Hi Haleh, I've added some comments below.

Regards, Graham.

2008/9/3 haleh mahbod <[EMAIL PROTECTED]>:
> Hi Graham,
> It looks like from the other thread that David was using this information.

Do you mean Daniel?  If not, can you point me at the thread?

> Is this something that can get generated automatically?

It is something we could generate automatically (the data, that is).
I don't know about automatic uploading to manyeyes.  The code I used
to work out the dependencies is licensed under apache 2, but I don't
have the rights to contribute it, so if we want this in our build, we
would need to write that part or ask the owner to contribute it to
Apache.

> It might be useful to make it available on the website. what do others
> think?
>
> btw. I was able to see the generated data.
>
> Haleh
>
>
> On 8/28/08, Graham Charters <[EMAIL PROTECTED]> wrote:
>>
>> Hi Raymond,
>>
>> The links work for me so I'm not sure why they don't work for you :-(
>> .  Glad you found them anyway.
>>
>> Regards, Graham.
>>
>> 2008/8/28 Raymond Feng <[EMAIL PROTECTED]>:
>> > I searched the site by "osgi" and found your entries.
>> >
>> > Thanks,
>> > Raymond
>> >
>> > --------------------------------------------------
>> > From: "Graham Charters" <[EMAIL PROTECTED]>
>> > Sent: Thursday, August 28, 2008 9:09 AM
>> > To: <[email protected]>
>> > Subject: Visualization of Tuscany OSGi dependencies
>> >
>> >> Hi All,
>> >>
>> >>
>> >>
>> >> I've been investigating ways to visualize the Tuscany runtime OSGi
>> >> dependencies.  I've managed to write out dependencies in a format
>> >> which can be viewed using manyeyes [1].  Manyeyes has a Network
>> >> Diagram view whose input format is a simple tab separated table where
>> >> the first column is the 'dependant' and the second is the 'dependee'.
>> >>
>> >>
>> >>
>> >> I've made two datasets and visualizations available on manyeyes:
>> >>
>> >> Dataset with 3rd party dependencies:
>> >>
>> >> http://services.alphaworks.ibm.com/manyeyes/data/SKsauPsOtha6bxU_c_O0Q2~
>> >>
>> >> Visualization of dataset with 3rd party dependencies:
>> >>
>> >> http://services.alphaworks.ibm.com/manyeyes/view/SKsauPsOtha6hxE8m3O0Q2~
>> >>
>> >> Dataset without 3rd party dependencies:
>> >>
>> >> http://services.alphaworks.ibm.com/manyeyes/data/SKsauPsOtha6sxUxXIO0Q2~
>> >>
>> >> Visualization of dataset without 3rd party dependencies:
>> >>
>> >> http://services.alphaworks.ibm.com/manyeyes/view/SKsauPsOtha6yxkjmRO0Q2~
>> >>
>> >>
>> >>
>> >> One of the nice features of this approach is we only need to calculate
>> >> the full dependencies then we can subset just by editing the table
>> >> data.  We can also experiment with visualizing merging bundles by
>> >> simply renaming two (or more) to the same name in the table.
>> >>
>> >>
>> >>
>> >> Perhaps we could use this to demonstrate candidate bundle
>> >> merging/refactoring prior to making the changes in the codebase?
>> >>
>> >>
>> >>
>> >> Does this seem useful? Thoughts?
>> >>
>> >>
>> >>
>> >> [1] http://services.alphaworks.ibm.com/manyeyes/app
>> >>
>> >>
>> >>
>> >> Regards,
>> >>
>> >>
>> >>
>> >> Graham.
>> >
>
>

Reply via email to