Real Linux advocates see shades of grey.

So, it appears the last post has attracted substantial attention, with both conflicting and harmonized opinions floating to the surface.

The only differing opinion that contains content beyond a knee-jerk reaction is Joe Brockmeier’s post, “Real Linux advocates educate, not humiliate”, and it’s what this article is a response to.

Block-quoted material is sourced from his posting and duplicated here to facilitate response to individual sections.

The Ubuntard opinion (as outlined in Choosing an OS is Not a Team Sport) is that Linux is not well-served by gaining the average user. It would seem that few are cut from an appropriate cloth to contribute to the free software ecosystem — The majority are content to exist with a modicum of technical knowledge, deferring to their more technical friend(s) when problems occur.

Inclusionism does not serve free software well: Would the Linux kernel be where it is today if half-baked solutions from novice programmers were readily integrated into mainline? The kernel is a shining star of what’s achievable through open development because it’s developed by some of the sharpest minds in software development, with their collective oversight ensuring high-quality results.

To make my point at a more end-user level, look at the bug tracker for any major piece of software, and for every legitimate bug, there will be multiple WONTFIX or INVALID bugs, as inexperienced end-users are being taught to report whenever they encounter an issue, though they may lack the knowledge to provide adequate documentation, reproducibility, and other qualifiers. All this form of inclusionism does is waste valuable development time.

So, here we have member of the Linux community that’s dripping with contempt for people who cover technology that aren’t as expert as they are.

And, indeed, a case can be made that this happens frequently: A writer or journalist that covers tech “discovers” Linux and decides they want to spread the word. In their enthusiasm, they bite off a bit more than they’re ready to chew, and get some (often many) things wrong, and put up a piece that more expert Linux users recognize as flawed. The results of which can vary from mildly amusing to somewhat harmful if the author has given either a very wrong impression or faulty advice that might lead others to futzing up their systems.

Technology knowledge tends to be highly compartmentalized. There exists no person who can explain every facet of each kernel subsystem, let alone write code for each of them. Likewise, a database administrator may not be the person to talk to when setting up BIND. The key is that people should play to their strengths, and not attempt to exceed their own capabilities. The author here has delved far too deep with apparently zero oversight.

So, by all means, a response is called for. It’s the flavor of response that I disagree with. Where some might approach the author and gently correct their errors, and perhaps even offer help and resources to allow them to create better materials, others take the piss out of the hapless author and try to demonstrate their superiority by ripping mercilessly on the author.

Ubuntard is not about being nice, coddling people and patting them on the head when they make mistakes, telling them, “You’ll do better next time, champ!”. Line-by-line analysis of technical articles is supposed to illustrate the author’s incompetence through many egregious mistakes. As has been said before, someone who posts such factually-inaccurate dreck is likely to be a lost cause.

Which brings me to the title of this post: Real Linux advocates educate, not humiliate.

This is the same sort of exclusionism that was being criticized moments earlier. To dismiss all other view points as invalid is to put your own opinions on a tremendous pedestal. The fact is, certain actions do deserve bitter rebukes, regardless of original intent. A user who tells another that you can copy a disc via dd if=/dev/cdrom of=/dev/sda deserves punishment because their misinformation would likely cost a trusting newbie their boot disk. While the advice given out in the original article isn’t quite that egregiously wrong, teaching rm -rf dir $DIR as the appropriate way to remove a directory is bad.

As for the title of this post, I am referring to the fact that more exists than the extremes. Criticism is not useless. While highly acerbic, the past post did provide the exact reasons why each piece of advice was flawed, painting it a nice shade of grey. It’s at once mocking, yet informative.

The Linux community is, unfortunately, saddled with a reputation for being populated nearly exclusively with arrogant elitists who enjoy mocking noobs and generally being asses to people who know less than they do. The “Ubuntard” response to this author simply reinforces this attitude.

Worse yet, instead of helping to correct the situation they’ve put a writer or journalist on the defensive and made it entirely likely that the author will either stop writing about Linux and FOSS out of fear of getting it wrong again, or ensured that they now have a negative attitude about Linux which is unlikely to benefit the community at large should they choose to cover Linux again.

We should be better than this.

What’s being described here is not what’s happening in my earlier post. There, the original author is mocked for getting things consistently wrong, taking the time to type the individual commands into a terminal yet not bothering to actually test them, and, most importantly, spreading this disinformation in a location where it’s likely to be encountered by newbies who may well perpetuate the disinformation cycle.

Furthermore, with the substantial exposure the original post got, it’s almost assured that the author was notified of the inaccuracies, yet no effort has been made to correct them. I don’t see that as a well-intentioned newbie posting misguided tips, I see that as a mediocre journalist posting a hacked-together list to get a few pageviews without regard for the aftermath.

There is nothing inherently wrong with a lack of knowledge, but turning to evangelism with insufficient knowledge does nothing but damage the reputation of what you’re attempting to promote. I’m firmly not in the “Linux for everyone” camp, but even for those that are, I can’t see poorly fact-checked top-10 lists being of any use to the Linux community at any level.

Not for a minute am I suggesting that the article go uncorrected. By all means, drop the author a note or leave a comment that gently corrects the errors. For added bonus points, offer to help spot-check the next piece they care to write, or point them in the way of some better resources.

I give the original author credit for a few things:

• They actually sign their name to the things they write. In my book, you lose several thousand points of credibility unless you’re willing to use your name when writing, unless you have genuine reasons for not doing so. (i.e., someone living in an oppressive regime that fears reprisal for political commentary — not someone taking the piss out of named tech authors online.)
• They were actually trying to be informative. OK, minus many points for missing the mark, but they tried.
• They have enthusiasm for the command line. Getting some CLI stuff wrong doesn’t irk me half as much as the “OMG, not the CLI!!” pieces that pop up so often about how Linux is unsuitable because you might need to touch the shell once in a great while.

I don’t see how anonymity (or lack thereof) should factor into this. Choosing to withhold your details does not say anything about your character, contrary to what “If you aren’t guilty, then what do you have to hide?” fear-mongers would have you think.

The results of ‘trying’ have been discussed several times, and I believe spreading disinformation vastly outweighs the benefits of one advocate.

The last point is valid, in that it’s rare for a Windows-centric journalist to not immediately deride *nix for its archaic command line roots. However, as above, teaching the rm -rf-as-default doctrine does nothing but harm.

That’s material that can be worked with. Yes, the original article was flawed. But if approached correctly, many of the writers who want to cover Linux can be helped to get it right. And what we want, at least what I want, is more people spreading the word about Linux. The odds of the humiliation approach being effective are zero. You’ve just gotten off on the wrong foot with a potential advocate and this approach will do absolutely nothing to correct the next newbie journalist before they write their first flawed article, because they’re not going to read this sadistic diatribe before penning their first piece. It’s only perpetuating the arrogant elitist stereotype, which many others have been working quite hard to correct.

Our opinions differ hugely on this matter, so I won’t address the first bit, as I can’t do so objectively. It’s true that humiliation causes people to retreat, but that’s not far from the intent. The elitist attitude is not all-encompassing. Most of the cynical “RTFM” answers given out in IRC channels and on forums have been quashed, and I, along with many, many others, spend time helping new users that are legitimately interested in Linux, yet have difficulties that can’t be solved with a five-second search.

The fact is, Ubuntard is not mainstream Linux reading material. It’s cynical, it’s caustic, and it’s an outlet, albeit an entertaining one for those with similar perspectives. Generally, it draws mostly from the latter half of “contructive criticism”, and that’s fine.

It’s important to remember that most journalists are generalists, not experts. And right now, most of those generalists are scrambling like mad to meet deadlines and try to pump out as much material as possible. Yes, they’ll get things wrong from time to time. Most people eff up in their jobs from time to time, the primary difference is that journalists and bloggers do so in full public view. Nearly every error is going to be on public display. (Plus the ones made by subsequent editors and people in the editorial chain, assuming one is working for a publication that still has an editorial chain, but I digress…)

As I mention in response to the first paragraph, technology knowledge is hypercompartmentalized. Combine this with the generalism of journalists and the often-minimal editorial oversight of online publications, and authors are suddenly left largely to their own devices. Those that do not adapt to this deserve to be ridiculed, for they are spreading disinformation by not checking their own facts at any level. If you’re a new user and encounter a guide such as this and can’t differentiate fact from fiction, you’ll likely blame the software, not the author, as it’s commonly assumed that journalists have some amount of integrity and check their own facts.

It’s okay to get a few facts wrong when you’re writing a thousand-word piece on climate change, but writing a short top-ten list wherein more than half the commands are demonstrably broken? That’s shoddy by any metric.