4 ms·

I'm sure Datasift saw it coming. I researched this last summer. Even before I was explicitly told all the original firehose contracts were five years long, I c

by frabcus 12y ago

I'm sure Datasift saw it coming. I researched this last summer. Even before I was explicitly told all the original firehose contracts were five years long, I could infer it from attitudes and behaviours. I'm excited that Datasift are launching interesting new Facebook features. They're a great company, with good technology. I hope Twitter's "no data resale" attitude leads to them becoming irrelevant, as it is so destructive of much innovation.

bane 12y ago
Yeah, I'm sure it was significant source of talk within the company for a while. But what can you do? It's not like there are other Twitters out there. The value of Twitter is that there's only one. It looks and sounds like they're trying to diversify out to other sources, but none-of-them are Twitter and for whatever reason Tweets are like crack to market research firms.
- frabcus 12y ago
  Well, Instagram and Twitter are converging in terms of feature set. They're pretty well indistinguishable now. They've not totally converged in terms of audience, but I think have surprising overlaps. Lots of Twitter data use cases you can get from Instagram, I bet.
  - MichaelGG 12y ago
    And Facebook/Instagram has no official data policy, right? They don't offer paid data from what I can see. You can sign up as a dev, email them if you need more access, otherwise...? So as far as data access goes, Twitter and FB/Instagram are pretty much on the same page. Except Twitter's docs are way more accurate and nicer, probably because it makes money for them and isn't a "throw it out there and see what happens" kinda thing.
  - untog 12y ago
    Well, Instagram and Twitter are converging in terms of feature set. They're pretty well indistinguishable now. I don't see how that's true at all. While Twitter now has well-developed image capabilities it's still not the core of what people use it for. Also Instagram has no equivalent of retweets.
MichaelGG 12y ago
>I hope Twitter's "no data resale" attitude leads to them becoming irrelevant, as it is so destructive of much innovation. Are you serious? As in, do you honestly believe that end users are gonna say "what, Twitter won't let people have access to 'our data'? Screw that I'm gonna go to ... <some competitor I've not heard of>?"
- wpietri 12y ago
  I think the actual answer is implied by the last clause of the sentence you quote. Twitter gained a lot early on from having all sorts of innovative things happen on top of their APIs. They clearly believe that's irrelevant now, as they have energetically screwed over all sorts of API users. I think the hope is that users will gradually say, "Oh, I love cool tool X for competitor Y; I'll switch." I don't think it will happen next week, but it's a plausible path to long-term failure. Missing important waves of innovation is how a lot of tech companies die.
  - MichaelGG 12y ago
    Can you list some X and Ys that exist or have existed? I don't use social media at all (except when people link to G+ or Twitter from HN). But looking around, I don't see anyone offering the same kind of service as Twitter. And even last month, didn't they just offer video streaming, right around the time Meerkat got to market? So it's unsure that they need 3rd parties. Twitters got how many thousands of employees? Surely if they organize well they can innovate inside. Or just keep an ear to the ground and hear about neat-sounding things and clone them. Or buy them if they have to. What value add did DataSift bring? They were taking 80% of the revenue. If those DataSift customers were really using the Twitter firehose, there's no real replacement. That means Twitter just upped revenues on those customers by 400%.
    - wpietri 12y ago
      You mean X and Y for Twitter in the future? It's impossible for me to say; if I could, it wouldn't be particularly innovative. Historically, though, platforms die all the time. AOL and Compuserve lost out because they couldn't innovate at anything like the speed of the open web. I don't think they need third parties immediately. They needed them in the past, though; they were important to Twitter's early success. They'll need them again at some point, too. The notion that companies can innovate because a) they want to innovate, and b) they have lots of people is... unproven. How many years has Yahoo been floundering now? What's the last innovative thing Microsoft did? Does IBM even exist anymore? Lots of large companies open innovation centers and intrapreneurship programs. Small companies still regularly kill large ones. That's because large companies are mainly optimized for turning the crank on their current money machine. Things that interfere with that rarely survive, including most innovation efforts.

by frabcus 12y ago

bane 12y ago

Yeah, I'm sure it was significant source of talk within the company for a while. But what can you do? It's not like there are other Twitters out there. The value of Twitter is that there's only one. It looks and sounds like they're trying to diversify out to other sources, but none-of-them are Twitter and for whatever reason Tweets are like crack to market research firms.

frabcus 12y ago
Well, Instagram and Twitter are converging in terms of feature set. They're pretty well indistinguishable now. They've not totally converged in terms of audience, but I think have surprising overlaps. Lots of Twitter data use cases you can get from Instagram, I bet.
- MichaelGG 12y ago
  And Facebook/Instagram has no official data policy, right? They don't offer paid data from what I can see. You can sign up as a dev, email them if you need more access, otherwise...? So as far as data access goes, Twitter and FB/Instagram are pretty much on the same page. Except Twitter's docs are way more accurate and nicer, probably because it makes money for them and isn't a "throw it out there and see what happens" kinda thing.
- untog 12y ago
  Well, Instagram and Twitter are converging in terms of feature set. They're pretty well indistinguishable now. I don't see how that's true at all. While Twitter now has well-developed image capabilities it's still not the core of what people use it for. Also Instagram has no equivalent of retweets.

MichaelGG 12y ago

>I hope Twitter's "no data resale" attitude leads to them becoming irrelevant, as it is so destructive of much innovation. Are you serious? As in, do you honestly believe that end users are gonna say "what, Twitter won't let people have access to 'our data'? Screw that I'm gonna go to ... <some competitor I've not heard of>?"

wpietri 12y ago
I think the actual answer is implied by the last clause of the sentence you quote. Twitter gained a lot early on from having all sorts of innovative things happen on top of their APIs. They clearly believe that's irrelevant now, as they have energetically screwed over all sorts of API users. I think the hope is that users will gradually say, "Oh, I love cool tool X for competitor Y; I'll switch." I don't think it will happen next week, but it's a plausible path to long-term failure. Missing important waves of innovation is how a lot of tech companies die.
- MichaelGG 12y ago
  Can you list some X and Ys that exist or have existed? I don't use social media at all (except when people link to G+ or Twitter from HN). But looking around, I don't see anyone offering the same kind of service as Twitter. And even last month, didn't they just offer video streaming, right around the time Meerkat got to market? So it's unsure that they need 3rd parties. Twitters got how many thousands of employees? Surely if they organize well they can innovate inside. Or just keep an ear to the ground and hear about neat-sounding things and clone them. Or buy them if they have to. What value add did DataSift bring? They were taking 80% of the revenue. If those DataSift customers were really using the Twitter firehose, there's no real replacement. That means Twitter just upped revenues on those customers by 400%.
  - wpietri 12y ago
    You mean X and Y for Twitter in the future? It's impossible for me to say; if I could, it wouldn't be particularly innovative. Historically, though, platforms die all the time. AOL and Compuserve lost out because they couldn't innovate at anything like the speed of the open web. I don't think they need third parties immediately. They needed them in the past, though; they were important to Twitter's early success. They'll need them again at some point, too. The notion that companies can innovate because a) they want to innovate, and b) they have lots of people is... unproven. How many years has Yahoo been floundering now? What's the last innovative thing Microsoft did? Does IBM even exist anymore? Lots of large companies open innovation centers and intrapreneurship programs. Small companies still regularly kill large ones. That's because large companies are mainly optimized for turning the crank on their current money machine. Things that interfere with that rarely survive, including most innovation efforts.