Because it's not just the big things that are the same; it's the little points of style that a "marketplace of ideas" wouldn't think to share or copy.
You'd expect, if you allowed someone to study your notes before a test, that they'd possibly end up using similar turns-of-phrase to yours on the test. You wouldn't expect that they'd end up learning to perfectly duplicate your handwriting. It'd both be too hard, and not worth it, to do so; and they already have their own handwriting style. So why would they?
Ah, but if you both learned to write from the same person, then you would expect the handwriting to look similar (even if you didn't let them see your notes).
And I think that might be going on here. Take a look at a Chinese-language web site, like http://news.baidu.com
You'll see that it has the "niche" punctuation you mentioned, lots of point form descriptions with emoji bullets, and nearly all the latin script on the page is composed of short strings of all-caps text (many of which are acronyms like "IPO" or "AEX" that would be nonsensical if you didn't already know what they meant).
Some of these stylistic elements are naturally going to bleed over into Amazon listings too.
That's certainly an alternative possibility; more likely, IMHO, than sellers copying one-another perfectly. However...
> You'll see that it has the "niche" punctuation you mentioned
It does, but there are only two examples of it on the page right now. (It is a thing common to Chinese text generally, but it's not the first thing you'd reach for.)【】gets used on this page as a sort of "tag" or "section" for a story. In Amazon product descriptions, meanwhile, it's being used to form a sort of two-tier "【title】body" text; as if compressing a slide-deck slide onto a single line. That's not what those characters are "for", in Chinese. It's a misuse. A Chinese reader would be confused.
> lots of point form descriptions with emoji bullets,
There are no emoji bullets on Baidu; there are styled bullets. But also, when I say "emoji bullets", I don't mean that they use emoji as bullets; I mean that they use regular bullets, and then use emojis as additional "decorations" for each point. Like this: https://i.imgur.com/xW3uPEP.png . AFAIK, nobody does this, anywhere on the Internet, Chinese or otherwise, other than on these brands' Amazon product pages. Because it's silly.
Consider also: if this was just "the way Chinese people write product descriptions on marketplace websites", then you'd expect to see it happening on e.g. AliExpress, or among Chinese sellers on Wish.com. But you don't. On both of those sites, product listings (from Chinese sellers) just use regular, random+inconsistent styling, with a diffusion of different stylistic techniques spreading via natural selection of sellers; with none of these particular techniques being among them. It's only a certain implicit web of a few thousand Amazon product brands, that have this extremely-consistent style.
Can't believe "【Title tags】body text" notation is washing up here. This has to be, to describe in a William-Gibsonian description, an old Japanese practice in email titles heavily used in 2ch/5ch later became popular in Rakuten listings mimicked by Chinese listings on Amazon JP that ultimately leaked into Amazon US via machine translations.
I have never seen it done in anything other than in Japanese text, and translations of. Not sure about double bullet points, but this usage of 【】 is distinctly Japanese.
> You'd expect, if you allowed someone to study your notes before a test, that they'd possibly end up using similar turns-of-phrase to yours on the test. You wouldn't expect that they'd end up learning to perfectly duplicate your handwriting.
Consider a foreign seller who perhaps doesn't have a great grasp of the English language or the cultural context of the US. When you add your own spin to someone else's idea you are leveraging a lot of implicit knowledge to be able to spin it in a way that makes sense. If you don't have that implicit knowledge (and are not going to be penalized for verbatim copying) why try something different from what you have seen be successful already?
What I'm saying is that it would take an immense amount of effort, looking at literally tens of thousands of Amazon listings, to even realize that "this is what everyone else is doing." I noticed these patterns because I personally went through the top 100 items in every leaf-node category of the Amazon store a few months back (because I treat "finding obscure solutions to problems I didn't know I had" as a hobby.) No new Amazon seller is going to do that; and so no new Amazon seller is going to notice every detail of the pattern.
Or, to put that another way: if this were a "marketplace of ideas", there'd be a certain amount of mutation, of copying error, to be expected, from individual sellers not noticing all of the stylistic quirks other sellers use; and instead substituting something random.
But instead, what you see is perfect copying of style, with no mutation or variation, among what are ostensibly thousands of distinct sellers/brands. That's implausible.
(Also, for a bit of a knock-down argument I maybe should have pulled out sooner: when there's an update to the "optimal style" used by these brands? They all change. All at once. Thousands of different brands got rid of the 【】 — replacing it with [] — on the same day, some time last year. Real independent sellers, even if they notice tiny changes in popular style like that, can't react that fast, and don't have time to be constantly updating all their product listings. But a SaaS sales platform with a post-maintenance bot sure does!)
> What I'm saying is that it would take an immense amount of effort, looking at literally tens of thousands of Amazon listings, to even realize that "this is what everyone else is doing."
You're assuming cause & effect here.
To this point, imagine:
- The crowd following the leader / a particular success story
- Most individual sellers are not actually "individual": Same owner or network
- "Educational" boot camps to "get rich". There are towns where people gathered to learn from "the pros" (search for related documentaries)
- Learning from the same person, e.g. via tiktok-esque platform
- Subscribed to update packages
- Generator to fill forms
> That's implausible.
'thousands' is tiny in terms of scale at China.
Imagine yourself as a seller within a network of an extremely competitive network, with a "succeed or hunger" mindset, would you follow and immediately implement any changes that can improve your chance of success?
e.g. New gossip of the day: "Amazon is going to ban any seller using the characters 【】"
Try to relate why being on the first page on HN can bring certain websites down?
I think you are also underestimating what 'travem' mentioned about language and literacy.
Again, imagine you have a limited or zero grasp of the English language. The alphabetical letters are just gibberish to your eyes, but you know you can copy them as your native language characters on your PC/smartphone. Would you attempt to be creative or play it safe?
> Most individual sellers are not actually "individual": Same owner or network
> Subscribed to update packages
> Generator to fill forms
...are fully in line with my argument. My hypothesis wasn't specifically that this was a SaaS system doing this. Rather, my hypothesis was that there's a Single-Point-of-Failure entity or platform that these listings go through — a Borg Queen that could be taken down; and that doing so would stop thousands of sellers in their tracks. I'm agnostic to what form that entity or platform takes.
You might be interested to hear my own guess as to a possible mechanism for this, though, as it's not listed among your alternatives. That guess is that there are a very small number of Chinese companies that advertise their services as writing/managing English-language Amazon product listings, for companies that have no English speakers. These companies use human labor, not automation, but they have a strict style guide, which both informs the format of their output, and the type of input they require from their clients. The network of sellers whose listings look the same, are all the work of one such English-language post-localization company (the largest/most popular one), and so all adhere to one uniform style guide. This company, at least, has the seller themselves register with Amazon; but from there, takes over responsibility for creating products in their account, managing returns claims, etc. They promise to "take care of" every interaction with Amazon FBA that requires English knowledge, and to only bother the seller for things that are really important.
This would explain the reluctance of these sellers to engage through Amazon customer service (instead sending cards with their products that say "please report any issues to <email address>") — they don't manage their own product listings (rather, the contractor does); and they don't trust the English-language-product-listings contractor to know enough about their product to do customer service; and it's hard to coordinate their separate English-language customer-support contractor (the one you reach via the email on the card) to be able to receive + respond to messages on postings managed by the product-listings contractor.
It would also make an interesting prediction: that you'll only see the particular style I described in my top post, in Amazon's English-language product listings, because each contract company would likely focus on selling product-listing localization services for a particular language, so different language ⇒ different company ⇒ different style guide.
> e.g. New gossip of the day: "Amazon is going to ban any seller using the characters 【】"
You'd expect some people to miss that news. And even for people who see the update — no matter how "hungry" they are — you'd expect some sellers to let the news slip past them. For every such disseminated "Amazon Seller pro tip", you'd expect less than perfect 100% engagement. And yet engagement with these changes is 100% — at least within this network of sellers where engagement has historically been 100%.
(Just to beat this point to death, consider the ultimate in centralized top-down "do it exactly one way" skill dissemination: driving. Does every 16-year-old who is highly motivated and hungry to get out on the road, learn every rule of the road + constantly execute their learned driving skills perfectly? Sadly, no. Humans are not good at perfectly absorbing skills, and are also fallible at executing them.)
> a possible mechanism for this [...] They promise to "take care of" every interaction with Amazon FBA that requires English knowledge, and to only bother the seller for things that are really important.
Very likely.
In my previous comment, I focused on the perspective of individual sellers (i.e. entrepreneurial) who have bulk contracts with the local manufacturers. They are the group who are unlikely to commit to "all-inclusive" packages, as they value financial cost overwhelmingly more than personal time, relative to other groups. Think college students, retirees, unemployed, people from poorer areas, etc. They are ubiquitously known as 'wang dai', literally online retailer or agent. These are the first wave of significant online retailing in China. As they want to expand abroad, those who gained some first experience are packaging courses at minimal entry fee (~10USD!). There were a long period where these courses are bombarded via WeChat. (There still are, but the trend has shifted to other topics)
Naturally, the manufacturers want a bigger piece of the cake and now dominates their own online presence in China. To further expand beyond abroad, as they probably do not have adequate language/platform in-house expertise. This is a perfect match to the services you described - outsourcing to e.g. Amazon/eBay specialists.
In either case the point still stands. There's a snowball effect of following the lead of whoever is known to be successful and became the 'authority'.
> you'd expect less than perfect 100% engagement
I agree imperfect engagement is expected on the whole, but how the 100% metric is derived can be misleading. Is it 100% of the top 100 items i.e. the cream of the crop? How different would it be if the procrastinators or failures or one-offs are also included?
Also, I don't think analogue activities like driving is, analogous. Probably closer to Pride Day / French Flag photo overlay across various social platforms. Or SEO.
> a Borg Queen that could be taken down
I think we are aligned that there's a Borg Queen, but not on the rigidness of the hierarchy.
I believe it's weakly/organically structured, unless there's a monopoly, that I'm not aware of, happened.
Not so: the change happened across all these thousands of brands that had the exact style; but it didn't happen to the minority of posts that were from "real" independent Chinese marketplace-of-ideas sellers, who had copied the style with errors, or independently reinvented it. Those other listings still use the dictionary-headword brackets. So no Amazon-side canonicalization was performed.
This must be something that materializes out of the seller interaction with Amazon, because the same aesthetic does not appear on eBay, or for that matter Aliexpress.
You'd expect, if you allowed someone to study your notes before a test, that they'd possibly end up using similar turns-of-phrase to yours on the test. You wouldn't expect that they'd end up learning to perfectly duplicate your handwriting. It'd both be too hard, and not worth it, to do so; and they already have their own handwriting style. So why would they?