[OSGeo-Discuss] [Geo4All] Draft of Open Letter on the importance to protecting independent peer review frameworks for Scholarly publications of Scientific Associations

Peter Baumann p.baumann at jacobs-university.de
Thu Aug 2 01:49:13 PDT 2018


Dear Maria,

On 28.07.2018 20:08, María Arias de Reyna wrote:
> Dear Peter,
>
> Thanks for the explanation, I'm a software engineer and I know very well how
> the process work. SQL is indeed a good example of a good standard almost
> everyone follows. But I would bet that's an exception in software.

I cannot completely disagree, unfortunately - much software is not built to such
rigorous processes, and in particular not with an exact specification. During my
work in the ISO SQL group (where we brought in datacubes) I have learnt to
deeply respect the rigor of work there.

>
> I raised the intended bugs thing because in recent years it had been very
> common to find intended "bugs" in proprietary licensed software to steal data
> or open ports to monitor usage. You never know how that will affect your
> tests. When you use software that you don't know what is doing in the
> background, you can't be really sure what the result will be.

hm, this still is unspecific allegations. Two thoughts:
- can you name concrete cases?
- IMHO we are talking of completely unrelated things here. I understand that
some people feel uneasy about a monitoring (although this is very helpful also
for open-source projects where you are in need of statistics to claim your
importance - we refrained from it after internal discussion). What you describe
would not affect accuracy / correctness of tool results in any way, though.

cheers,
Peter


>
> Kind regards, 
> Maria. 
>
>
> El sáb., 28 jul. 2018 19:48, Peter Baumann <p.baumann at jacobs-university.de
> <mailto:p.baumann at jacobs-university.de>> escribió:
>
>     Dear Maria,
>
>     On 25.07.2018 10:06, María Arias de Reyna wrote:
>>
>>
>>     On Wed, Jul 25, 2018 at 9:57 AM, Peter Baumann
>>     <p.baumann at jacobs-university.de <mailto:p.baumann at jacobs-university.de>>
>>     wrote:
>>
>>         Hi Christian,
>>
>>
>>         while I could not agree more to what you say there is one point to
>>         disagree with:
>>
>>
>>         On 24.07.2018 18:43, Christian Willmes wrote:
>>>
>>>         Dear Suchith,
>>>
>>>         I understand your point, and I also support your views on this, but
>>>         this is from my perspective a too personal/particular issue, as to
>>>         have it as an "OSGeo open letter". Also, because this is more of an
>>>         ICA and not so much an OSGeo issue, I think.
>>>
>>>         First, I would keep it more general. You address a particular issue
>>>         (UN SDG book published by esri), and also some personal background
>>>         (this should not matter to the addressed subject). I would recommend
>>>         you keep it from being personal and denouncing proprietary GIS
>>>         vendors. If a company plays by the rules of science, there is
>>>         nothing wrong about that company publishing a scientific book. I.e.
>>>         almost all book publishers are commercial companies with interests
>>>         somehow and somewhere.
>>>
>>>         You need to “attack” scientific “wrong doing” by that particular
>>>         company in conducting the editing and publication of that book.
>>>         Publishing books if done correctly is not wrong, even by a vendor
>>>         with vested interests. But if you witness, for example, that
>>>         submissions using open source GIS solutions are disadvantaged
>>>         against the submissions using products of the proprietary GIS vendor
>>>         publishing the book, that would be the point to raise and attack.
>>>
>>>         Second, better write about how it should be done to avoid this
>>>         negative “Fake Science” things from happening. Here the idea of Open
>>>         Science and Reproducible Science is key, i.e. the most openness and
>>>         transparency possible. We just need more transparency in science and
>>>         also in the whole process of editing/reviewing and publishing a
>>>         book. And this is where OSGeo can contribute. Basically, real
>>>         reproducible and open science is not possible without open source
>>>         software. If you can’t see how something is implemented, you can not
>>>         really reproduce the results.
>>>
>>
>>         No. Open science and open source software are fundamentally different
>>         things. For example, if you derive stats from some data set via SQL
>>         it does not matter whether it comes from open-source PostgreSQL or
>>         from proprietary Oracle. Because the SQL language in its syntax and
>>         semantics is standardized, and it is assured thereby that both
>>         systems will deliver the same results. So standards actually are a
>>         prerequisite for science to be comparable, but surely not open source.
>>
>>
>>
>>     If you use proprietary products and can't verify that the result is not
>>     due to a bug (even an intended bug ), you are missing an important step
>>     on verifiability. Open Source (as in "I can see the code") is an
>>     important piece of open science.
>
>     that's not what software engineers would do normally. If you feel a tool
>     has a bug you'd
>     - try to isolate through a minimal failing example
>     - possibly try with another tool (in the case of PostgreSQL, maybe try
>     MariaDB) for verification
>     - definitely contact the support list (in the case of PostgreSQL, Regina &
>     friends)
>
>     Unless it is some simple scripting issue you (that is: I) normally don't
>     stand a chance to dive into the code. Honestly, would we / could we spot a
>     bug in the source code for executing an index-only plan of a distributed
>     SQL query, after heuristic and cost-based optimizers have done their work?
>     I could not.
>
>     Good software engineering practice is to work specification-based, not by
>     trying to hack yourself into code.
>
>     And both of that _can_ work well with both open-source and proprietary
>     tools. Again, SQL is the shining example: a good specification says it all.
>
>     BTW, why do you raise, on the fly, the accusation that there may be
>     "intended bugs"? Any evidence for this? I'd like to learn more about such
>     cases.
>
>     cheers,
>     Peter
>
>
>
>
>
>>
>>
>>     _______________________________________________
>>     GeoForAll mailing list
>>     GeoForAll at lists.osgeo.org <mailto:GeoForAll at lists.osgeo.org>
>>     https://lists.osgeo.org/mailman/listinfo/geoforall
>
>     -- 
>     Dr. Peter Baumann
>      - Professor of Computer Science, Jacobs University Bremen
>        www.faculty.jacobs-university.de/pbaumann
>     <http://www.faculty.jacobs-university.de/pbaumann>
>        mail: p.baumann at jacobs-university.de <mailto:p.baumann at jacobs-university.de>
>        tel: +49-421-200-3178, fax: +49-421-200-493178
>      - Executive Director, rasdaman GmbH Bremen (HRB 26793)
>        www.rasdaman.com <http://www.rasdaman.com>, mail: baumann at rasdaman.com <mailto:baumann at rasdaman.com>
>        tel: 0800-rasdaman, fax: 0800-rasdafax, mobile: +49-173-5837882
>     "Si forte in alienas manus oberraverit hec peregrina epistola incertis ventis dimissa, sed Deo commendata, precamur ut ei reddatur cui soli destinata, nec preripiat quisquam non sibi parata." (mail disclaimer, AD 1083)
>
>

-- 
Dr. Peter Baumann
 - Professor of Computer Science, Jacobs University Bremen
   www.faculty.jacobs-university.de/pbaumann
   mail: p.baumann at jacobs-university.de
   tel: +49-421-200-3178, fax: +49-421-200-493178
 - Executive Director, rasdaman GmbH Bremen (HRB 26793)
   www.rasdaman.com, mail: baumann at rasdaman.com
   tel: 0800-rasdaman, fax: 0800-rasdafax, mobile: +49-173-5837882
"Si forte in alienas manus oberraverit hec peregrina epistola incertis ventis dimissa, sed Deo commendata, precamur ut ei reddatur cui soli destinata, nec preripiat quisquam non sibi parata." (mail disclaimer, AD 1083)


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/discuss/attachments/20180802/f97c9502/attachment.html>


More information about the Discuss mailing list