START here

Nonsense in the cycling industry

My thoughts about why and how the patents that are not very reasonable, or outright unsafe, arrive and remain in the cycling industry. This is my opinion (based on my education, knowledge, and experience).

Table Of Contents (T.O.C.):

  1. The story of the threadless (“Ahead”) forks & stems
  2. Cycling industry trends
    2.1. Ignoring common sense and ready-made solutions
    2.2. Complicated and expensive, but at least unreliable 🙂
  3. “You ride too hard”
  4. A way forward?
  5. Zeitgeist and “AI
  6. Instead of a foreword
    6.1. Great article, I agree
    6.2. You state your own opinion as a fact
    6.3. Hung up on disc brakes or other specific technical aspect
    6.4. You are against technical innovation


1. The story of the threadless (“Ahead”) forks & stems

The story of how threadless forks (and stems) have replaced threaded forks with quill stems (well, for the most part) nicely depicts how the cycling industry works.

Quill stem vs ahead stem
Quill stem (1); bolt that holds it in place (2); quill (3); “ahead” (“threadless”) stem (4)
Picture 1

I’ll start by stating these three facts:

  1. Threaded forks with quill stems have been used for decades.
  2. Threaded fork systems make headset-bearing preload adjustment slow and tedious.
  3. Quill stems are unsafe.
    If that’s not clear from the picture above, see the explanation in my article about unsafe cycling industry patents.

I’ve had a quill stem bolt break. During a ride. That urged me to look into whether that could be prevented. I wrote my findings in the above-linked article about unsafe cycling industry patents – briefly: there is practically no preventive maintenance or pre-ride checks that can negate the risk of the one quill-stem bolt snapping during a ride. Over the past decades of fixing bikes and getting feedback from other mechanics and cyclists, my conclusion is that it does happen. Not very often, but it is enough for it to happen once… on a downhill. I am mentioning this because, whenever I talk about this particular problem, there’s always at least that one guy who says “Quill stems have been in use for decades and no one has had those problems!”

Today, we have threadless forks (and matching stems). Did someone in the cycling industry design them in order to fix the quill-stem problems? A bike manufacturer like Giant, Schwinn, or Cannondale? A frame manufacturer like Bianchi, or a frame builder like Tom Ritchey or Richard Sachs? A component manufacturer like Shimano or SRAM? No! 🙂

Bicycle and frame manufacturers were perfectly happy as long as the sales were good.

It was an MTB racer enthusiast John Rader who came up with the elegant, strong, reliable, and light (I will come back to this last part) solution that we call “threadless fork” today.

The solutions I could come up with were very non-elegant (awful-hack ™ 🙂 ).

He made a prototype (used for his racing) and managed to arrange an appointment with Dia-Compe‘s USA branch representative Peter Gilbert. He brought his MTB to the meeting and demonstrated how quickly he could remove the fork from its frame by just loosening two bolts. That was impossible to do with a threaded fork.

Gilbert was a salesman, but he did see mechanics work and struggle, so he realized how this patent would speed up the work (lower weight was also a big plus). That, not the quill stem safety concerns, got him to seriously consider the new patent.

However, an elegant solution for adjusting the headset bearing preload was yet to be found. They thought about cutting threads on the inside of the steerer tube, and had several similar non-elegant ideas. Then, Dia-Compe USA’s product manager, Doug Beeler, had a wheel fall off his office chair.

While he was looking at how to solve that “malfunction,” he saw a star insert inside the chair’s tube. Eureka! 🙂 The star-nut was born!

Of course, with modern carbon forks, you don’t hammer in a star-nut, but use a system similar to the one used to keep quill stems in place. The major difference is that the insert is used only to help adjust the headset bearing preload, while the stem is mounted and held on securely with two bolts.

The patent was initially named “Light Set,” but Dia-Compe’s marketing wizards renamed it to “AheadSet.” Thanks to their already existing cooperation with the RockShox company, the production and sales of threadless forks went well and soon overtook the market.

The Dia-Compe’s USA department became independent in the meantime and named themselves Cane Creek.

Related article:
Bicycle fork and headset bearings’ types and construction.

– T.O.C. –


2. Cycling industry trends


2.1. Ignoring common sense and ready-made solutions

You can draw your own conclusions from the story above. My impression is that the cycling industry is driven by sales and marketing primarily, with a rather poor engineering side of things.

The story of the threadless forks is with a happy ending. The story of the modern road bicycle calliper brakes is not so good (the problem with modern road bicycle rim brakes).

I almost forgot the well-known and still present pedals-to-cranks mounting problem. This problem was solved about half a century ago in the automobile industry. What problem am I talking about?

Automobile right wheel bolts were prone to unscrewing, and all the wheel bolts were prone to breaking. The problem was solved by introducing a tapered interface between the wheels and the bolts/nuts (that we can see on all the modern lug nuts and wheel bolts).

Automobile wheel to lug-nut tapered (conical) interface
Automobile wheel to lug-nut tapered (conical) interface
Picture 2

This looks so simple and obvious, but until the invention of this patent, the right wheel nuts/bolts had to use a left-handed thread to keep them from loosening (today, all the wheel nuts/bolts use the standard right-handed thread). There were also problems with bolts breaking due to more “motion” (metal fatigue).

Here is how the late Jobst Brandt improved his pedal-to-crank interface in order to prevent fretting damage to the cranks:

Jobst Brandt's modified pedal to crank interface
Jobst Brandt’s modified pedal-to-crank interface
Picture 3

This patent also removes the need for the left pedal to have a left-handed thread to prevent unscrewing – making pedal manufacturing simpler, cheaper (left, and right pedals would be identical), and making life easier for any tandem bicycle riders.

Let me add one more example:
In the instructions for mounting square taper cranks, most manufacturers and industry authorities recommend a dry mount. I wrote a tutorial on how to mount square taper cranks. There, I explained in great detail the theoretical (engineering) and practical reasons why I use and recommend anti-seize paste for mounting these cranks.

Now who is crazy here?! If everyone keeps shouting that the “Earth is obviously flat,” should I play dumb and accept that?

When I talk about these problems with other cyclists or mechanics, most pretend like they (the problems 🙂 ) don’t exist, i.e. “that hasn’t happened to anyone.” To me, that looks like putting one’s head in the sand and eating the cycling industry marketing with a spade.

– T.O.C. –


2.2. Complicated and expensive, but at least unreliable 🙂

The cycling industry mostly invests in marketing, not engineering. When I dare to criticize some of the “modern inventions,” I often get angry rants from cyclists who pay (needlessly?) large sums of money for those. That is how powerful marketing is.

Some of the things I find “delightful:”

Disc brakes, even on road bikes, or bikes used mostly in dry weather. Today, it is difficult to find a new bike with mid or high-end parts without the disc brakes. I wrote about the bicycle disc brake pros and cons (compared to rim brakes). Based on that, I would say that disc brakes make little sense on road bicycles, but most companies today leave you no alternative – you must go with the more expensive and more complicated option.

Talking about road bikes, modern brifters (integrated shifter and brake levers) are very expensive, complicated to repair (made not to be repaired), sensitive to impacts and easily damaged. Marketing has convinced cyclists they should shift gears every second (as if the human body is an inline four motorcycle engine).

I actually did manage to repair a few Shimano STIs, for example, but that is precisely why I’m convinced they weren’t designed to last or be repaired.

If I haven’t been riding for decades in mountains, mud, rain, and snow with rim brakes and friction shifters, maybe I too would buy into the hype.

In and of itself, none of that is a problem. The problem is what happens when those things break down, which leads us to the next chapter:

– T.O.C. –


3. “You ride too hard”

A great majority of cyclists ride very little and very gently (low strength, stamina, and too little experience and skill to ride aggressively). Bicycles are often used for light recreation and a couple of joy rides. That’s it.

Now, if you ride a lot, far, often, or if you are using your bicycle as the main means of transport (bicycle commuting), the situation changes. You are a lot more likely to put the frame and all the parts to the test and see what breaks. When the complex, expensive, and too-sensitive stuff malfunctions or breaks, the blame is usually put on the cyclist. “You ride too hard” and similar nonsense.

The thing is that many (most?) bicycles, frames and parts are not built for riding (except maybe a few times per year, in fair weather, to the local park). As soon as you “step on them,” they show their weakness, and then you, the cyclist, get the blame.

– T.O.C. –


4. A way forward?

In the automobile industry, the situation was similar to the above-described, until the popularization of Formula 1 racing. F1 brought enormous amounts of sponsor money, and winning the Grand Prix brought prestige that boosted sales. Only then did the top engineers and teams get hired and paid to do some real improvements, that “trickled down” to the commercially sold automobiles.

Bicycle racing doesn’t bring nearly as much money. Also, the emphasis is put on the cyclists, while any equipment improvements are strictly controlled and regulated (to not provide any unfair advantage).

For all those reasons, the cycling industry has no incentive to build better, especially not more durable and robust stuff.

The elegant and reliable quick-release is being replaced by “thru-axles.” Friction shifters are next to impossible to source (if you aren’t a fan of the indexed shifters or road bike brifters). There’s suspension and hydraulics (and electronics) on everything, even seat posts! Carbon fibre parts have replaced the good old steel, even when it makes no sense (spokes for example)…

In my opinion, a bicycle should serve me, not the other way around. It should be robust, reliable, cheap and simple to repair. I like riding to the edge and crashes are a normal thing. My concern should not be if I’ll need to take a bank loan to fix my bike after a classic, normal wipe-out and slide. 🙂 “Don’t buy upgrades, ride up grades,” to quote Eddy Merckx.

However, the cycling industry is driven by marketing and what can be sold well, for a lot of money. Combine that with the average cyclist who doesn’t ride their bike very much, and any problem or malfunction can easily be blamed on “riding too hard,” instead of fixing the design (or quality control) problems.

I understand that many (perhaps most) cyclists enjoy the fancy parts and apparel. That’s OK. Everyone should choose for themselves. But I don’t like having fewer and fewer “robust and reasonable options” (to put it that way) when shopping.

They’ll gladly sell you a helmet, but won’t do a decent frame and fork quality control.

Relja Novović – to immodestly quote myself

– T.O.C. –


5. Zeitgeist and “AI”

I already touched upon this line of thinking on my blog, in the article: Zeitgeist: “Fake it ’till you make it!”

The above-described problems have been common in the cycling industry since the 20th century. In the 21st century, they have become prevalent across practically all industries and areas of life. I can confirm that in my areas of expertise (mechanics and computers/IT), that definitely is the case.

People don’t read anymore (you can check yourself by asking how many books you’ve read over the last 365 days?). They “skim” articles and social network posts, absorbing just the occasional image, heading, or short video. I am convinced that this affects how children’s and young people’s brains (synapses) are developed (though I’m sure it doesn’t help older adults stay sharper either).

Besides that, people are no longer worried (or scared) when they don’t understand how (and why) something works. I’m not talking about my art-historian aunt not understanding what happens when she presses “send” in her Gmail account. I’m talking about the average professional web developer not caring to learn about it because “it just matters that the website works” (that is just one of the countless examples).

With the development of technology and artificial intelligence, appliances are getting smarter, while people seem to be getting stupider and more enslaved. I’m saying this as an enthusiast who actually loves technology and tries to use it for the good. There seems to be an epidemic of superficiality.

The consequence of all that is that people can’t think critically – check information, make logical deductions, or find errors in logic and conclusions.
My article about the freedom of speech, science, and critical thinking

All that is among the things that make people easy to manipulate and enslave. Just think of the countless “(Internet) misinformation tackling” campaigns and measures. Apparently, the folks in charge don’t trust us to be able to realize that the person saying the Earth is flat is an idiot. The tragedy is that their assessment, for too large a percentage of people, is probably correct. An even greater tragedy is that there are no real efforts towards raising consciousness, improving education, and enlightenment – it all goes towards even more dumbing down.

The funny and ironic thing is that I don’t believe that to be some conspiracy or a conscious planned effort. No. It looks more like a combination of going with the flow, not really caring (or thinking forward, while not being selfish), and simple human stupidity (even, or especially, at the top levels). It is in human nature to invent a computer, just as it is to become lazy with a computer and stop & forget how to calculate “on paper.”

I will now briefly touch upon artificial intelligence (AI). Inventing AI is also in human nature – we are urged to see what we can do and make. Let me make a brief digression:

In 1945, Americans invented the atom bomb. After that, Russians had to make their own. Neither of the superpowers was willing to fall behind in the nuclear arms race (out of fear of the other one). As I’m sitting and writing this, some humans have the power to press a red button and kill us all.

Don’t forget we are talking about species so primitive that they think electric cars are a jolly good idea, while some of its most intelligent members, in their prime years, spend their time, energy, and intellect trying to figure out how to boost the sales of product XY (while the dumber ones are writing articles 🙂 ).

Now, where was I? Oh, yes – the AI. Like with nuclear arms, no one dares to fall behind in the AI development race. I mean, we can’t let the Americans/Russians/Chinese/You-Name-It overtake us, can we?!

The problem is that even those who are developing AI don’t really understand how that thing works, and how exactly it “learns*.”
* For a more detailed explanation of this aspect, see the comment by my friend who remained anonymous (“M, as in Tangerine”) below, with whom I mostly agree.
What we do know is that AI can improve exponentially. Exponential growth is like British humour – strange, and no one really gets it. Alas, AI does not have to become self-aware to ruin us. Nope. It just needs to become a powerful program/machine and get a bug. It’s a program. Programs get bugs. And no one does proper beta-testing anymore, they just rush to production to meet the deadlines, so the higher-ups can get their bonuses. Combine that with a “program” that learns and grows exponentially quickly, without proper control, and you could end up in a tricky situation.

Hence, in the best-case scenario, AI will let humans become even more stupid (a lot more than digital calculators, computers and smartphones did). We will no longer need to Google, read, try to see if it makes sense and figure it out – AI will just give us all the answers, specifically for us, in the exact context of our question(s).

In the worst-case scenario, AI will calculate that the optimal solution is to press that red button (due to a bug in the program, it needn’t involve any consciousness). The monkeys making and perfecting it can’t really control it because they don’t understand how – “it’s important that it works.”

Of course, the fact it is all going to hell does not mean you should not enjoy life as much as you can. After all, you might be lucky enough to die before robots take over the world, and humans forget how to read. Who knows? 🙂

This started out as a story about the bike industry, but I spontaneously felt like expanding the topic a bit. Those who mind that have probably not even reached this part, so it’s all good. 🙂

If you’ve actually read this whole article, I believe you are among the 1% “strange” people who are not “in tune with modern times.” Thank you. I am curious to read your thoughts about this (pun intended 🙂 ) – in the comment section below or on the BikeGremlin forum.

– T.O.C. –

6. Instead of a foreword

This article received a lot of praise, but also some criticism (most of it via email, various messengers, and Facebook groups). That is good. It is nice to see people reading and thinking. Some feedback, both positive and negative, was quite passionate. I’d like to thank everyone who’s praised, criticized or commented.

Most of the feedback fits into just several categories. Let me discuss each:


6.1. Great article, I agree

The basic idea of this article was to share my opinion and perhaps nudge people to think about it. With that in mind, I would rather hear you disagree, but think with your own head (and politely & constructively explain where you think I was wrong).

Of course, it is nice to know there are people who think alike and see the problems that I’m seeing. Noticing problems is the first step towards solving them.

– T.O.C. –


6.2. You state your own opinion as a fact

A vast majority of my articles are objective and technical, with information, tables, and step-by-step tutorials (for more details, see: “Is BikeGremlin a reliable source of cycling and mechanics information?“). This article however is in the section called “My projects, work and thoughts,” and at the very start, in bold font, it says: “This is my opinion (based on my education, knowledge, and experience).”

Imagine what the article would look like if I started each sentence with “In my opinion…” 🙂 This is an article with a bit more liberal form compared to a scientific paper, with all the limitations, pros, and cons of that.

I would also add that flat Earth has been a “fact” for centuries – i.e. you could argue that facts are “opinions on which a majority of experts in a field agree.” I discussed this in more detail in the article “Freedom of speech and science.”

That’s about it for the opinions in general. Now, about my opinion(s):

I have a lot of knowledge and experience in the topics I write about (a part of that can be seen in the hundreds of articles and videos I’ve published so far). You could even say I’m an expert. 🙂

In addition to that, thanks to my work, workshop, websites and YouTube channels, over the past decade I’ve answered thousands of questions and either fixed, or helped thousands of people fix their bikes.

That’s the essential distinction to make. Let me make an example:

  • I don’t like “And Quiet Flows the Don” by Mikhail Sholokhov.
    That is an opinion based on personal tastes and preferences. I’m not disputing its literary quality and importance. It’s just that it didn’t “click” for me.
  • I don’t like bicycle disc brakes (at least for many use cases).
    That is an opinion based on knowing and weighing all the pros and cons. I would call it “an objective opinion.” When a company pays a consultant to improve its security, a good consultant will use a similar “objective opinion” to make recommendations, based on investigating and understanding the company and all the available solutions’ pros and cons.

So, my question is:
If I can’t or shouldn’t state “my opinion” about the problems and downsides I notice, then who can, who should?

You are free to disagree. In fact, I would encourage that. Constructive discussions and criticisms are a way to improve knowledge and come up with new ideas. But it is “my opinion” that you should not forbid me to state “my opinions” (in a polite and constructive manner). 🙂

This leads us to the next type of objections:

– T.O.C. –


6.3. Hung up on disc brakes or other specific technical aspect

I’ll use disc brakes as an example, but the principle is the same for any other technical aspect.

So, using disc brakes as an example, here are my arguments:

They are not the optimal choice for every use case.
Like everything else, disc brakes have their pros and cons. For some use cases, the cons outweigh the pros.

It’s great if you prefer disc brakes. It’s not that great that manufacturers are leaving us less and less choice if we don’t want them (it’s very hard to find a bicycle with decent-quality equipment that doesn’t come with disc brakes and a suspension, or “at least” a carbon fibre fork).

Marketing brainwashes people into believing that disc brakes are superior (in every aspect) and that they shouldn’t dare to ride off-road in the mountains without disc brakes. If I hadn’t been riding rim-brake bikes for decades, I too would probably have believed that.

– T.O.C. –


6.4. You are against technical innovation

Nope. I love innovation. I am against poor (and unsafe) engineering, and brainwashing people to buy expensive, unreliable and unsafe stuff for profits – when it is easy (technology already exists) to make them safe, robust, and reliable (instead, those are being withdrawn from the market).

Shimano has finally publicly admitted that there is a problem with (some of) their cranks:
https://bike.shimano.com/en-US/information/customer-services/corrective-actions/11-speed-hollowtech-ii-crankset-recall-notice.html

Is it forbidden to criticize other stuff until a big company’s CEO approves that?

BikeGremlin forum discussion about this article:
https://www.bikegremlin.net/threads/nonsense-in-the-cycling-industry-article-comments.35/


Help BikeGremlin
stay online & independent

This website is educational, free, objective, and not commercial
(sponsors don’t enjoy paying if you mention all the product downsides that you notice 🙂 ).

How much does a WordPress website cost?

If you find this site to be good and helpful,
and if $5 per month is what you can afford to set aside,
please consider supporting my work with a Patreon donation:


patreon.com/bikegremlin


– T.O.C. –

4 thoughts on “Nonsense in the cycling industry”

    • Cloudflare have published their report on the problem.

      They always seem straightforward in explaining the problem causes, which is admirable.

      The full report (Cloudflare’s blog link):

      https://blog.cloudflare.com/post-mortem-on-cloudflare-control-plane-and-analytics-outage/

      Now for the less admirable part. The problem seems to have been caused by two things:

      – Power outage at one of the CloudFlare’s data-centres (electrical company had a problem, and data centre operators made some mistakes).
      – CloudFlare’s fail-over systems didn’t work.

      A “jewel” from the electric company:

      “…the overnight shift consisted of security and an unaccompanied technician who had only been on the job for a week.”

      CloudFlare’s “jewels”:

      We believed that we had high availability systems in place that should have stopped an outage like this, even when one of our core data center providers failed catastrophically.”

      “We were also far too lax about requiring new products and their associated databases to integrate with the high availability cluster. Cloudflare allows multiple teams to innovate quickly. As such, products often take different paths toward their initial alpha.”

      This ironically touches upon what I wrote about in this article. Believing instead of testing (apparently), and fast innovation before really understanding how stuff works.

      To be fair, CloudFlare is probably among the least bad “cloud” companies, they always learn from their mistakes in order to prevent them from re-occuring, and, for the most part, they seem to know what they’re doing (which is rare nowadays).

      A huge number of appliations and websites rely on CloudFlare.
      BikeGremlin sites are configured so that I could move away from their service relatively easily, but at the cost of slightly lower performance, and higher expenses (and this is a non-profit, independent website, so I don’t like the idea of increased expenses to keep it running).

      Relja RetroGrouch Novović

  1. Relja,

    Congratulations on the article you wrote. I really like your writing style: direct and straightforward. I agree with everything you wrote in the section five.

    Here are my two cents on the subject. My view on AI is quite pessimistic and I think that what we now call AI is actually a dead end from which we will get out only when the invested funds pay off through some nicely packaged software solutions. In addition, I think that the notion of artificial intelligence should be reconsidered, because it seems to me that we need another name for what we call AI today, and at the same time we are still looking for something that could be called intelligent. Theoretically, AI is divided into symbolic AI (based on formal knowledge representation) and machine learning (based on statistical models). This first category is also called good-old-fashioned AI and is most similar to teaching students to solve some tasks (we first teach them a problem-solving method, and then we give them tasks in which they should apply the given method). The second approach, which trains artificial neural networks to recognise some patterns in the data, is the most similar to the way babies learn (we show them how they should or shouldn’t do something, until they learn it).

    As reliable as the first approach is, it requires a huge initial effort to prepare that knowledge base. As you said yourself, today few people delve into a problem, preferring to look for instant solutions. In addition, the accessibility of computer equipment and the availability of data have led to the fact that it is easier for us to let the machine “learn something by itself”, but I don’t know who is ready to take responsibility for the decisions that such trained software will make. Explainability of those decisions is also one of the biggest challenges in machine learning. It passes only because nowadays hardly anyone is interested in detailed explanations, because we are not able to follow an even more complex chain of causality nor to critically review the information (in support of the story about the flat-Earthers you mention). It’s as if the only thing that matters is that the answers the AI gives sound good, and their truthfulness has become a secondary concern. Just as you described web developers who don’t care how exactly their application works (and they don’t even distinguish the Web from the Internet), so today’s AI allows you to develop software even though you don’t have the necessary domain knowledge, but you somehow got a large enough training data set.

    In any case, while we wonder if we are close to a scenario from a sci-fi movie, AI can at best only imitate some already seen patterns of reasoning, so I don’t even know how well it fits into the understanding of intelligent behaviour. This brings me back to the question from the beginning – do we need another name for what we call AI today?

    If you need a suggestion on what could be improved in the text, it would be the sentence: “The problem is that even those who develop AI do not know exactly how this thing works and how it learns.” Basically, I believe they know how AI learns, but in practice AI is mostly a black-box.

    That’s why I would rather write:

    “The problem is that those who develop AI have an idea of how this thing learns, but not always an explanation of why it works the way it does.”

    • Thank you for taking the time to write this comment.

      I’ve included a link to this comment in the “disputed” article section, because the comment nicely corrects, explains and expands the topic for those interested in learning more; without making the article longer and broader.

      The comment above was written by my friend for whom I can freely say that they are an expert in this field and one of the most intelligent humans I know (since the comment is anonymous, I guess you’ll have to take my word on that 🙂 ). It was originaly written in Serbocroatian, and this is my free translation to English.

      Relja

Comments are closed.


Please use the BikeGremlin.net forum for any comments or questions.

If you've found any errors or lacking information in the article(s) - please let me know by commenting on the BikeGremlin forum.
You can comment anonymously (by registering with any name/nickname), but I think it is good to publicly document all the article additions (and especially corrections) - even if their author chooses to remain anonymous.

Skip to content