Re: GNU IceCat & SoC

[Giuseppe Scrivano]

The problem is that we don't have clear ideas about the first
suggestion too :)

This is why the first part of this job is about investigate new cases
that can be a problem for the privacy of users.

At the moment two possible scenarios are covered by our privacy
features, as described on the project homepage: 

  1. Some sites refer to zero-size images on other hosts to keep track 
  of cookies. When IceCat detects this mechanism it blocks cookies
  from the site hosting the zero-length image file. (It is possible to
  re-enable such a site by removing it from the blocked hosts list.)

  2. Other sites rewrite the host name in links redirecting the user
  to another site, mainly to "spy" on clicks. When this behavior is
  detected, IceCat shows a message alerting the user.

At the end of the task I think the student will have to detect and
implement at least other two or three possible cases.
Especially in the first part, the student will have to communicate
with the mentor to discuss new possible situations and only later,
when they are well defined, continue to implement them.

I know this task has a vague description, but unfortunately we can't
give more details as we don't know them too; if it is too vague for a
possible Summer of Code task then we should drop it. 

Regards,
Giuseppe



"James Youngman" <address@hidden> writes:

> On Sun, Mar 9, 2008 at 12:49 PM, Giuseppe Scrivano <address@hidden> wrote:
>> Hello,
>>
>>  For GNU IceCat I have two ideas:
>>
>>  1) Privacy features:
>>  *  Investigate new cases and implement them.
>>  *  Leave in the low level code only the observers and re-implement the
>>    rest as an extension.
>>
>>  2) Port GNU IceCat to Firefox 3:
>>  *  Update IceCat source code to Mozilla Firefox 3.
>>  *  Update the plugins list, create .xpi packages where it is needed.
>>
>>
>>  I can be mentor for them.
>
> Thanks for your suggestions!
>
> The second suggestion seems well-defined, but the first one (privacy
> features) is very open-ended.  That in itself is fine, but very often
> students just choose an idea and volunteer to do it.   They assume
> that because the idea is on the suggestions page, the idea itself is
> acceptable.  However, my own assumption is that a SOC proposal which
> just repeated the text of the suggestion would be rejected as being
> too vague.    So, we must expect the student to expand the suggestion
> into a more detailed proposal.
>
> Could you please update your suggestion to make that more explicit?
> That is, I'd prefer the suggestion to tell the student that their
> proposal needs to specify in more detail what they propose to do.
>
> If the suggestions also included pointers to more information
> (information the student could read in order to figure out what the
> project might entail, for example) that would be good, too.     The
> student's proposal would need to include answers to these kinds of
> questions:
>
> 1. What will the output of your project be?   What things will be
> improved/created?
> 2. How will you know when you're done?
> 3. What will the project look like at the half-way evaluation point?
> 4. Of the skills you need to complete the project, which do you
> already have and which will you need to learn?
>
> Obviously, I'm not suggesting that we must hand that info to students
> on a plate, but I think it would be useful for the suggestion to
> provide a little help.  I believe that the quality of the proposals we
> get will improve if we set out clear suggestions.
>
> Thanks,
> James.