4 ms·
> Remember Digg? Unlike the companies you listed, FB is user-facing - just like Digg. Digg was abandoned because it broke itself. The codebase that all those
by cookiecaper 9y ago
> Remember Digg? Unlike the companies you listed, FB is user-facing - just like Digg.
Digg was abandoned because it broke itself. The codebase that all those users knew and depended on was completely trashed (the original author hangs out here sometimes), the site dropped a number of its key functions, and the user experience was redesigned to maximize Digg's ability to show ads, no matter how obnoxious that got, with many people saying that it was almost entirely ads by this point (I had personally stopped using Digg years before the big v4 debacle, so I can't speak to this directly).
People want the path of least resistance. They'll tolerate some ads, but their tolerance ends when it's just easier to click over to the other guy's site and get a comparable experience. If you neuter/break Digg, and reddit is right there, well...
The path of least resistance for Facebook users is still Facebook, regardless of the news cycle. Facebook will remain dominant until it is no longer the path of least resistance, just like everything else.
Users have different levels of tolerance. The moment a switch is the path of least resistance differs for everyone, usually corresponding to their level of emotional investment in the platform and their individual tech savvy. Facebook has a lot of very invested, very non-tech-savvy users.
The right types of questions to ask are along the lines of whether we're sure competitors have sufficient ground to catch up to massive incumbents and keep them honest (and in this case specifically, whether certain misrepresentations or violations occurred that should be made to materially affect market standing through government intervention). For the most part, tech is non-competitive due to terrible legislation like the CFAA.
- narag 9y agoIf you neuter/break Digg, and reddit is right there, well... That case is about an almost identical twin, but I would also consider sites that are not so similar but serving the same function. I haven't had a Facebook account so not sure what it is exactly, but people are happy as long as I have Whatsapp.
- waylandsmithers 9y ago> The right types of questions to ask are along the lines of whether we're sure competitors have sufficient ground to catch up to massive incumbents and keep them honest (and in this case specifically, whether certain misrepresentations or violations occurred that should be made to materially affect market standing through government intervention). I don't see how there could ever be any viable competition. Facebook already has many of the top engineering minds in the world who are pushing the edge of what computers are capable of, not to mention the largest database and neural network of human behavior ever created.
- jstarfish 9y ago> I don't see how there could ever be any viable competition. Facebook already has many of the top engineering minds in the world who are pushing the edge of what computers are capable of, not to mention the largest database and neural network of human behavior ever created. How is any of that necessary? I don't give a shit about their bleeding-edge neural network or army of PhDs. I just want to send my mom pictures of my kids and maybe send an occasional party invitation to some friends. What does Skynet contribute to this workflow? I would argue that all that technical bloat only made the Facebook experience worse. Facebook itself had humble beginnings. All a competitor needs is a LAMP or MEAN stack, an experience worth migrating to, and enough ethical sense to not treat its userbase like a science experiment.
- cookiecaper 9y ago>I would argue that all that technical bloat only made the Facebook experience worse. Facebook itself had humble beginnings. All a competitor needs is a LAMP or MEAN stack, an experience worth migrating to, and enough ethical sense to not treat its userbase like a science experiment. It also needs the ability to make the transition plausible for consumers. There's really no reason that Facebook should be able to enforce a stringent walled garden and bar specific user agents that aren't negatively impacting the network and aren't exceeding the access of the user they represent. But Facebook did exactly this in Facebook v. Power Ventures, leaving an entrepreneur with a ruined business and a bill for $3M in damages for creating a system that exfiltrated only one's own personal, copyrighted data from Facebook. It wasn't even taking other peoples' feeds or trying to multiplex data. It was simply trying to make one's own data accessible and portable. Large tech companies have done this for decades, abusing the CFAA and the Copyright Act to enforce de facto platform monopolies. These companies know that if they can't assert ownership over the data, they'll have to compete on merit in the open marketplace, and who wants to deal with that? This is also why you don't see new protocols like POP3 or NNTP anymore. Academics trying to build a transparently-syndicated collaboration platform are out. Big companies looking to hoard your data and assert some type of incidental, indirect ownership over it are in, and they want to make sure that you don't really have the option to leave. Lock-in is king. Once a critical mass outside of the tech-savvy early adopter crowd is obtained, aggressive legal methods like those used against Power allow incumbents to hold onto power in perpetuity. Granny doesn't care about all of that tech mumbo-jumbo. She just knows that she can't see Auntie's pictures anymore because you broke it by switching her to F+. Since they're legally forbidden from multiplexing Facebook's content into a universal "social media reader", Granny's options are to stare at a blank feed or go back to Facebook -- even though Auntie would be perfectly willing to allow Granny to see her content through F+, and it's really more of an accident that Facebook is the platform that received the initial post. What is Granny gonna do in this situation? Why have we awarded tech companies the right to hold user-generated content hostage like this? Wrap anything in custom HTML and it's suddenly very difficult for a third-party to access it without infringing your copyright on the exterior HTML wrapping, regardless of the status of the content inside. In the US, it's infringement to load copyrighted content into RAM, per the RAM Copy Doctrine. Good luck convincing a technologically-illiterate judge that your software is no more disruptive than Chrome or Firefox, especially when you have to do so over the objections of Facebook's $X000/hr law firms and previously-referenced army of PhDs. Such rules have stifled online innovation for decades now, and they're the reason that the internet is becoming a walled garden. Under the GDPR, which mandates all companies provide a machine-readable export of user data within some fairly short timeframe (less than a week iirc), things may become marginally more fluid for people in EU countries, but I wouldn't count on it. Disclaimer: not a lawyer