Re: NAMD Non-features

From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Wed Jan 21 2015 - 15:54:49 CST

On Wed, Jan 21, 2015 at 4:24 PM, JAVAD NOROOZI <noroozijavad_at_yahoo.com> wrote:
> Axel,
>
> I never meant to make you guys angry.I know that before solving a problem,

i am not angry, but if you make a gross generalization like you did
and effectively blaming somebody else for not doing your due
diligence, you have to be able to deal with people disagreeing with
you and telling you where you went too far.

> first one need to find a suitabe tool or be smart enough to build his/her
> own tool.
> Non-features are usually is discovered after setting up a system , running
> the simulation and getting an error messages.

no. if you understand the methods you want to use (and you better do
so or else you'll have a big mess at your hands), you have to be able
to tell whether a software has all the features you need or not. there
may be issues with some of the technical details and limitation of the
implementation, but all the things that you list are clearly issues
that you could have determined to be not available way ahead of time.

your logic is inverted. a software should state what it *can* do, and
NAMD does so. everything else is by definition a non-feature. if you
cannot tell the difference, then that is your problem. don't blame
others for that.

> Lammps has listed some of its Non-features in the begining of the manual, and you could perhaps do so.

that really isn't a statement of "what LAMMPS cannot do". that list
would have to be much, much longer. rather that list is (obviously) a
political statement. its purpose is to tell, "this is what we do not
*expect* LAMMPS to be able to do". ...and if you look closely, you'll
see that at least one point, visualization, is no longer true. i gotta
tell steve to update the page... ;-)

> Finally i got what i wanted from NAMD, but this would help people to make a
> proper choice.

again, i doubt it would make a difference for most people. like i
wrote in my previous response, most people will not even look for a
sports car when they need a van. and you don't have to be a genius to
tell the difference.

axel.

> Javad
>
>
>
>
>
>
>
>
>
>
>
> ________________________________
> From: Axel Kohlmeyer <akohlmey_at_gmail.com>
> To: NAMD list <namd-l_at_ks.uiuc.edu>; JAVAD NOROOZI <noroozijavad_at_yahoo.com>
> Sent: Wednesday, 21 January 2015, 14:54:41
> Subject: Re: namd-l: NAMD Non-features
>
> On Wed, Jan 21, 2015 at 2:54 AM, JAVAD NOROOZI
> <noroozijavad_at_remove_yahoo.com> wrote:
>> To dear developers,
>>
>> In a few month working with NAMD , i came across with some important
>> Non-features which hinder working with NAMD.
>
> strangely enough a ton of people *are* working with NAMD and quite
> successfully and happily so, thus what you call "important" doesn't
> matter to them. you have to rephrase this to "hinder *me* working with
> NAMD".
>
>> NAMD does not support...
>>
>> 1. non-equilbirum MD (multiple tcouple , multiple tcl forces, examples
>> are
>> measuring thermal conductivityand shear viscosity)
>>
>> 2. dummy or virtual sites (except for TIP4 water)
>>
>> 3. Rigid bond between two heavy atoms
>>
>> 4. Rigid bodies
>>
>> 5. Efficient solvation free energy methods( i.e , Expanded Ensemble , BAR
>> ,...)
>
> you sound like the guy with six children who bought himself a sports
> car and now complains that it doesn't seat whole family. that is why
> software has documentation and why smart people carefully study that
> documentation and what other people do with a software before they
> spend weeks of their time trying to make it work (unless they are
> developers, of course, and have the intent of implementing the
> features they are missing).
>
> i understand your frustration, but you have to realize that this is
> self-imposed and could have been easily avoided.
>
> axel.
>
>
>
>
>>
>> Javad
>
>>
>>
>>
>>
>>
>>
>>
>
>
>
> --
> Dr. Axel Kohlmeyer akohlmey_at_gmail.com http://goo.gl/1wk0
> College of Science & Technology, Temple University, Philadelphia PA, USA
> International Centre for Theoretical Physics, Trieste. Italy.
>
>
>
>

-- 
Dr. Axel Kohlmeyer  akohlmey_at_gmail.com  http://goo.gl/1wk0
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2015 - 23:21:34 CST