Thank you for the explanations! Your final proposal looks good to me.
On Thu, Sep 19, 2013 at 6:55 PM, Christian PERRIER <bubu...@debian.org> wrote:
> Quoting Thiemo Nagel (thiemo.na...@gmail.com):
>
>> "Overwriting ${DEVICE} with random data to prevent meta-information
>> leaks from the encrypted volume. This step may be skipped by pressing
>> 'Cancel', albeit at the expense of a slight reduction of the quality
>> of the encryption."
>
>
> Well, no either..:-)
>
> The rationale for this is that "Cancel" may depend on the debconf
> interface that is used so, in general, it is always highly discouraged
> to make explicit reference to debconf interface widgets. The same
> rationale stands for makeing reference to "Yes" or "No" in boolean
> templates.
>
> And, even if one assumes that all debconf interfaces have "Cancel"
> buttons, internationalization also brings a problem: a translator
> might translate the word Cancel in her own language in the template
> translation...while debconf has no translation for that language and
> will then show a "Cancel" button.
>
> My proposal:
>
> Overwriting ${DEVICE} with random data to prevent meta-information
> leaks from the encrypted volume. This step may be skipped by cancelling
> this action, albeit at the expense of a slight reduction of the quality
> of the encryption.
>
> I'm also worried by the first sentence not being a sentence. Indeed,
> this is (IIRC) a progress message, so the sentence not being one is
> fine....if the sentence is alone. However, as another one follows, it
> looks weird.
>
> So my final proposal:
>
> The installer is now overwriting ${DEVICE} with random data to prevent
> meta-information leaks from the encrypted volume. This step may be
> skipped by cancelling this action, albeit at the expense of a slight
> reduction of the quality of the encryption.
>
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
