Crossfire Multiprocess, Cross-Browser, Open-Web Debugging Protocol

[John J Barton]

Preprint of a paper on Crossfire is available:
http://blog.getfirebug.com/2011/04/28/crossfire-multiprocess-cross-browser-open-web-tools-protocol/

jjb

[pd]

The 'abstract' in the Abstract worries me if it is really meant to
mean 'obfuscate'.

Whilst being able to use my debugging client of choice with any
browser acting as the 'server', there's limited value if this protocol
obfuscates though standardization the variations of the various
browsers.

Good luck getting a certain vendor to cooperate :)

On Apr 29, 7:48 am, John J Barton <johnjbar...@johnjbarton.com> wrote:
> Preprint of a paper on Crossfire is available:http://blog.getfirebug.com/2011/04/28/crossfire-multiprocess-cross-br...
>
> jjb

[John J Barton]

On Apr 30, 1:23 am, pd <an0n1m...@gmail.com> wrote:
> The 'abstract' in the Abstract worries me if it is really meant to
> mean 'obfuscate'.

I'm sorry I do not understand this sentence.

>
> Whilst being able to use my debugging client of choice with any
> browser acting as the 'server', there's limited value if this protocol
> obfuscates though standardization the variations of the various
> browsers.

On the contrary, our intent is to highlight different features of
different browsers. We plan to implement client side features for the
best server side features and advise users when the browser they
happen to use does not support a feature they try.

>
> Good luck getting a certain vendor to cooperate :)

We welcome and encourage vendor cooperation, but we don't plan on
waiting for it. The only case I feel a bit bad about is Opera where
they created a remote debug system but we don't plan to use their
protocol.

jjb

[Maciej Jaros]

John J Barton (2011-05-01 17:02):

>
> On Apr 30, 1:23 am, pd<an0n1m...@gmail.com> wrote:
>> The 'abstract' in the Abstract worries me if it is really meant to
>> mean 'obfuscate'.
> I'm sorry I do not understand this sentence.
>
>> Whilst being able to use my debugging client of choice with any
>> browser acting as the 'server', there's limited value if this protocol
>> obfuscates though standardization the variations of the various
>> browsers.
> On the contrary, our intent is to highlight different features of
> different browsers. We plan to implement client side features for the
> best server side features and advise users when the browser they
> happen to use does not support a feature they try.
>
>> Good luck getting a certain vendor to cooperate :)
> We welcome and encourage vendor cooperation, but we don't plan on
> waiting for it. The only case I feel a bit bad about is Opera where
> they created a remote debug system but we don't plan to use their
> protocol.

But isn't this the only way to recieve notifications about errors on the
page? Opera doesn't have/fire onerror event. Is there some other way to
catch them?

Regards,
Nux.