[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#662736: ITP: maxwell -- entropy-gathering daemon



On 12-07-18 04:43 PM, Martin Eberhard Schauer wrote:

Package: wnpp
Followup-For: Bug #662736
Owner: "Pedro I. Sanchez"<psanchez@fosstel.com>

Package: wnpp
Followup-For: Bug #662736
Owner: "Pedro I. Sanchez"<psanchez@fosstel.com>

Package: wnpp
Followup-For: Bug #662736
Owner: "Pedro I. Sanchez"<psanchez@fosstel.com>

Package: wnpp
Severity: wishlist
Hi Pedro,

if I understand things right, you missed a CC to

|  control@bugs.debian.org

and a separate line

    thanks

to make your control commands become effictive.
|
* Package name         : maxwell
   Version              : 1.2-1
   Upstream Author      : Sandy Harris<sandyinchina@gmail.com>
* URL                  : ftp://ftp.cs.sjtu.edu.cn:990/sandy/maxwell/
* License              : GPL v2
   Programming Language : C
* Description          : entropy-gathering daemon

Daemon to gather entropy from a timer and feed it to random(4).

maxwell collects randomness from the small variations in a system
timer, distills it into a concentrated form, and sends it to random (4).

The amount of output varies with the parameters chosen, but is
generally a few K bits per second. The quality is intended to be very
high.
Both volume and quality should be adequate for most applications.

There are a number of other ways to feed entropy to random (4).
The advantage of maxwell is that it is small, simple and only minimally
hardware-dependent. The other methods also have advantages, and in many
cases one of them will be preferable to this one.
Please consider an informal review of the package description on
debian-l10n-english as well.

To me the first paragraph is an incomplete sentence and its
information is contained in the second as well. How about

     The maxwell daemon collects randomness from the small variations in a
     system timer, distills it into a concentrated form, and sends it to
     random(4).

And most of the last two paragraphs seems vague and trivial.

     Both volume and quality should be adequate for most applications.

A typical use case would help.

     The other methods also have advantages, and in many cases one of
     them will be preferable to this one.

The advantages are trivial. The alternate methods are not mentioned.
What are the criteria?

And perhaps some words on "why the name" would not hurt. Some people
primarly think of electromagnetic fields when they hear/read "Maxwell".

BTW: I strongly suggest s/random (4)/random(4)/ and appreciate the
hint to the manual page.


Martin



Hi Martin,

Regarding the missing CC to control@bugs.debian.org, I didn't know about it. But I used reportbug to send the ITP bug update and I assumed that any standard CC would be taken care of by reportbug. Anyway, shall I copy the ITP bug to control@bugs.debian.org now? or is it too late?

Regarding the package description, I just copied the verbatim the text from the upstream maintainer. I'll take your comments into consideration for the next revision of the package.

Thank you for your comments.

--
Pedro I. Sanchez


Reply to: