A Hub for Twitter

14 views
Skip to first unread message

Julien Genestoux

unread,
Mar 1, 2010, 6:03:11 PM3/1/10
to Pubsubhubbub
Hey all,

I think everybody agrees that Twitter needs a hubs. We (Superfeedr) want to build it.

Please help us : http://bit.ly/hub4twitter the more people in there, the more comments, the more we can convince Twitter to use ours or even host their own. This has just been wayyyy to long.

Any comment, question, suggestion and intro is more than welcome. We can get them down this road.

Julien


Julien

unread,
Mar 2, 2010, 10:58:25 AM3/2/10
to Pubsubhubbub
Hey, I was expecting a little more debate around that :p

After a small day of poll, here are some results :

Do you currently use the Twitter Streaming API?
Yes 18 53%
No 16 47%

Would you use a Twitter PubSubHubbub hub if it was available?
Yes 33 97%
No 1 3%

Have you already implemented PubSubHubbub?
Yes 24 71%
No 10 29%


Obviously, 34 is _not_ a big enough number that I think we have a
representative panel of respondant, but we also have "big" names in
here, (including some who have access in the firehose), which makes me
think that PubSubHubbub should be a viable option for Twitter.

If you read this, please take some take to respond :

http://bit.ly/hub4twitter

Thanks all.

Cheers,

Julien

On Mar 1, 3:03 pm, Julien Genestoux <julien.genest...@gmail.com>
wrote:


> Hey all,
>
> I think everybody agrees that Twitter needs a hubs. We (Superfeedr) want to
> build it.
>

> Please help us :http://bit.ly/hub4twitterthe more people in there, the

kang

unread,
Mar 2, 2010, 11:10:54 AM3/2/10
to pubsub...@googlegroups.com
Hi, Jullien,

I think Twitter may have its own plan about real-time web. But I support your hub!

Kang
--
Stay hungry,Stay foolish.

Andrew Badera

unread,
Mar 2, 2010, 11:03:44 AM3/2/10
to pubsub...@googlegroups.com
As I said over on twitter-dev ... if streaming gives me everything I
need, (which it almost does now, and supposedly may do in the future,
plus commercial considerations may solve my problems) and it's really
not that complex (which it's not) then why replace it with an
alternative that doesn't, to me at least, seem to offer any hugely
distinct advantages? Or, what advantages am I missing? Don't get me
wrong, I <3 PuSH, but Twitter Streaming is already in place and
invested in ...

∞ Andy Badera
+1 518-641-1280 Google Voice
∞ This email is: [ ] bloggable [x] ask first [ ] private
∞ Google me: http://www.google.com/search?q=andrew%20badera

Julien Genestoux

unread,
Mar 2, 2010, 11:24:30 AM3/2/10
to pubsub...@googlegroups.com
As I said on the Twitter Dev Group. I think you miss the fact that it's a "standard" and "inter-operable". It's very counterproductive to have 36 different APIs for 36 different services. The web is what it is because people have agreed to build websites with the same standards, which makes it easy to consume : HTTP, RSS/Atom, email... 

There are also obvious technical advantages, like the fact that you can actually subscribe and receive notifications at the same time, which is not the case with the streaming API.

Jeff Eddings

unread,
Mar 2, 2010, 11:41:47 AM3/2/10
to pubsub...@googlegroups.com
As a very interested lurker, I'm with Julien on this one.  There are many reasons why it's good for all of us to have Twitter implement an agreed-upon standard like PubSubHubbub, while there are many good reasons for Twitter to implement using their own proprietary API.  Ultimately Twitter using an available standard is better for everyone involved (including them), but in the short run, Twitter has every reason not to use it, and developer solidarity around a standard helps us all.

Jeff

richardlusk

unread,
Mar 2, 2010, 1:16:31 PM3/2/10
to Pubsubhubbub
Yes.

> > Please help us :http://bit.ly/hub4twitterthemore people in there, the

Jason Walniuk

unread,
Mar 2, 2010, 1:28:08 PM3/2/10
to Pubsubhubbub
As a casual observer to the list and fan of PuSH, I can't help but
wonder why on earth Twitter would want to make a small 3rd party
startup responsible for something they could turn around just as well
in a couple days.

PuSH is dead simple. If they wanted to supported they could, very
easily. Considering how well they seem to be managing their stream
API, I would much rather have THEM manage a PuSH-based delivery option
as well.

Jason

On Mar 1, 6:03 pm, Julien Genestoux <julien.genest...@gmail.com>
wrote:


> Hey all,
>
> I think everybody agrees that Twitter needs a hubs. We (Superfeedr) want to
> build it.
>

> Please help us :http://bit.ly/hub4twitterthe more people in there, the

tyler gillies

unread,
Mar 2, 2010, 1:45:52 PM3/2/10
to pubsub...@googlegroups.com
I had the same thoughts
--
Everyone Loves Tea
http://www.everyonelovestea.com

Julien Genestoux

unread,
Mar 2, 2010, 1:54:22 PM3/2/10
to pubsub...@googlegroups.com
Jason, Tyler... read the first lines of the first message:
"Please help us : http://bit.ly/hub4twitter the more people in there, the more comments, the more we can convince Twitter to use ours or even host their own. This has just been wayyyy to long."

Obviously, using ours can be a nice proof of concept for no work (the problems today are more EULA related than tech related) and since our hubs are fully "open", they can export all the data at any time, if they want to go down the route of implementing theirs. We even offer the ability of using CNAMEs, which would make the switch to theirs (or another provider) fully transparent.

I think everybody wins if they decide to implement theirs or use ours : an open protocol for realtime that Twitter wouldn't be using would still be a 3 legged horse, no matter what. We (Superfeedr) bet on that horse, I want it to have 4 legs.

tyler gillies

unread,
Mar 2, 2010, 2:10:54 PM3/2/10
to pubsub...@googlegroups.com
One potential problem i see is that twitter has started selling access to their streaming API, which is kinda their version of PuSH.
Wouldn't implementing PuSH go against their monetization strategy?

Julien Genestoux

unread,
Mar 2, 2010, 2:15:13 PM3/2/10
to pubsub...@googlegroups.com
I disagree... PubSubHubbub cannot be compared to a firehose, unless you subscribe to all the twitter feeds, which is quite hard to do.
PubSubHubbub has the exact same limitations that they have with their different "levels" of free streaming API access.

Julien

Andrew Badera

unread,
Mar 2, 2010, 2:20:18 PM3/2/10
to pubsub...@googlegroups.com
On Tue, Mar 2, 2010 at 2:15 PM, Julien Genestoux
<julien.g...@gmail.com> wrote:
> I disagree... PubSubHubbub cannot be compared to a firehose, unless you
> subscribe to all the twitter feeds, which is quite hard to do.
> PubSubHubbub has the exact same limitations that they have with their
> different "levels" of free streaming API access.
> Julien
>

Why is it hard? Easy enough to subscribe to millions of Buzz feeds, as
I've been working on this past week ..

--ab

Julien Genestoux

unread,
Mar 2, 2010, 2:23:14 PM3/2/10
to pubsub...@googlegroups.com
It's easy to subscribe to millions, but it's hard to subscribe to all. I don't think anybody was ever able to subscribe to _all_ the feeds from a given hub, simply because there is no discovery mechanism.

James Holderness

unread,
Mar 2, 2010, 4:08:24 PM3/2/10
to Pubsubhubbub
Julien Genestoux wrote:
> As I said on the Twitter Dev Group. I think you miss the fact that it's a
> "standard" and "inter-operable". It's very counterproductive to have 36
> different APIs for 36 different services. The web is what it is because
> people have agreed to build websites with the same standards, which makes it
> easy to consume : HTTP, RSS/Atom, email...

My understanding of the Twitter streaming API is that it's a client-
server protocol that can be used from desktop applications. PuSH is
only a server-to-server protocol so it doesn't really serve the same
purpose.

Even if you think that PuSH is somehow more of a "standard" (which I
think is debatable), it just doesn't fit their requirements (IMO).

Julien Genestoux

unread,
Mar 2, 2010, 5:27:10 PM3/2/10
to pubsub...@googlegroups.com
On Tue, Mar 2, 2010 at 1:08 PM, James Holderness <j4j...@gmail.com> wrote:
Julien Genestoux wrote:
> As I said on the Twitter Dev Group. I think you miss the fact that it's a
> "standard" and "inter-operable". It's very counterproductive to have 36
> different APIs for 36 different services. The web is what it is because
> people have agreed to build websites with the same standards, which makes it
> easy to consume : HTTP, RSS/Atom, email...

My understanding of the Twitter streaming API is that it's a client-
server protocol that can be used from desktop applications. PuSH is
only a server-to-server protocol so it doesn't really serve the same
purpose.

That's a fair point. However, I'm not sure of _any_ desktop app that uses the streaming API yet, hwoever, I know many web app who have been forced to use it :(

Julien

unread,
Mar 3, 2010, 3:10:40 PM3/3/10
to Pubsubhubbub
I just posted the results : http://blog.superfeedr.com/API/PubSubHubbub/Twitter/feeds/streaming/a-hub-for-twitter/

And shared them by email with John and Ryan at Twitter. I think we can
only wait for them to make the right decision now :)

Cheers!

Julien

On Mar 2, 2:27 pm, Julien Genestoux <julien.genest...@gmail.com>
wrote:

Reply all
Reply to author
Forward
0 new messages