[Date Index] [Thread Index] [Author Index]

Re: Re: Mathematica: Long divison for polynomials

• To: mathgroup at smc.vnet.net
• Subject: [mg84044] Re: [mg83975] Re: Mathematica: Long divison for polynomials
• From: Daniel Lichtblau <danl at wolfram.com>
• Date: Fri, 7 Dec 2007 03:12:14 -0500 (EST)
• References: <200711301018.FAA04867@smc.vnet.net> <firel5$rsq$1@smc.vnet.net> <200712060737.CAA10055@smc.vnet.net>

Caren Balea wrote:
> Thank you for your answers!
> 
> As the reply in the newsgroup is quite slow I did try a different newsgroup
> and receive an answer right after 21 minutes (rather than to have to wait
> a whole day in this newsgroup):
> 
> http://groups.google.it/group/sci.math.symbolic/browse_thread/thread/bd8e9a5475af5fe9?hl=it
> 
> There is also an interesting discussion about how meaningful it is
> to have a moderation in this particular newsgroup.
> 
> Frankly, I agree with what they are saying.
> I'm curious though whether my post is going to be displayed or not.
> 
> [I am offering a totally free service to those who want to use it.
> You should probably go elsewhere if it does not meet your needs.
> -- Moderator] 

There has been substantial commentary in sci.math.symbolic, going back 
several years, over the various ups and downs of having a moderated 
forum comp.soft-sys.math.mathematica. Here is a URL to a reply I made in 
one such thread, a few months ago.

http://groups.google.com/group/sci.math.symbolic/browse_frm/thread/302a47fd9af78063/cf8bfd0f577dba67?hl=en&lnk=gst&q=Lichtblau+mathgroup+moderated#cf8bfd0f577dba67

I've had some involvement with this since the beginning so I think I can 
clarify and address a few issues. Bear in mind that I am replying as one 
who sometimes writes to the forum; neither this reply, nor other posts 
to the group, are part of my official job responsibilities.

First, the group began life, and remains, as a moderated mailing list 
forum known as MathGroup. It was extended to a Usenet group around 12-13 
years ago. I think this was partly in response to a need to remove the 
many Mathematica-centric posts from sci.math.symbolic. And also there 
was a desire to extend the mailing list reach to more people for whom it 
was an appropriate venue, but who might not otherwise find it.

So what are the gripes about this moderated group? They fall into a few 
categories.

(1) Turnaround time. It takes a day or so for a post to appear. So what 
are the disadvantages to the relatively slow turnaround? One is that 
feedback to the person posing questions is slow. This, I concede, is a 
very real drawback. But short of cloning the moderator, I don't see a 
good way around it. Suffice it to say, the advantages to having a 
moderator (or at least to having the one we have), in my opinion, far 
outweigh this. More below.

(2) Another perceived drawback to the turnaround time is replication of 
effort in responses. I've seen this issue raised but frankly I think it 
is mostly baloney. The vast majority of responders know good and well 
that others are also likely to respond, and they do so anyway. The fact 
is, Mathematica is a complex program, and often there are multiple ways 
to achieve a stated goal. Responders know this and oftentimes "best 
practice", or several reasonable forms thereof, emerge in the 
multiplicity of responses. About the only actual drawback is that some 
responders might delay an extra day to see if others answer first, and 
this can prolong the process per (1) above.

(3) There has been persistent griping about the moderation amounting to 
"censorship". Depending on how you define the terms, I suppose perhaps 
this can be a valid sort of remark. But the upshot is that whether you 
call it moderating or censorship, the moderator keeps a huge amount of 
garbage out of the group. If you have not seen similar groups 
effectively crippled by rants, be happy. Likewise with spam. Suffice it 
to say that these problems have arisen and persisted for long stretches 
of time. But never, absolutely never, has such a fate befallen 
comp.soft-sys.math.mathematica. The group has remained consistently 
useable, with outages only in rare times (like two or three over a 
period of a dozen years) when the moderator's systems or ISP have 
suffered grievous damage.

(4) One particular censorship gripe concerns the rule against mention of 
competing programs. I myself am a bit sympathetic to this one; if there 
were a single restriction I could change, this would be it. That said, I 
think that unrestricted allowance of such could pose a problem. That is, 
we might see the sort of degeneration into rants ("Why can't Mathematica 
do X? Program MZZZ does it?" followed, a day later, by "Why can't 
Mathematica do Y?..."). In past I've seen at least some tendency toward 
such posts, and it is generally thwarted by this restriction, which 
means the restriction has some merit. Also it helps to keep the focus on 
Mathematica per se, which is what the group is all about.

As to what are the best places to take Mathematica related questions, 
answers will vary and to some extent it can depend on what you need. 
MathGroup aka comp.soft-sys.math.mathematica remains a compelling venue 
when the one day turnaround is acceptable, and high quality of response 
is needed. It is the forum with the most expertise. I think this is in 
large part because the moderator keeps out garbage, thus making the 
signal-to-noise ration quite high. Were it to degenerate to some of the 
things I have seen, I doubt most of us would maintain interest for long. 
So yeah, the turnaround time is an issue, but there are advantages to 
the moderating that, for myself at least, far outweigh this drawback.

Daniel Lichtblau
Wolfram Research

• References:
  • Re: Mathematica: Long divison for polynomials
    • From: Caren Balea <caren_balea@xemail.de>