The sooner the better.
I long for the headline “self driving Tesla runs over musk”.
Self driving teslas only run over kids and POC
Point of contacts?
The fact it hasn’t imploded a long time ago is proof that digital platforms need to be regulated to enforce interoperability.
Since this shitshow started, I have not heard from anyone that wanted to be on Twitter. In anything resembling a free market, these customers (both advertisers and users) could freely go to a competitor.
But due to the way platforms work, no one can compete, once a dominant platform emerges. A platform has a monopoly on all the things people built on top of the platform (content, software etc.). This monopoly kills the free market. Enforced interoperability would reduce this platform effect and help out competitors.
The EU is starting to tackle that, with the Digital Markets Act, but very few companies are targeted so far, even though the whole industry is plagued by quasi-monopolistic platforms that are universally agreed upon to be trash.
That’s a seriously interesting idea. For context, I’m a middle-aged, Southern, American white guy. “FREE speech! CAPITALISM!”
“That how dad did it, that’s how I do it, and it’s worked out pretty well so far.” ~Tony Stark.
High time to start looking at ideas like yours. If Europe and California have to impose these things? So fucking be it.
Might make me uncomfortable, might not understand it completely, too bad for me. I will vote for the world I want my children to live in. They’re 8 and 10, I’m 52. Done my time, coasting out. Y’all’s turn.
And if you want to hold forth on the notion of “enforced interoperability”, I’m listening.
Sure, yeah. The way I imagine this would work out best for humanity, is if companies are forced to open up platforms they provide, when they have e.g. more than 40% market saturation with that.
Most small platforms will want to strive for interoperability with the dominant platforms anyways, so this threshold is just to keep the burden of regulation low.
In practice, this might mean that Twitter would be forced to allow federation with Mastodon.
Or that Microsoft is forced to open-source the code for the Windows API.
Or that Reddit is blocked from closing up their third-party API.Ultimately, I don’t think, it even needs to be as concrete. I feel like even a law stating that if you’re providing a platform, you need to take special care to keep competition alive (along with some detailing what this entails), and then leaving it up to a judge to decide, would work.
The GDPR is implemented like that and while most larger companies are IMHO in violation of the GDPR, I also feel like most larger companies actually did go from atrocious privacy handling to merely bad privacy handling, which is an incredible success.
That’s effectively all I’m hoping for, too. That dominant platforms can’t just stagnate for multiple decades anymore. That they do have to put in at least a small bit more effort to stay in that dominant position.
I liked Twitter. I know it’s a cesspit, but as a software engineer it was always the top company I wanted to work at. It didn’t work out (for several funny reasons), but for that selfish reason I’ll never forgive Musk.
IMO, Musk needs help. If he were a normal person, someone would have pushed him to leave work and find help. As the owner of three companies, responsible for tens of thousands of employees, no chance is he getting that help. He’s constantly baited and prodded by his fan boys, people like Rogan and Chappelle who can deal with that kind of fame, and the press that get content from his antics.
As for Twitter, I don’t see it dying, until it fails to have a use for Musk. My initial belief was that his “everything app” would use Twitter’s account system to get all of its users, and then he’d sell Twitter and continue with the users - but that app isn’t ever happening. It’s just something he’s desperate to ditch, but his vanity and poor mental health won’t let him do it. For that reason, it’ll just be a zombie app.
I dont think an “everything app” will ever work.
You can make one thing that does one thing very well and better than the competition, and you will get users. Or you can do one thing that will try to do 10 things half assed, and it will fail to impress users. This happens because you have to divert your resources (time, money, people) for development, maintenance, new ideas, design etc. across all your “everythings”. The more everythings you have, the less resources each one gets, however the costs for maintenance, bugfixes, updates etc. stay the same.
This happened to Yahoo in the early 2000s, where it tried to be Search, News portal, Email, Web directory, Weather, games and whathaveyou, however it failed because none of it’s parts was better than the competition.
The better approach for an app would be to do it’s own thing it is supposed to do, but support other apps that can enhance your product by allowing it to interact with outside data, and also give his data back out to other apps: use mailto:links/email instead of inventing your own messaging protocoll, support exporting to standard calendar files instead of implementing your own calendar that is oblivious to the schedule on the users phone. Support exporting datasets into common formats the user knows from his everyday tasks (excel, csv) so he can run his own data analysis on it, instead of baking some half-assed “analytics” module that only has 10% of the features the user needs.
Of course it works. See WeChat in China.
But I doubt it’ll be X that will make it work in the rest of the world.
That is a good example, but as the other commenter pointed out I dont think you can compare weChat with Twitter. Twitter is a startup trying to make money from it’s service. WeChat is a tool for the chinese goverment to track each persons chats, money transactions and purchases, and as such will pretty much receive all the funding it needs. Being profitable is not the main objective of WeChat.
The Chinese government can track any app in their country. Their laws just give them acces to the data if they want it (the same as the usa basically). They don’t give a shit about WeChat.
Sure, but withg wechat you can link each user to the real person, bank account, phone number and find his friend circle on we chat. This might not work so well on other apps where any user can sign up via vpn and a random email address…
Yes