[Q-e-developers] qe package

Adrian Hühn adrian.huehn at thch.uni-bonn.de
Thu Aug 18 12:34:57 CEST 2016


I hope, I am not impudent if I interfere your discussion but maybe I can 
contribute with my personal perspective as newcomer (I used QE for my 
master thesis last year and was on my own, as nobody in my workgroup 
actually uses it). Before I built QE for the first time manually, I took 
a build script from the community of Arch Linux (the script automates 
built and installation to common bin and lib paths). So it made several 
days of fiddling as the installation directories and the paths in the 
documentation differ. I end up only using the built directory and never 
install via "make install".

I think that it would be beneficial if the packaging is more close to a 
layout typically found in Linux distributions, i.e. separating source 
code, documentation and examples. I see several advantages to this:
- people familiar with Linux/UNIX but new to QE would have a much 
smoother learning curve
- package maintainers have an easier way to set up installation directories
- file path in the documentation would be more consistent for "regular" 
linux package


Am 18.08.2016 um 11:20 schrieb Stefano Baroni:
> Possiamo graduare la distinzione fra pacchetti “core” e gli altri (ad 
> esempio distribuendo per default neb.x, ph.x, e pochi altri), ma 
> secondo me il meccanismo attuale va mantenuto. SB
>> On 18 Aug 2016, at 11:13, Filippo Spiga 
>> <filippo.spiga at quantum-espresso.org 
>> <mailto:filippo.spiga at quantum-espresso.org>> wrote:
>> I am *PERSONALLY* aligned with Nicola's way of thinking. A single 
>> package would simplify a lot, including the perception to the public 
>> about what QE is and what is part of QE distribution. We can continue 
>> to have third-part packages following this "on-demand" model (West? 
>> EPW? SaX?) but NEB, PH, TDDFT and others packages that exist since 
>> ages can be collected under the same umbrella.
>> I have always believed that the reason we had many packages is to 
>> avoid a monolithic heavy distribution. Based on what I see, the core 
>> source code is not "that big" in size.
>> I personally see some beauty and some practicality in changing the 
>> packaging process. The 6.0 will continue to follow the current 
>> process unless the majority of contributors agree differently. But 
>> because 6.0 is going to introduce already some new stuff, I 
>> personally think this is a good time to review the packaging process 
>> as well.
>> Just my 2 cents ...
>> On 18 Aug 2016, at 09:51, Stefano Baroni <baroni at sissa.it 
>> <mailto:baroni at sissa.it>> wrote:
>>> Then we have simply to beat the drum by claiming that our “virtual” 
>>> (or whatever fancy adjective you may find) distribution model is 
>>> “innovative” and much better than the old-fashioned tar balls … SB
>> -- 
>> Filippo SPIGA
>> * Sent from my iPhone, sorry for typos *
> _______________________________________________
> Q-e-developers mailing list
> Q-e-developers at qe-forge.org
> http://qe-forge.org/mailman/listinfo/q-e-developers

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20160818/c06c8090/attachment.html>

More information about the developers mailing list