incubator-esme-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ethan Jewett <esjew...@gmail.com>
Subject Re: Is doing PubSubHubub worth it ?
Date Sat, 04 Sep 2010 10:53:21 GMT
I think we just need the authenticated RSS feeds first. Another issue
dependent on the RSS feed issue would make sense. I'll try to remember
to do it when I get home.

Agreed on it being in the Backlog, not 1.2.

Ethan

On Saturday, September 4, 2010, Richard Hirsch <hirsch.dick@gmail.com> wrote:
> Do we want to create another JIRA issue for Pubsubhubbub publishing?
> Don't know whether it would be 1.2 though:
>
> D.
>
> On Sat, Sep 4, 2010 at 10:49 AM, Imtiaz Ahmed H E <in.imtiaz@gmail.com> wrote:
>> I think I'll just go ahead and do it and the rest of you can review my
>> completed stuff and it's working in the svn pubsubhubub-2 branch and decide
>> whether to go ahead and integrate with trunk or not.
>>
>> I'll try to deliver quickly now!
>>
>> Imtiaz
>>
>> ----- Original Message ----- From: "Ethan Jewett" <esjewett@gmail.com>
>> To: <esme-dev@incubator.apache.org>
>> Sent: Saturday, September 04, 2010 1:40 PM
>> Subject: Re: Is doing PubSubHubub worth it ?
>>
>>
>>> It's a good question - perfectly reasonable. As Dick mentioned, I
>>> don't think that list is complete, but it's true that not everybody
>>> has implemented PuSH.
>>>
>>> In my mind, Pubsubhubbub is a nice-to-have at this point. It's
>>> definitely not one of the highest priority items. It's something that
>>> I was interested in implementing as an experiment and it would
>>> immediately deliver some additional nice functionality in the form of
>>> instantly updating actions and less server time spent downloading RSS
>>> feeds every X minutes.
>>>
>>> Longer-term, there are more interesting implications if we get
>>> PuSH-enabled, authenticated feeds of ESME user timelines (with ESME as
>>> a publisher and hub). Once this happens, it is another good way to
>>> integrate ESME with other systems on a user-by-user basis. Eventually
>>> it might be a way to federate ESME instances, but there are a lot of
>>> complications that are not easily resolved, including what is done
>>> about pools in this scenario and how authentication works (PuSH
>>> authentication is not a very well-resolved problem).
>>>
>>> So, that's just my partially-formed thinking on the topic. It was
>>> pretty much a science-project for me that didn't get very far. I
>>> definitely think of it as a nice-to-have rather than a "must-have".
>>>
>>> Ethan
>>>
>>> On Sat, Sep 4, 2010 at 9:58 AM, Richard Hirsch <hirsch.dick@gmail.com>
>>> wrote:
>>>>
>>>> There are some other companies not on the list but present in the
>>>> comments (RabbitMQ and status.net) - there are also some interesting
>>>> other possibilities (for example, linking it to xmpp (
>>>> http://grack.com/blog/2009/09/09/pubsubhubbub-to-xmpp-gateway/) or
>>>> this blog about pubsubhubbub and twitter
>>>> (http://blog.persistent.info/2009/08/twitter-pubsubhubbub-bridge.html)
>>>> Since we've already started the implementation, I'd really be
>>>> interested in continuing it but I'm open for discussions on the topic
>>>> as well.
>>>>
>>>> D.
>>>>
>>>> On Sat, Sep 4, 2010 at 9:11 AM, Imtiaz Ahmed H E <in.imtiaz@gmail.com>
>>>> wrote:
>>>>>
>>>>> Please don't misconstrue me, I would have (and already started having)
>>>>> my usual ball doing it!
>>>>>
>>>>> But at http://code.google.com/p/pubsubhubbub/wiki/Companies
>>>>> there are only a handful of companies who have advertised themselves
as
>>>>> doing (supporting a hub for their feeds) it.
>>>>>
>>>>> Of course, Ethan probably knows better...
>>>>>
>>>>> What do you think ? Dick ?
>>>>>
>>>>> Maybe there's lots of other stuff that would really help Esme. But,
>>>>> pubsubhubub is a really worthwhile api/concept in itself...
>>>>>
>>>>> Imtiaz
>>>>>
>>>>>
>>>>
>>
>>
>

Mime
View raw message