5 ms·

We've had open-source alternatives like http://identi.ca/ http://identi.ca/ for years; and in my memory, these have always been faster to implement new features

We've had open-source alternatives like http://identi.ca/ http://identi.ca/ for years; and in my memory, these have always been faster to implement new features that Twitter itself has been. Yet Twitter, not identi.ca, caught on in the public. So the question is really not whether there is an opening for a Twitter alternative and what features we would want it to have, but rather why the existing alternatives with strictly more features failed to catch on. My take is that Twitter had already passed a threshold by the time these alternatives hit the scene, and their first-mover advantage carried them a long way in defending themselves against other similar-but-incompatible platforms. Put another way, the reason Twitter is singular and email providers are not is that email providers communicate with each other over a common open protocol, whereas Twitter communicates only using its (closed) API. The difficulty would not be in defining an open protocol to doing what Twitter does, naturally; existing ones could be tweaked slightly to achieve the same result. (Look at what Shortmail is doing while maintaining 100% backwards-compatibility with traditional email). The problem is that Twitter has no incentive to turn its "open" (read: closed) API into an completely open means of communication like email, because doing so would squander the one huge asset it already has: a captive userbase. (For the record, the same arguments could apply just as well to Facebook; people just don't think about Facebook that way, because we're all so used to them being a walled garden).

smashing 14y ago
Commercial efforts tends to beat committee efforts in the marketplace.
- badgergravling 14y ago
  I'm not sure the two are mutually exclusive there. Commercial doesn't negate committees (I've been involved in a fair few), and open source doesn't have to be a committee effort. I blogged about the problem as I perceive it with existing efforts - they always touted themselves as 'the open source version' and most non-developers and non-technologists don't care about that. They care if it's easy to use and has reached critical mass amongst their networks.
- shuzchen 14y ago
  Identi.ca isn't exactly a committee effort though. The vast majority of development comes from Status.net, which makes money (apparently) by selling/setting up private instances for enterprise usage.
  - flatline3 14y ago
    Status.net should fund the development of OSS mobile/desktop clients that work with multiple Twitter-like services, including Twitter itself.
- erikpukinskis 14y ago
  You might be right, in the "more than 50% of the time" sense of "tends to", but this kind of generalization is not really prescriptive at all. I could easily rattle off dozens of cases where "committee efforts" beat commercial efforts (linux, webkit, TCP/IP, ARM, DNS, yada yada yada). On second thought, if you go by percentage of operations handled by, say, your cell phone, I wouldn't be surprised if more than 50% of those ops are "committee efforts".
chipsy 14y ago
Although it can look dark to have a monopoly in some space, like Twitter has achieved in microblogging, we actually have the same problem with older open protocols like email and IRC; they can be greatly improved with a redesign, but network effects are holding them back. This isn't really an issue of technical worth or underlying philosophy so much as it is finding the right "trojan horse" that can keep pushing new, worthy protocols into the mainstream. XMPP, for one example, stayed nearly unused for years and years and only gradually started making inroads as a public network after being picked up as an "add-on" to existing popular apps(e.g. GMail). Bittorrent, on the other hand, has been the defacto monopoly in mass-downloading of copyrighted material since it was released, and the closed/commercial response has been app stores, streaming, etc. Or, to put it another way, all of these technologies are ultimately "transitional." We just don't know where we're transitioning _to_ exactly. And we do seem to follow an "open/closed" "centralized/decentralized" wheel-of-reincarnation type of cycle in how we go about reinventing them.

smashing 14y ago

Commercial efforts tends to beat committee efforts in the marketplace.

badgergravling 14y ago
I'm not sure the two are mutually exclusive there. Commercial doesn't negate committees (I've been involved in a fair few), and open source doesn't have to be a committee effort. I blogged about the problem as I perceive it with existing efforts - they always touted themselves as 'the open source version' and most non-developers and non-technologists don't care about that. They care if it's easy to use and has reached critical mass amongst their networks.
shuzchen 14y ago
Identi.ca isn't exactly a committee effort though. The vast majority of development comes from Status.net, which makes money (apparently) by selling/setting up private instances for enterprise usage.
- flatline3 14y ago
  Status.net should fund the development of OSS mobile/desktop clients that work with multiple Twitter-like services, including Twitter itself.
erikpukinskis 14y ago
You might be right, in the "more than 50% of the time" sense of "tends to", but this kind of generalization is not really prescriptive at all. I could easily rattle off dozens of cases where "committee efforts" beat commercial efforts (linux, webkit, TCP/IP, ARM, DNS, yada yada yada). On second thought, if you go by percentage of operations handled by, say, your cell phone, I wouldn't be surprised if more than 50% of those ops are "committee efforts".

chipsy 14y ago

Although it can look dark to have a monopoly in some space, like Twitter has achieved in microblogging, we actually have the same problem with older open protocols like email and IRC; they can be greatly improved with a redesign, but network effects are holding them back. This isn't really an issue of technical worth or underlying philosophy so much as it is finding the right "trojan horse" that can keep pushing new, worthy protocols into the mainstream. XMPP, for one example, stayed nearly unused for years and years and only gradually started making inroads as a public network after being picked up as an "add-on" to existing popular apps(e.g. GMail). Bittorrent, on the other hand, has been the defacto monopoly in mass-downloading of copyrighted material since it was released, and the closed/commercial response has been app stores, streaming, etc. Or, to put it another way, all of these technologies are ultimately "transitional." We just don't know where we're transitioning _to_ exactly. And we do seem to follow an "open/closed" "centralized/decentralized" wheel-of-reincarnation type of cycle in how we go about reinventing them.