[RFC/P] Port System Re-Engineering

Aryeh M. Friedman aryeh.friedman at gmail.com
Mon Dec 3 09:43:01 PST 2007

Hash: SHA1

Ade Lovett wrote:
> On Dec 03, 2007, at 08:38 , Aryeh M. Friedman wrote:
>> Coding before the problem is well understood is the worst of all
>> possible solutions...
> Congratulations on snipping the relevant part of my email which
> indicated how fundamentally flawed your "survey" was.  Until such
> time as you understand the concept of self-indicatory statistics,
> which, by their very nature, exclude most of the relevant
> population, in this case, the consumers of the FreeBSD ports and
> packaging system, then you're doomed to failure.

If the questions are flawed then point out where.    If the general
concept of a survey vs. user stories vs. what ever then state which
you think is more productive.  If your problem is the medium/forum the
data is being gathered in see below.
>> specifically in many ways thats how to the port system got into
>> such a bad state....
> Further congratulations on a single sentence that, whilst you may
> not realize it, tramples all over the not inconsiderable amount of
> time that a large number of folks have supplied, for free, for your
>  benefit.

That is one reason for determining if there even is a big enough
problem to do re-engineering on (not to have their time be in vein)
> Just exactly what have you done for FreeBSD to the point where you
> believe you are able to state such a sentence?  I certainly
> wouldn't dream of writing something like this -- you can check
> ports/ commit logs for my background.

If the system was not in some state of disrepair ("a mess") then no
one would brought the issue up so by definition the sentence is a)
accurate, b) not meant to put anyone or anything in a worse light then
needs to be.
>> I have plenty ideas but am purposefully not stating them so as
>> not to color responses to the survey.... basically the idea is
>> start assume that we know nothing and gather data without passing
>> judgment and refine stuff based on what we find until there is a
>> set of features and perhaps even a roughed out framework that we
>> can be agreed on by the large majority of people... *ONLY* then
>> can we start to examine what kind of code is even needed.
> So, like I said before, start a wiki.  Mailing lists do not work
> for this kind of collaborative study.  That has already been proven
> time and time again.
> The thought of developing some kind of new-technology system, using
>  a really rather outmoded form of communication, particularly given
>  recent advances in such "social engineering", does seem to pose
> something of a conundrum.

Then the community needs to make up it's mind because I have been
criticized for making a wiki for a similar issue (SATA issues on
ICH9(R)).... You do sound like Marie Antonetta.   For the time being
as far I can tell the consensus is to keep most things on mailing
lists (not my choice).
>> Aryeh M. Friedman FloSoft Systems Developer, not business,
>> friendly
> Not business friendly?  Sorry, son, in which case you're screwed.
> You're full of talk about how things should be more
> user-orientated, yet use this as your tagline?

Developer friendly == not having to redo your work because the user
demands something they forgot to ask for.
> Which are you?  Developer, or business (and thus, by association,
> user) friendly?

Business Friendly == a lot pretty pictures that do absolutely nothing.

- --
Aryeh M. Friedman
FloSoft Systems
Developer, not business, friendly
Version: GnuPG v2.0.4 (FreeBSD)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org


More information about the freebsd-ports mailing list