We need a better web, not an open Twitter(blog.superfeedr.com)
blog.superfeedr.com
We need a better web, not an open Twitter
http://blog.superfeedr.com/better-web/
7 comments
The problem isn't siloing, that's a symptom. The problem is hosting. No matter which way we spin it, major services still need to pay for the physical hardware on which it runs. Given the size of Twitter and Facebook, this means thousands of machines that need power, space, cooling, and maintenance. None of that stuff is free. This, of course, invites business thinking and money, both of which want to further silo it's systems.
The other major problem is the client-server nature of the web itself. I, as a user, must request information/page/JSON from somewhere where it's processed by the thin terminal of the browser. Backbone.js and such technologies are trying to beef up the browser, but until the entire paradigm changes we are only applying bandages to the problem. That invites scaling problems to solve and the possibilities of a single outage bringing down entire services. We all remember the AWS outage a month ago, but problems arose in the platform before. Anybody using Tumblr last year will remember the tongue-in-cheek Tumblbeasts when it would go down several times a day.
I really do believe P2P technologies are the paradigm that will deliver the openness that the web promises, but it's also very hard to build. Concurrent multi-threaded programming hard. Security and veracity are a nightmare. Just analyzing how to propagate information is difficult. But it brings with it a built-in solution to scaling and reliability. Unlike central services, a P2P service experience gets better when more people are involved because it makes the network effect an actual strength. Bittorrent is the classic case. I believe mobile phones, with their native applications and built in network hardware, could be leveraged into an ad-hoc P2P/mesh network.
I talk about some of this here: http://blog.jamesleonis.com/says/the-end-of-an-age/
The other major problem is the client-server nature of the web itself. I, as a user, must request information/page/JSON from somewhere where it's processed by the thin terminal of the browser. Backbone.js and such technologies are trying to beef up the browser, but until the entire paradigm changes we are only applying bandages to the problem. That invites scaling problems to solve and the possibilities of a single outage bringing down entire services. We all remember the AWS outage a month ago, but problems arose in the platform before. Anybody using Tumblr last year will remember the tongue-in-cheek Tumblbeasts when it would go down several times a day.
I really do believe P2P technologies are the paradigm that will deliver the openness that the web promises, but it's also very hard to build. Concurrent multi-threaded programming hard. Security and veracity are a nightmare. Just analyzing how to propagate information is difficult. But it brings with it a built-in solution to scaling and reliability. Unlike central services, a P2P service experience gets better when more people are involved because it makes the network effect an actual strength. Bittorrent is the classic case. I believe mobile phones, with their native applications and built in network hardware, could be leveraged into an ad-hoc P2P/mesh network.
I talk about some of this here: http://blog.jamesleonis.com/says/the-end-of-an-age/
"lack of knowledge about how things work" and "The problem is hosting". Are you people insane?!
An open information policy, regardless of it's supposed higher morality, reduces a social network's ability to compete, because a) it gives competitors an advantage in acquiring your users and retaining them, and b) the vast majority of users do not care about this. We have the current equilibrium in the social networking market because of many forces, including ease-of-use, social feature set, up-front monetary cost, and social group inertia, but information policy has empirically proven not to be one of these factors. It has, however, been an important factor in shaping the self-publishing/blog market. There are different solutions suited to different markets.
And you cannot explain people's indifference to this issue by proclaiming (arrogantly) that they have a "lack of knowledge". People have other more pressing concerns in their lives. Making the rent this month. Dealing with a child who is not performing well in school. Internalizing stress from their job because they are not trained enough to get a better one. Even if the average person did know about information accessibility, privacy, and sale to 3rd parties, why would this issue rise to a level comparable the issues listed above? The fact that they can smile for half an hour when they get home from work by reading/looking at a personalized listing of information they find valuable, and do this for free, is all they need to know about the web. This is not for better or for worse. It's reality in all its complexity.
If or when this free-but-centralized/closed system fails them, by directly affecting their pocketbook or by itself has becoming insolvent, is when they'll take notice. And if this does ever happen, it is probably decades away.
Lastly, if you aim your service at just "the people that care about information access", don't overestimate your market size. You going after a much smaller pool. With social/communication services, we're not just talking about a pool of individuals, we're talking about a pool of friend/colleague/interest groups. In this case, the specific pool is basically all friend/colleague/interest groups which are comprised entirely of libertarian-minded nerds that actually want a communication service beyond email/mailing lists, github, and blogs/comment threads.
Ferchrissake, this comment thread on Hacker News only generated 19 comments; what do we need a whole social networking service for?!
An open information policy, regardless of it's supposed higher morality, reduces a social network's ability to compete, because a) it gives competitors an advantage in acquiring your users and retaining them, and b) the vast majority of users do not care about this. We have the current equilibrium in the social networking market because of many forces, including ease-of-use, social feature set, up-front monetary cost, and social group inertia, but information policy has empirically proven not to be one of these factors. It has, however, been an important factor in shaping the self-publishing/blog market. There are different solutions suited to different markets.
And you cannot explain people's indifference to this issue by proclaiming (arrogantly) that they have a "lack of knowledge". People have other more pressing concerns in their lives. Making the rent this month. Dealing with a child who is not performing well in school. Internalizing stress from their job because they are not trained enough to get a better one. Even if the average person did know about information accessibility, privacy, and sale to 3rd parties, why would this issue rise to a level comparable the issues listed above? The fact that they can smile for half an hour when they get home from work by reading/looking at a personalized listing of information they find valuable, and do this for free, is all they need to know about the web. This is not for better or for worse. It's reality in all its complexity.
If or when this free-but-centralized/closed system fails them, by directly affecting their pocketbook or by itself has becoming insolvent, is when they'll take notice. And if this does ever happen, it is probably decades away.
Lastly, if you aim your service at just "the people that care about information access", don't overestimate your market size. You going after a much smaller pool. With social/communication services, we're not just talking about a pool of individuals, we're talking about a pool of friend/colleague/interest groups. In this case, the specific pool is basically all friend/colleague/interest groups which are comprised entirely of libertarian-minded nerds that actually want a communication service beyond email/mailing lists, github, and blogs/comment threads.
Ferchrissake, this comment thread on Hacker News only generated 19 comments; what do we need a whole social networking service for?!
I generally agree, our web is increasingly made of huge corporations, silos and less and less made of small indie websites which would guarantee that no one is ever powerful enough to decide for everybody else.
Yet, I'm not as pessimistic as you seem to be. We can make things better: we just have to do it.
We should definitely try, but centralization does give some dangerous benefits.
It would be difficult to have trending topics in a completely decentralized twitter (everybody would have to report to some central "trending topic collector" service, right?). You could still notify someone of @ mentions, but then @foo turns into an email address (or something with lookupable routing information). Not very pretty.
With tumblr, half the benefit of the service is seeing who posted the same thing somewhere else. Unless you are Google, we can't quite cross reference the entire web in real time to see who posts the same thing or who forked it from whom.
Centralization gives us many nice features we can't get from distributed systems, but it also gives us single points of failure, arbitrary internet dictatorships, and fad driven startup cargo cult fetishism.
It would be difficult to have trending topics in a completely decentralized twitter (everybody would have to report to some central "trending topic collector" service, right?). You could still notify someone of @ mentions, but then @foo turns into an email address (or something with lookupable routing information). Not very pretty.
With tumblr, half the benefit of the service is seeing who posted the same thing somewhere else. Unless you are Google, we can't quite cross reference the entire web in real time to see who posts the same thing or who forked it from whom.
Centralization gives us many nice features we can't get from distributed systems, but it also gives us single points of failure, arbitrary internet dictatorships, and fad driven startup cargo cult fetishism.
Trending topics are this decade's flagpole sitting. Replace it with something just as dumb-joinable-fun that's only possible with a decentralized web, and people will forget about the old thing after a while. Easier said than done, of course, but not an inherent advantage to centralization.
And Google used to be impossible too. There's always a chance that someone reading this site will do to Google exactly what Google did to AltaVista, and for the exact reasons you cite here.
And Google used to be impossible too. There's always a chance that someone reading this site will do to Google exactly what Google did to AltaVista, and for the exact reasons you cite here.
Trending topics are this decade's flagpole sitting
I don't think that's true. Identifying trending topics can be hugely important. I forget where I saw it, but someone did an analysis of tweets about "earthquake", when one occurred on the east coast of the US some months ago. It followed the path of the quake almost exactly. When people are posting about events before the news does, it's great to be able to detect it.
I don't think that's true. Identifying trending topics can be hugely important. I forget where I saw it, but someone did an analysis of tweets about "earthquake", when one occurred on the east coast of the US some months ago. It followed the path of the quake almost exactly. When people are posting about events before the news does, it's great to be able to detect it.
It would be difficult to have trending topics in a completely decentralized twitter (everybody would have to report to some central "trending topic collector" service, right?).
A decentralized Twitter would probably be like StatusNet: Server-to-server, not client-to-client. It wouldn't be that hard to subscribe (using PSHB) to each server's trending API and get callbacks from all of them.
So yes, there would be trending topic collecting services, but multiple, not a central one.
You could still notify someone of @ mentions, but then @foo turns into an email address (or something with lookupable routing information).
You can hide the routing information from the user, as message metadata.
With tumblr, half the benefit of the service is seeing who posted the same thing somewhere else.
It's not really that hard: when you re-publish a post from another server, yours can just subscribe to get new events about that post (like new reposts).
A decentralized Twitter would probably be like StatusNet: Server-to-server, not client-to-client. It wouldn't be that hard to subscribe (using PSHB) to each server's trending API and get callbacks from all of them.
So yes, there would be trending topic collecting services, but multiple, not a central one.
You could still notify someone of @ mentions, but then @foo turns into an email address (or something with lookupable routing information).
You can hide the routing information from the user, as message metadata.
With tumblr, half the benefit of the service is seeing who posted the same thing somewhere else.
It's not really that hard: when you re-publish a post from another server, yours can just subscribe to get new events about that post (like new reposts).
> It's not really that hard: when you re-publish a post from another server, yours can just subscribe to get new events about that post (like new reposts).
Because trackback/pingbacks worked so well.
Because trackback/pingbacks worked so well.
I'd argue that all the "benefits" of centralized web are probably just easier to implement when everything is centralized, but can also be implemented in a decentralized fashion. Trending topics can be done by agregating local trending topics, for example.
Conflict is good for creativity. I agree: we do just have to do it. We have to use these conditions to our advantage in paving the way forward.
The world doesn't have a single necessary outcome. We ought to view our constraints as provocations.
The world doesn't have a single necessary outcome. We ought to view our constraints as provocations.
"The goal is a better open web. One on which I can chose which tools I use, without forcing my friends to do the same decision..."
Very well said. This is exactly how it started with emails and it continues as is. You can't force people, to make them follow your social network stream, you can only join the stream which should be independent.
Very well said. This is exactly how it started with emails and it continues as is. You can't force people, to make them follow your social network stream, you can only join the stream which should be independent.
Didn't identi.ca the attempt to do just that, through AtomPub and OStatus? The OStatus page even mentions PubSubHubbub as a future plan.
http://status.net/wiki/OStatus
Sadly, it seems like the Twitter compatible API gets much more love...
http://status.net/wiki/OStatus
Sadly, it seems like the Twitter compatible API gets much more love...
You are right, Ostatus is an excellent stack. However, as you pointed, developer decided to go with the "sexier" looking Twitter. Needless to say that they eventually all got rejected by the cutie.
I think we have seen this trend for a while now. The fundamental question seems to be around how open/closed an essential part of the web experience can be. As always there is a delicate balance somewhere between crazy open and crazy closed.
Without leaning towards one side or another, it seems to open side seems to be losing over time. We have seen it with discussions about open standards like OAuth:
http://hackerne.ws/item?id=4294959
It seems (maybe) like there is a ultimate trade-off between openness and ease of use/convenience and we as users usually decide with out wallet where to go. And it seems like we have voted indeed.
Without leaning towards one side or another, it seems to open side seems to be losing over time. We have seen it with discussions about open standards like OAuth:
http://hackerne.ws/item?id=4294959
It seems (maybe) like there is a ultimate trade-off between openness and ease of use/convenience and we as users usually decide with out wallet where to go. And it seems like we have voted indeed.
We just need all information in one place with an easy way to add original thoughts and an easy way to curate and discover what has been said already. Information is repeated and all over the place in pieces or outdated.
So that the people running that "one place" can then lock it down and demand money for the content?
Centralized or decentralized, just a resource everyone will know.
You mean Distributed Social Networking Protocol
https://github.com/ilhanyumer/DSNP
We need something like a next generation Usenet.
http://es.wikipedia.org/wiki/Usenet
We've reached a dangerous point. Everybody in the world joining the internet combined with a lack of knowledge about how things work means the unwashed masses gravitate towards centralized services. Instead of building distributed competitive services on top of a distributed fault-tolerant bi-directional network, we have rampant mass centralization. Protocols and interoperability are dead.
People aren't building billion dollar companies on top of application level protocols because open protocols would invite competition. Why compete when you can do a land grab for average people mindshare and lock 'em in forever?
We are fighting pressure of hundreds of millions of new pocketputers being sold every year. Our tiny idealistic cries of "more decentralization, no ads, and open data!" have the importance of a fish farting in the ocean. The teenagers and moms want to access their monkeysphere gossip. Their only care about technology is if somebody they find attractive is using it.
We can now almost legitimately say "The Internet is down!" when AWS goes tits up. People can't get news from their favorite boy bands when mysql hoses twitter (for the 7,000th time). Apple is centralizing document and data structure storage. Google could reproduce the thought patterns of a billion people if they cared enough. Distributed openness lost about five years ago.
I forgot my point. Anyway, go read http://craphound.com/overclocked/Cory_Doctorow_-_Overclocked...