What makes $BigTechCompany's products bad, anyway?
24 points by mond
24 points by mond
If I remember correctly, google got caught artificially decreasing the quality of search results to force users to make multiple searches per query. Downright unethical & exploitive practices like that make products far worse for me and the "median user.' For me, that's a way larger problem than "oh no the product doesn't cater to my niche usecase."
Yes. The hard parts here are 1. lack of visibility into this from outside the company, 2. it's hard to structurally prevent changes like that.
In Google's defense, Google claims to have some internal separation between its Search and Ads divisions, presumably to provide some structural defense against ads incentives hurting search quality.
It's hard to tell from the outside how true that is (or how much it changed over the years), so I'd take it with a grain of salt, but at least the principle of such a structural separation is great, with the obvious caveat that it's hard to maintain under pressure.
In Google's defense, Google claims to have some internal separation between its Search and Ads divisions, presumably to provide some structural defense against ads incentives hurting search quality.
Unfortunately, this doesn't matter at all if Search has an org-level OKR like "Increase Search volume by xx%". Which they almost certainly do. The result is indistinguishable from Ads telling Search to do anything they can to get more searches.
On principle I agree, but honestly, I'm worried about this pattern on addictive platforms (e.g. Youtube, Instagram). For feedscrollers an increasing volume strictly maps to "make the platform more addictive (and show more ads)".
Google Search is just not addictive or scrollable, so I'm not as worried about it.
Yes, Doctorow was telling that story about the court case revealing that on every pod he was on in his last book tour if you're looking for a source.
Imagine there's a knob sitting there and a product manager or an executive notices that your KPIs improve the moment you start turning that knob. Do you think they could ever resist turning that knob regardless of what the knob actually does.
Or maybe there was an anomaly in the KPIs yesterday, they were abnormally high. Turns out that was a side effect of a bug. Imagine how enthusiastic they will be to get the bug fixed.
What makes $BigTechCompany's business practices excusable, anyway?
You joke, but I think that this is an interesting question! It's easy to say "Microsoft bad", and it turns out that Microsoft is (in fact) bad, but it's a bit harder to pin down which precise things are bad in a way that could be generalized to laws or regulations (that have any remote chance of passing).
e.g. The switch from a license-based business model to a cloud subscription based one. It's not obvious that subscriptions are worse, you have to argue that, for example, cloud subscriptions increase vendor lock-in, concentrate power, or are problematic due to the ability of the provider to increase prices on a short notice.
In the same vein as "Seeing like a bank" by Patrick Mackenzie and "Seeing like a State" by James Scott, I think it's high time for someone to write a "Seeing like a $BigTechCompany" that highlights how to the long chains of accountability sinks build up decisions processes that wind up into enshittified outcomes.
My understanding is that no single individual ever decided that Google Search had to be bad, but that they A/B tested themselves into the result that made their line go up in detriment of everything else.
You may already be familiar with this, but Sean Goedecke took a stab at this prompt: https://www.seangoedecke.com/seeing-like-a-software-company/
This is the rough direction I'm trying to move in here, but it's a pretty complex topic. I've read both of these.
FYI, if you've not read https://apenwarr.ca/log/20190926-what-do-executives-do-anyway, I recommend it on this topic.
"Careless People" by Sarah Wynn-Williams fits in here, I think. It's not an exact match, but it does talk a bit about a growth-at-any-cost mindset at Facebook.
Google is, quite literally, evil (remember the motto change?).
Google will happily ruin the internet for users like us for a profit, then later ban their 15 year old accounts for being a "bot" for no reason (and hecking stop asking questions ok?) — happened to me.
The core issue here is a misalignment of expectations. OP is trying to justify what is happening in economic terms, while I do not agree with the things happening in the economy today being acceptable at all.
So yeah, Google radicalized me. And I'm not having any of this capitalist corpo bootlicking.
Describing behavior in economic terms is not justifying it anymore than providing an analysis of a piece of malware is encouraging spreading it.
The core issue here is a misalignment of expectations. OP is trying to justify what is happening in economic terms, while I do not agree with the things happening in the economy today being acceptable at all.
Perhaps I missed it, but I don't remember the article trying to argue that what these companies do is necessarily good. It's just pointing out why they do what they do.
I agree that people shouldn't really get to complain about Google making their products worse for their specific use case, especially when they most often aren't paying them specifically for it. But I feel like that's a bit of a strawman.
If we accept your point that we should still call out Google for being unethical or enshittifying, then what remains is essentially a call to not be like the spacebar xkcd. For the record, I agree that it's important to support median-user features like fuzzy search finding Nietzsche (I cannot spell that name) from "neechee" --- and for Google that it is both explicable and reasonable to prioritize them over power-user features. But if we "follow the money," as you suggest, then wouldn't it make sense to keep the median user happy enough to turn up the dial on ad revenue, which is exactly the premise of enshittification?
This ignores as well the (hairy, but I believe legitimate) ethical issues people have raised about big tech. Which is fair, it was clear pretty quickly that by "bad" you meant ineffective, but I clicked expecting a slightly different article.
I think this is upside down. "Why is bigco product bad" is a complex and nuanced question. "Why am I using bigco's product at all" is way simpler.
We only all need to deal with things that are mid for everyone if there's only 2 things and we can't meaningfully make another one. This whole discussion takes for granted that bigco exists in the first place and that bigco has positioned themselves such that competing with them takes complexity-they-introduced-based economies of scale and lawfare war warchests that only they have. "We have to get used to things that suck cause making things for everyone is hard." Yeah. Stop trying. "The median user" is a concept that only exists if your users are effectively imprisoned. If they weren't, the ones who don't like the product would use a competitor's.
There's no good reason that my grandma and my mom and me and my nieces all need to use the same WhatsApp client. If maps thinks my address is 4 blocks south of where it is, why can't I make a "go to my house" app without mapping the entire planet? The data's there. It'd have about 12 customers and they'd all love it.
As long as bigco refuses to interoperate meaningfully, this is entirely their fault and they deserve the flak for it.
None of this matters when it's not centralized or when the frontends and protocols and network operators aren't welded together by legacy economics.
There's no good reason that my grandma and my mom and me and my nieces all need to use the same WhatsApp client
I suppose the reason is findability. How do they hear about Whatsapp existing? From someone already using it, presumably. This starts at the early adopters, who are probably technically savvy. They say "well, this is what I use" and recommend the same to their friends. Or maybe they go with something more palatable to the "non-technical user". But as soon as you have non-technical users talking to each other they are not even going to know (or care about) the ins and outs of different clients, so you get a popular client to win out. It's a type of network effect of popularity rather than technical necessity.
The word "monopoly" doesn’t appear at all in that piece and that’s probably all you need to know given that it is written by a Google employee.
The very first problem of "Big Tech" is that you can’t escape them or at a great cost. They are controlling most governments.
Perhaps. It is also true that people choices are perhaps the biggest motivator. Take Google Search: it's been the most unusable unfriendly ads-infested search engine for many years, and for many years there have been good alternatives (I'm very happy with DuckDuckGo for example), and despite all of that, most of the users still choose Search because $reasons. Why should Google do anything differently at all?
People keep coming back for more heroin so why should dealers do anything differently at all?
As they do for any other drugs, illigal or not (that's a technicality by the way, heroin should be regulated like every other potentially dangerous substance, making it illigal just makes crimial organizations profit from it, with all the consequences).
Human nature won't change, corporations like Alphabet and such only care about profit and shareholders happyness, unless we come up with a new model to replace capitalism, that won't change for sure.
What we can change is our habits, I do change mines as I go, and avoid bad applications/services as soon as there's a better option.
As a company grows, “scale” becomes more important than making everyone happy, and non-median users are essentially invisible in the metrics.
In other words, the bigger the company, the shittier the product. They can't make any products which anyone loves because they can't be focused. I don't really understand why these are so popular though. I understand that network effects may lead someone to use a product for compatibility reasons, or for example inside large organizations with a diverse user base, but why would one willingly choose (or keep using) a shitty product for themselves?