Re: [tlug] Open Access Journals

[Attila Kinali <attila@example.com>]

On Mon, 24 Mar 2014 11:59:24 +0900
"Stephen J. Turnbull" <stephen@example.com> wrote:

> Attila Kinali writes:
> 
>  > In some fields, i have the impression that 90% of the papers were
>  > just published because a poor student was given a bad project and
>  > told to publish a paper out of it or he will fail.
> 
> That's because the professors themselves typically can't evaluate each
> others' research and just count publications (and nowadays, grants
> more than publications).  (The advisor has strong incentive to promote
> his students so he doesn't really count.)

How come professors cannot evaluate each others work? If they are
working in the same (or at least nearby) field, then they should
know enough to judge whether something is potentially usefull or not?
Or am i just to naive in this regard?

And even if the advisor has a strong incentive to promote his students
(what is that incentive anways?), shouldn't he try to get his students
to do "real stuff" instead of doing just another expedition in an already
known dead end, with no guidance and even less incentive to try new things?

Ok, i think i have to differentiate here a little bit. For a master thesis
it's not a bad idea to replicate something other people have done. Even
if it has been done multiple times. It's a good teaching tool and gives
the student a hands on experience on how to do stuff (i'm thinking here
about taking known method X from literature, implement it and see how it
works and perform. Especially evaluating its strengths and weakenesses,
which are usually not mentioned). And for the professor it is a good tool
to see whether the student has any potential to perform in any meaningful
way in a PhD.

But during a PhD (or any other research position), i expect people to do
something that andvances the field in some way. Be it by exploring things
no one has done before, refining known methods or simply summarizing and
evaluating other peoples work.

>  > And yet these papers get published. I am not sure
>  > why, but they are there.
> 
> Sure.  The reason they get published is that second- and third-tier
> researchers have discovered that publishing many papers is more
> important in achieving status and promotion (for them) than publishing
> good ones is.

I always wondered how this system came to be. What external motivation
is there to publish as much as possible? Ok, if you don't publish (much)
your name will not be known. But isn't a good paper a year worth more
than 10 bad ones?

I'm by far not an expert. But reading a good paper from someone gets
me to look at his publication list and read his other papers too.
While reading a bad paper makes me put someone on my "blacklist".


>  With creating a new journal becoming cheaper all the
> time, there's an incentive to ensure that (a) everything you write
> gets published and (b) multiple times ;-), a goal that can be achieved
> if you and a bunch of buddies review and approve and cite each others'
> papers in these journals.

Yeah. Cronyism is everywhere. Works great to get you promoted, 
until the moment someone from outside reviews your performance,
or you are cut of your team. Unfortunately, science works in a
way where outside review is hard, if not impossible. Especially
if you are an university administrator.

>  > And i don't think that any change in publishing system will make
>  > this much better or much worse.
> 
> I think the change in cost structure over the last 50 years has made a
> big difference in many fields, I'm not sure if that's included in what
> you mean by "publishing system," though. 

No it doesn't. I'm not that old yet, to know what has been going on 
in the last 50 years :-) I've been reading scientific papers for about
15 years, give or take a couple. And ever since, i accessed them over
the local university. Which makes me oblivious to the real cost of those
papers. Also, i've never published anything, so i don't know how that
works either.

Can you tell us what happend to the cost structure in the past and how
it affected science?

> I think it needs to be
> addressed by changing the culture of scientific research.  Mitchell
> Waldrop's "Complexity: The Emerging Science at the Edge of Chaos" is
> an interesting read here.

Hmm? How is it interesting? Beside mentioning of the Santa Fe Institute
it does not say anything about how research is done or how it is
paid for.


			Attila Kinali

-- 
The trouble with you, Shev, is you don't say anything until you've saved
up a whole truckload of damned heavy brick arguments and then you dump
them all out and never look at the bleeding body mangled beneath the heap
		-- Tirin, The Dispossessed, U. Le Guin