On Cerebro, Telepathy, yokes and whites (was Re: cerebro in sugar)

Robert McQueen robert.mcqueen at collabora.co.uk
Tue Jun 10 21:49:47 EDT 2008


Ricardo Carrano wrote:
> On Tue, Jun 10, 2008 at 3:36 PM, Benjamin M. Schwartz
> <bmschwar at fas.harvard.edu> wrote:
>> Polychronis Ypodimatopoulos wrote:
>> | I thought we were talking about collaboration. MSN, IRC etc are
>> | basically chat protocols. Cerebro has little to do with such protocols;
>> | its goal is to provide efficient and scalable presence and data sharing
>> | in an ad-hoc, mobile environment where even IP addresses are a burden to
>> | maintain.
>>
>> AIM, MSN, IRC, and Cerebro are all protocols that provide message
>> channels, file transfer, and presence info.  Cerebro is different in that
>> it is designed to run on networks that do not have strong routing
>> guarantees, but it ultimately provides a very similar set of features.
> 
> A swiss army knife is never as simple as a knife. When what you need
> is a knife, trying to find the proper thing to pull out your swiss
> army knife may cost your life. We've made a mistake if we decided that
> an all encompassing framework is what we needed in an XO. That's in
> the past and certainly there was a good reason by the time. The real
> question is: can we fix this mistake now?

If you say you need to see presence, send text messages, transfer files
*and* transmit streams of data for collaborative applications, then we
don't just need a knife, we need the screwdriver, corkscrew and can
opener. A swiss army knife isn't such a bad option, unless you want to
carry around a whole toolbox. :)

More seriously, I don't think it's fair to qualify this as a mistake,
and in fact I'm not that sure Telepathy should be described just as "an
all encompassing framework". It's primarily an API for accessing a
variety of functionality which are common to different communications
mechanisms.

I'm not (too) ashamed to admit that for various reasons, some errors in
judgement were made with the implementations we've ended up with, but I
don't believe these are inherent to the API in use, and we've got plans
for rectifying them. Making and testing a Telepathy implementation that
uses Cerebro is part of these plans.

> Cheers
> Ricardo Carrano

Regards,
Rob




More information about the Devel mailing list