If you are a coffee lover, you’re probably already well aware of the online coffee courses offered by Seattle’s ChefSteps. That is, unless you’ve been holed up in Guantanamo or you avoid most social media like the time-sucking plague that it is (though I do, and yet I couldn’t avoid the subject).

This week coffee legend James Hoffman blogged about the latest ChefSteps course he’s involved with, and today we’ve already witnessed the promotional marketing for it bleeding out to publications such as Eater and Food & Wine — complete with obligatory use of the word “perfect” in their article titles.

ChefSteps offers the 5-star-rated Espresso: The Art of Extraction

To read the general press and discussion about publishers like ChefSteps, you’d think we were entering a revolutionary era of coffee education. But having taken ChefSteps’ free Espresso: The Art of Extraction course, and having a lot invested in the subject of how online education works (and does not work), I’m just not feeling the love. At least yet.

Of computer mice and men

Let me explain. To begin with, I also need to start with a little about my “day job”. For over the past three years, I’ve been the co-founder and president of an online education start-up. We raised several million dollars from Khosla Ventures, a heavy investor in education technology, where Vinod Khosla himself sat on my board of directors. The only other board he served on was Square, so we were fortunate to have so much of his attention. (Side note: Khosla Ventures has more recently purchased the start-up outright for integration within its educational portfolio that includes the CK-12 Foundation, etc.)

Coincidentally, representing this start-up at TechCrunch Disrupt SF 2012 is how I first encountered Blossom Coffee. And for those of you who have seen the brilliant HBO show, Silicon Valley, TechCrunch Disrupt is exactly like that.

Now the mission of this start-up may have been teaching users to program rather than to make latte art, but many of the frameworks and principles apply regardless: delivering educational content online, complementing lessons with resources such as online video, embedding discussions and interactions with the faculty and other students throughout the course, etc. Thus, over some years of experience, I’ve learned a few things about what’s good, what works, what’s revolutionary, and what’s — well — not so much in online education.

Back to that latte art

First: ChefSteps’ content. Taking their Espresso: The Art of Extraction course, I didn’t come away learning anything I didn’t already know. But as an introductory course, that’s to be expected. The videos, hosted on YouTube, are slickly produced and include legit instructors: Charles Babinksi, of G&B Coffee (he’s the ‘B’) and a former USBC finalist, and Scott Callender of La Marzocco Home.

The chief questions I struggled with as I took the course: was the material any better than any book or series of online articles I’ve read before? Same question for the videos. And did all of that help me learn something? My answer was generally “no”, or at least “not really”.

CoffeeGeek legend Mark Prince chimed in last week on a little of his mixed experiences with ChefSteps — at least regarding the “Espresso Myths” videos in the course. So let’s take the video on crema myths:

Now ChefSteps’ byline is “cook smarter”, but being “smarter” usually involves a more thorough effort of evaluating multiple perspectives, checking out references, and maybe performing a little analysis on all that before drawing hard conclusions. What we have instead in this “crema myths” segment (at 1’11’) is Scott Callender saying this:

“I think one of the best examples of that is Italian roasters include robusta into their blends simply to add this really thick, dark crema on top of their shots so it looks beautiful. But if you ever just taste a single origin robusta, most people would not tell you that tastes like a very good espresso.”
— Scott Callender, Espresso Myths: Magical Crema

I can almost get past the fact that Scott has essentially stereotyped an entire coffee culture by suggesting that Italian roasters blindly add robusta to blends for the sole purpose of enhanced visuals. But what really makes my eyes roll is that Scott dismisses the idea that anyone might add a robusta component to a blend for something as insane as flavor balance or complexity.

Scott’s attitude is also rooted in the Puritanical myth — common to many myopic self-described Third Wavers — that the ultimate expression of coffee can only be found in a coffee bean’s genetically and geographically isolated single-origin, single-farm, single-row-of-shrubs heritage, unadulterated by external contaminants. This is essentially a lite version of Adolf Hitler’s purified master race doctrine as applied to coffee. And yet some of the greatest pleasures of coffee today come from an incestuously muddled history of genetic and geographic mash-ups; mash-ups that have given birth to everything from Bourbons to Catuais to Caturras to SL28s to SL34s to Typicas to even prized Gieshas transplanted to Panama as recently as 2000.

Charles Babinksi (who later uses big words like “quotidian”) then adds to this deconstructionist nonsense at 3’40” in the video:

“Also, it should be noted that crema tastes terrible. It’s one of the least enjoyable parts of drinking coffee. And more crema is not necessarily going to mean a tastier shot.”
— Charles Babinksi, Espresso Myths: Magical Crema

While Charles is factually correct, what he says reflects a deconstructionist and non-integrative approach to thinking — i.e., that any component that isn’t good individually in isolation is therefore potentially negative, detrimental to quality, and/or not important. This line of thinking borders on implying that nothing can be better than the individual sum of its parts, which is just plain wrong.

The multi-dimensional representative taste types of coffee, courtesy The Coffee Cupper's Handbook 4th Ed. from Ted Lingle

As Charles points out later in the course, in Taste the Extraction, progressively tasting an espresso extraction highlights how it transforms from sour-through-bitter notes and yet they all balance out in the end. That balance is arguably one of the most critical elements to a quality espresso and coffee in general: sour is important, sweet is important, salty is important, even bitter is important, and the balance between them all is what makes the beverage we obsess over.

I’m all for dismissing unnecessary espresso myths, but in the process you shouldn’t be creating new ones in their place.

Where ChefSteps succeeds

Where the course excels is in introducing the “three legs of the espresso stool”: brew ratio, brew time, and brew temperature. Again it’s nothing that hasn’t been repeated before dozens of times elsewhere on the Internet (despite many student comments in the course to the contrary), but it’s summarized well in a concise place and format.

Overall the course is a bit short and superficial (hey, it’s free), serving mostly to improve general awareness rather than to teach any skill, method, or technique. Segments such as Pulling a Great Shot, for example, do very little towards the mission of actually teaching. Instead, it repeats a lot of minimalist common knowledge to a soundtrack of lounge music more suitable for getting a hot stone massage at a spa:

But is it an actual “course”?

Last, we come to ChefSteps as a concept and overall learning format. Here’s where that rubbish about the day job kicks in. Is what ChefSteps offers any different or more effective than a book with a supplemental DVD of videos?

The threaded comment section to engage with the instructors and fellow students is helpful, but it feels a little wonky in the context of a course. It’s optimized more for commenting on Facebook posts than to facilitate any actual learning exchange, but it’s the easiest and most obvious thing to do in the early stages of any start-up learning platform. There are also Quartz-style contextual comments, but they barely get used.

The support component of any course — where students have questions or challenges that veer from the linear narrative of the program — is essential to its effectiveness. It is a core differentiator from merely reading a book or watching a video. One of my most critical insights was that many learning platforms are designed primarily as modified content delivery platforms, and support is often bolted on as an afterthought. My start-up’s platform was initially built around content delivery rather than support delivery, and effective learning platforms often require the reverse.

In other words: it’s the support, not the content, that’s often the primary driver of learning value. We choose university courses more for the professors and TAs than we do for the textbooks they use.

The discussion comments help, but is this a post or a course, really?

However, the primary glaring omission from ChefSteps as an effective learning platform is its complete lack of assessment or evaluation capabilities. A few years ago the Gates Foundation developed a compelling universal data model to represent learning systems and environments. While not yet made public, it was extensively shared among many tech-minded educators as a potential learning blueprint. In this fundamental data model, learning resources are paired with assessment or validation resources, and the pairs are typically combined in a sequential series. That much is so fundamental to the human learning model that the Gates Foundation proposed it as a standard.

And it makes obvious sense when you think about it. What education is there without quizzes or tests, midterms and finals? The human brain simply does not learn in the absence of useful feedback, without tightly integrating the practice of the very things you are supposedly learning.

You can only go so far learning to play a piano just by reading a book and watching videos. Hearing how you actually sound playing a piano — or better: having your piano teacher assess your performance — is the only way to know if you’re really progressing.

Hence why Charles Babinksi wisely suggested that ChefSteps streamline, if not eliminate, the course’s latte art section: teaching such a skill with the platform would be something of an abject failure. But without any assessment in the mix, you might make the same case for the entire course overall.

So it remains that ChefSteps is a nice reading resource, but it offers nothing you couldn’t recreate on Facebook other than its paywall. Accept it now for what limited reference value it provides. Any actual online education is a long, long ways off still.

UPDATE: August 13, 2015
Not to be outdone, Blue Bottle Coffee‘s training director, Michael Phillips, has developed a 59-minute online coffee brewing course through Skillshare: Coffee Education for the Masses from Blue Bottle’s Michael Phillips.

Skillshare, as with Chefsteps, fails to integrate a feedback loop for students to gauge and measure their progress and success — thus making Skillshare also no more of a learning platform than a Facebook page sitting behind a paywall.