[boost] Re: FC++ -- Formal Review Result

From: David Abrahams (dave_at_[hidden])
Date: 2004-04-06 14:18:05

"Andy Little" <andy_at_[hidden]> writes:

> "Mat Marcus" <mat-boost_at_[hidden]> wrote in message
> news:603536759.1081186934@[10.51.241.93]...
>
>> I respectfully ask you to refrain from posts of this nature in the
>> future as others on this list may also feel that they are rather
>> offensive.
>
> I think I would be happy at the moment,especially if this is the general
> view, to be officially notified, which I hope will happen in due
> course.

Please consider yourself "officially notified".

> If that is the case I am sure that this reflects badly on me rather
> than Mr MacNamara.

It certainly does reflect badly on you, but there's a lot more at
stake here for Boost than merely looking good.  We wish to remain a
community where people look forward to the rigorous experience of
having their libraries reviewed, without fear that they'll be
gratuitously insulted and condescended-to.

On a human level, I personally have some concern for Brian's feelings,
and of those who may empathize with him.  Fortunately, Brian is
emotionally tough, but that won't be true of everybody.  There's a
great saying in the Python community: "always remember that a few
people on this mailing list are only fifteen years old".  I don't know
whether we have any 15-year-olds on the Boost list, but we might.

I want to be clear that there's nothing wrong with dissent on
technical issues.  The problem I have is that most of your last two
posts seem to have very little technical content and lots of
ad-hominem subtext.

Since it's not clear whether you are being intentionally hostile or
whether you simply don't understand the effects of your words, I will
now attempt to break down the offensive elements of your last two
posts and explain their impact, at least on me:

>> > I would suggest that C++ is not the solution for you

This strikes me as condescending in the extreme.  Brian has been doing
work *of community interest* in C++ since at least 2000, and was using it
for who-knows-how-long before that.  I'm sure he's well aware of its
capabilities and limitations.  Further, he probably has exposure to
more languages than most of us, making him more than qualified to make
such determinations on his own.  

>> > but OTOH neither is the dogma of FP. 

This part is not particularly offensive, but it is narrow-minded and
is phrased in a way that takes for granted an idea (that FP is more
dogmatic than, e.g., OO, GP, top-down, bottom-up, aspect-oriented,
etc...) for which, I'm sure you are aware, there is strong
disagreement in the group.  The polite thing to do in that case is to
indicate somehow that you realize it's your opinion as opposed to
fact.

>> > Whether anyone has the guts to employ you I have no idea, but I
>> > wish you all the best for the future.

Condescension and insult.  The implied message is "you will have a
hard time finding a job because you are incapable of doing anything of
practical value".  Having worked a bit with Brian outside of Boost, I
can say that I'd be very happy to hire him if I could.  His
professionalism and attention-to-detail are second to none.

>From your follow-up message:

> Functional programming is a great idea.

It certainly comes as a surprise to hear you say that in light of your
earlier posts, to the point that it's hard to guess whether you are
being honest.  Not wanting to ascribe malice, I am happy to see that
you're trying to soften the effects of your original posting a little.

> The general response of the 'boost community', which purports to
> embrace FP

This expresses a general lack-of-respect for everyone here.  Putting
'boost community' in quotes adds an implicit "so-called" prefix to the
subtext, as if to imply that there really is no legitimate community
at all.  "Purports to embrace FP" offends in two ways: first, it
ascribes a single philosophical preference to the whole 'community',
which is really made up of individuals (as though there were a Boost
FP manifesto somewhere), and then it indicts the 'community' for being
dishonest about its stated philosophy.

> Brian took a lot of time out with me during the FC++ review. My
> previous post was to say thanks, I have learnt something, and I
> appreciate that

I was shocked to read that your previous post was meant as a "thank
you"; I hope you now understand that its effect was... something else
altogether.

> and despite my non-boost community attitude, which
> is increasing exponentially, I did gain a lot from it. 

I'm not sure how to take the part about "non-boost community attitude"
at all; it appears to say something like "I dislike you people more
and more as time goes by".

> However I get the impression, which I stated in the review that
> C++'s place is as an Assembly language for FP. To (try to) use FP in
> C++ expressions is not very productive.

Legitimate technical content, stated appropriately as opinion.  Thank
you.

> ... sorry lambda.

Somewhat condescending.  The Lambda library doesn't need to please
you in order to be successful.

-- 
Dave Abrahams
Boost Consulting
www.boost-consulting.com