Linux that exposes or resists what's not bullshit (as Frankfurt defines it), and that we can use the same kind of model in regarding what's happening in marketing today.

Not sure I get all of what you're saying, though. Also don't see the machismo. Are you saying the use of the word bullshit is macho somehow?

Doc Searls is the Editor in Chief of Linux Journal

Code like math

Wouter's picture

Good stuff. I've worn black most of my life, even removing labels from clothes. The lack of colour in Linux – just like math is colourless – makes it attractively pure. It has, in and of itself, no brand, no government, no flag, no political movement and no religion. Its neutrality imparts freedom. It simply is. The fact Linux, *BSD and other software is open-source makes it practically impossible for one brand or government to turn it into a tool of oppression, abuse privacy, force the user into unwanted directions or limit personal choice.

Please correct the spelling of "Wittgenstein" and "inapposite", though. ;)

Code without color

Doc Searls's picture

Interesting reply. Thanks. Making me think.

Just so others know (and fwiw), Wittgenstein and inapposite are spelled correctly. :-)

Doc Searls is the Editor in Chief of Linux Journal


Doc Searls's picture

I was wrong in the last comment. They are spelled correctly in the two comments above, but not in the piece. Thanks to RO, in another comment, for the second (and hopefully successful) pass at the spelling debug.

Doc Searls is the Editor in Chief of Linux Journal

Excellent but Questions

bruce w fowler's picture

Bully! (TRism, not harassment.) An excellent article. I shall network it. But I have some questions. Could you provide a definition of what you mean by Bull Shit? In particular I am wondering if what Canonical is doing with Ubuntu is BS? So far as I understand your thesis, Unity and Mir and the whole shtick seems congruent.

Definition of bullshit

Doc Searls's picture

Harry Frankfurt, while providing the only complete work on bullshit (which I cite in the column), does not provide a dictionary-ready definition of the word. He does, however, contrast it from lying:

It is impossible for someone to lie unless he thinks he knows the truth. Producing bullshit requires no such conviction. A person who lies is thereby responding to the truth, and he is to that extent respectful of it. When an honest man speaks, he says only what he believes to be true; and for the liar, it is correspondingly indispensable that he considers his statements to be false. For the bullshitter, however, all these bets are off: he is neither on the side of the true nor on the side of the false. His eye is not on the facts at all, as the eyes of the honest man and of the liar are, except insofar as they may be pertinent to his interest in getting away with what he says. He does not care whether the things he says describe reality correctly. He just picks them out, or makes them up, to suit his purpose.

In other words bullshit is something worthless except to the one producing it. Is Canonical doing that with Unity and Mir? Look up canonical+unity+mir+bullshit and you'll get a mixed collection of answers, but ones in which the word "bullshit" is used.

One post among those results is Martin Gräßlin's Mir in Ubuntu. In it he says, "What is not fine is causing a major disruption in the free software ecosystem by giving false technical arguments and doing bold statements about software Canonical does not contribute to. This is not acceptable." I believe what he's doing there is "calling bullshit" on Canonical, without using the word. (It appears in one of the comments, which is why the post came up in the search results.) His critique is one made often of advertising: that it's something that deceives both the speaker and the listener. But maybe he's just saying that Canonical is lying. If that's the case, maybe lying is the breed of bullshit in which the speaker knows the truth. Frankfurt would make them two breeds. But in usage they do seem to overlap.

I recommend Wikipedia's article on bullshit, btw.

Doc Searls is the Editor in Chief of Linux Journal

Very good

Gonzalo's picture