4 ms·

Except Twitter used to offer "extremely deep API support...allow[ing] folks to build applications that leverage [their] platform in unique ways." And then they

Except Twitter used to offer "extremely deep API support...allow[ing] folks to build applications that leverage [their] platform in unique ways." And then they couldn't make any money that way. I'm really not sure you have a good example of a loss of an open ecosystem here. NewsGator's plans were to pretty much create the equivalent of the (unofficial) Google Reader API, which, while not replacing RSS (the actual open format), does notably reduce its role. Moreover, for those who apparently don't remember, NewsGator actually did release their API. The API and their free reader software is what they shut down in 2009. You can blame that on Google, who did indeed retain the lion's share of users, or you could blame the self defeating business plan of providing a completely open API for free even while it costs money to run. A NewsGator API is not a loss in terms of a loss of an open ecosystem (and Bloglines was definitely not a loss). If you want open, you need a real federated protocol. Like RSS. Meanwhile, this article says "RSS industry" in the headline, but it's about nothing like that. The author says it himself: "No, Google Reader’s real competition back in its early days was not client software but services that aggregated RSS feeds and synchronized them across multiple devices." You can't embrace, extend, and extinguish services that have no standard format. RSS is doing just fine. What was lost here was a UI and a service, which leaves us with user migration pain and a loss of continuity of history (due to indefinite feed history). We'll likely always need some services. Yes, I can run a feed aggregator on my home machine, but it's kind of a waste, and not everyone has stable IP addresses they can use to sync across devices. Instead of giving up on services, what we need to learn from this is to make sure the next services we allow ourselves to depend on don't have control of our history, both individually and collectively. Notably Facebook, Twitter, and G+ don't fit this bill, or fit it only partially. (Unfortunately, our collective history on those services is complicated by visibility/access by person controls, so maybe there isn't a perfect solution. I know I can use takeout to download my posts on G+, for instance, but what about my comments on others' posts, and what about those posts themselves as context for my comments? And what about the major news of the day, even if I didn't comment? Sigh...we need a modern usenet...)

ahoyhere 14y ago
"And then [Twitter] couldn't make any money that way." Hold the phone. Did they actually try to make money that way?
- magicalist 14y ago
  I could change my post to say the didn't make any money that way, but the point is the same. The problem is access to a user's content being at the mercy of revenue. An API from a single vendor that has to be paid for is not a type of open ecosystem and 2005 was not a magic time of possibilities that have since died.
Evbn 14y ago
How can an unencrypted internet API be not open? We aren't talking about secret binary commands like Skype here.
- magicalist 14y ago
  It certainly can be, but it's not an open ecosystem. I hate when discussions dissolve into arguments over "open" just because the participants didn't have the same definition for the word "open" to start with, so I probably should have been more careful with using it. I was really just using it in direct reference to the OP's post, trying to say that "the relatively open ecosystem we had back in 2005" wasn't really that, at least in this context. If an API is the sole source of a service, and, even worse, if it's the sole source to access data that users put into a system (e.g. feed history for essentially all feeds, facebook posts, etc), you're totally dependent on that API making enough money to justify its existence based on the whims of the people who make that decision. Luckily there's google takeout this time, but some data is still being lost. One benefit of usenet was that everyone could have a copy of everything, though there's many privacy tradeoffs there (in the modern social networking sense). Anyway, it's not really a novel argument, just the same "dangers of monoculture/single-point-of-failure" point that has been made many times.

ahoyhere 14y ago

"And then [Twitter] couldn't make any money that way." Hold the phone. Did they actually try to make money that way?

magicalist 14y ago
I could change my post to say the didn't make any money that way, but the point is the same. The problem is access to a user's content being at the mercy of revenue. An API from a single vendor that has to be paid for is not a type of open ecosystem and 2005 was not a magic time of possibilities that have since died.

Evbn 14y ago

How can an unencrypted internet API be not open? We aren't talking about secret binary commands like Skype here.

magicalist 14y ago
It certainly can be, but it's not an open ecosystem. I hate when discussions dissolve into arguments over "open" just because the participants didn't have the same definition for the word "open" to start with, so I probably should have been more careful with using it. I was really just using it in direct reference to the OP's post, trying to say that "the relatively open ecosystem we had back in 2005" wasn't really that, at least in this context. If an API is the sole source of a service, and, even worse, if it's the sole source to access data that users put into a system (e.g. feed history for essentially all feeds, facebook posts, etc), you're totally dependent on that API making enough money to justify its existence based on the whims of the people who make that decision. Luckily there's google takeout this time, but some data is still being lost. One benefit of usenet was that everyone could have a copy of everything, though there's many privacy tradeoffs there (in the modern social networking sense). Anyway, it's not really a novel argument, just the same "dangers of monoculture/single-point-of-failure" point that has been made many times.