From marie-emilia.herbet at univ-lyon3.fr Tue Apr 4 11:35:41 2023 From: marie-emilia.herbet at univ-lyon3.fr (HERBET Marie-Emilia) Date: Tue, 4 Apr 2023 08:35:41 +0000 Subject: [Cod-bugs] Questions about COD In-Reply-To: <61f37242-f94a-18af-d906-bb92458f2f07@ibt.lt> References: <0051dbcd7fd940579d354680fe98ff41@univ-lyon3.fr> <61f37242-f94a-18af-d906-bb92458f2f07@ibt.lt> Message-ID: <6f8be2f8bb9b42f0a524a7faea034bcd@univ-lyon3.fr> Dear Saulius, Thank you for all the information provided below. I get back to you with another question regarding COD. In the framework of our project aiming at building up a data repository selection tool for reasearchers, I would like to know if COD ID has the same features and role as a DOI. In other terms, does it act a permanent identifier that can be cited and accessed even if the URL changes? Regarding my previous question related to licenses, the idea was not to promote copyright but rather to ensure that structures can be freely reused and cited using for instance CCBY license instead of depending on the license of the paper in which the structure is described (which often entails copyright rules applied by publishers.) Thank you in advance, Best regard, Marie De : Saulius Gra?ulis Envoy? : mardi 31 janvier 2023 18:31 ? : HERBET Marie-Emilia Cc : cod-bugs at lists.crystallography.net Objet : Re: [Cod-bugs] Questions about COD Dear Marie-Emilia, thank your for your inquiry about the COD! Below I will attempt to answer your questions. On 2023-01-31 18:08, HERBET Marie-Emilia wrote: Could you please tell me what is COD?s main supervising institution? Currently the hosting institution is Vilnius University, Life Sciences Center, Institute of Biotechnology. The COD scientific contents is supervised by the international COD Advisory Board (COD AB) whose members are listed on the COD Web site. I saw that the repository received donations from several organisations but I don?t quite understand what is its founding institution. The COD is founded by scientists who constituted its first Advisory Board, as a "grass-root" movement. It was hosted at the Le Mans University at that time. It is funded by various means ? research grants to it's development and applications, donations by companies that use it, donations of individuals and support of research organisations such as universities. The founders of the COD are physical persons, however ? scientists who overview and invest their time into the COD development. It is regarded that for the project of this type, no "founding institution" can be identified, or is actually necessary. Second point, where is the data hosting service based The data are currently hosted in the Vilnius University Data Center, supported by the Vilnius University. and what is the long-term preservation policy of the repository? The policy is to preserve the COD data for as long as it is need for the community and valuable for science. The projected life span is decades, depending on the developments of the crystallography. All COD history is recorded in the Subversion repository, allowing to trace all changes to the database that were done over the time, ensuring full data provenance. Finally, I would like to know why COD opted for CC0 license instead of CCBY? The Advisory Board and the founders of the database (Armel Le Bail, Daniel Chateigner) was strongly insisting that crystallographic data should be as open as possible, thus putting their collections of data under the "public domain" designation. Since "public domain" is not a legal license in Europe, it was eventually voted by the COD AB that CC0 should be adopted. The public domain license might deter some researchers from depositing their structures as this license does not require to cite the authors when reusing their structures. I think there is some misunderstanding here. Facts of nature, as we all know, are not copyrightable (i.e. everyone has an unrestricted right to post a statement that "NaCl unit cell length a is 5.640 ?" and similar facts, encoded in whatever form they like). The license therefore does not apply, and can not apply, to the information in the individual records. Only the database as a whole is copyrighted, and licensed under CC0 by its creators under sui generis "sweat of the brow" doctrine of the current copyright law. Citing the individual records, or the database itself when they are used in research is an accepted good practice of scientific scholarship. It is therefore independent, in my view, from the license under which the database is distributed. When researchers use individual papers or the database as a whole, they cite the relevant papers to provide information for their readers and to give credit to information sources. This alone should ensure that the antecedent work is properly cited. Hope this helps, Saulius -- Dr. Saulius Gra?ulis Vilnius University Institute of Biotechnology, Saul?tekio al. 7 LT-10257 Vilnius, Lietuva (Lithuania) fax: (+370-5)-2234367 / phone (office): (+370-5)-2234353 mobile: (+370-684)-49802, (+370-614)-36366 -------------- next part -------------- An HTML attachment was scrubbed... URL: From grazulis at ibt.lt Wed Apr 5 09:25:14 2023 From: grazulis at ibt.lt (=?UTF-8?Q?Saulius_Gra=c5=beulis?=) Date: Wed, 5 Apr 2023 09:25:14 +0300 Subject: [Cod-bugs] Questions about COD In-Reply-To: <6f8be2f8bb9b42f0a524a7faea034bcd@univ-lyon3.fr> References: <0051dbcd7fd940579d354680fe98ff41@univ-lyon3.fr> <61f37242-f94a-18af-d906-bb92458f2f07@ibt.lt> <6f8be2f8bb9b42f0a524a7faea034bcd@univ-lyon3.fr> Message-ID: <80d6a63a-61c9-1282-33fe-5a065e80a4ca@ibt.lt> On 2023-04-04 11:35, HERBET Marie-Emilia wrote: > > Thank you for all the information provided below. I get back to you > with another question regarding COD. In the framework of our project > aiming at building up a data repository selection tool for > reasearchers, I would like to know if COD ID has the same features and > role as a DOI. In other terms, does it act a permanent identifier that > can be cited and accessed even if the URL changes? > Yes, indeed, the COD IDs are stable permanent IDs. Even when the data are no longer available (e.g. the structure is retracted), the ID stays and the metadata are further available. The COD IDs function as any other permanent IDs required to identify a data record, e.g. Archival Resource Keys (ARK), Handle identifiers, URNs; and, yes, DOIs are in the same category (except that you need to pay money to get DOIs assigned :/ ). In addition to that, COD URLs (which are actually URIs) of the form 'https://www.crystallography.net/cod/2000000.html' or 'https://www.crystallography.net/cod/2000000.cif'are intended to remain stable into the future and can be used as COD structure identifiers. > Regarding my previous question related to licenses, the idea was not > to promote copyright but rather to ensure that structures can be > freely reused and cited using for instance CCBY license instead of > depending on the license of the paper in which the structure is > described (which often entails copyright rules applied by publishers.) > Well, this was discussed in the COD AB, and the current collegial decision is to put CC0 in the COD. If you feel very strong about the issue, you can submit a "COD Improvement Proposal" to the COD AB (or to the cod-bugs@ list), and I'll forward it to the COD AB and promise to organise an on-line discussion (you would be welcome to participate in it). We need to take into account that CC-BY is more restrictive for data users than CC0, so we need to look carefully as to not jeopardise their work flows. Most (all?) of our users, however, already now cite the original works since this is the custom in the academic community, so CC-BY might be not that disruptive after all. Sincerely yours, Saulius -- Dr. Saulius Gra?ulis Vilnius University, Life Science Center, Institute of Biotechnology Saul?tekio al. 7, LT-10257 Vilnius, Lietuva (Lithuania) phone (office): (+370-5)-2234353, mobile: (+370-684)-49802, (+370-614)-36366 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -------------- next part -------------- An HTML attachment was scrubbed... URL: From marie-emilia.herbet at univ-lyon3.fr Wed Apr 5 17:33:09 2023 From: marie-emilia.herbet at univ-lyon3.fr (HERBET Marie-Emilia) Date: Wed, 5 Apr 2023 14:33:09 +0000 Subject: [Cod-bugs] Questions about COD In-Reply-To: <80d6a63a-61c9-1282-33fe-5a065e80a4ca@ibt.lt> References: <0051dbcd7fd940579d354680fe98ff41@univ-lyon3.fr> <61f37242-f94a-18af-d906-bb92458f2f07@ibt.lt> <6f8be2f8bb9b42f0a524a7faea034bcd@univ-lyon3.fr> <80d6a63a-61c9-1282-33fe-5a065e80a4ca@ibt.lt> Message-ID: <702825244b22446289325384318c6893@univ-lyon3.fr> Dear Saulius, Thank you very much for all the information provided and the accuracy of your answers. On the license issue, I agree with you as researchers are supposed to cite properly other authors even if the CC-BY license is not specifically mentioned. Thank you as well for clarifying the properties of COD-ID, which is a sort of ?in-house? permanent identifier as far as I understand, different from ARK or DOI but having the same function. One of our criteria consists in checking the data quality management of repositories. In cristallography, it is common to have a two-step process: automated check of CIF files and manual check by experts. Can you confirm that COD still performs manual inspections as described in this paper: https://academic.oup.com/nar/article/40/D1/D420/2903497 Last but not least, I would like to know why some researchers might decide/prefer to deposit their structures in COD rather than CSD as both services are free and cover the same disciplinary field. I don?t know if COD has already carried out a survey amongst its users but it would be great if you could elaborate a little bit on COD approach vs CSD, whose financial model is quite different. Thank you again for your help and insight. Best regards, Marie Marie-Emilia HERBET ------------------------------------------------------ Responsable du d?partement d?appui ? la recherche et aux projets BIBLIOTH?QUES UNIVERSITAIRES UNIVERSIT? JEAN MOULIN LYON 3 ADRESSE G?OGRAPHIQUE > Manufacture des Tabacs | 6 cours Albert Thomas | LYON 8e ADRESSE POSTALE > Biblioth?que de la Manufacture | 1C avenue des Fr?res Lumi?re | CS 78242 - 69372 LYON CEDEX 08 Ligne directe : +33 (0)4 78 78 79 04 E-mail : marie-emilia.herbet at univ-lyon3.fr --------------------------------------------------- www.bu.univ-lyon3.fr | Suivez-nous > Facebook | Twitter| Instagram L'Universit? Jean Moulin est membre de l'Universit? de Lyon De : Saulius Gra?ulis Envoy? : mercredi 5 avril 2023 08:25 ? : HERBET Marie-Emilia Cc : cod-bugs at lists.crystallography.net Objet : Re: [Cod-bugs] Questions about COD On 2023-04-04 11:35, HERBET Marie-Emilia wrote: Thank you for all the information provided below. I get back to you with another question regarding COD. In the framework of our project aiming at building up a data repository selection tool for reasearchers, I would like to know if COD ID has the same features and role as a DOI. In other terms, does it act a permanent identifier that can be cited and accessed even if the URL changes? Yes, indeed, the COD IDs are stable permanent IDs. Even when the data are no longer available (e.g. the structure is retracted), the ID stays and the metadata are further available. The COD IDs function as any other permanent IDs required to identify a data record, e.g. Archival Resource Keys (ARK), Handle identifiers, URNs; and, yes, DOIs are in the same category (except that you need to pay money to get DOIs assigned :/ ). In addition to that, COD URLs (which are actually URIs) of the form 'https://www.crystallography.net/cod/2000000.html' or 'https://www.crystallography.net/cod/2000000.cif'are intended to remain stable into the future and can be used as COD structure identifiers. Regarding my previous question related to licenses, the idea was not to promote copyright but rather to ensure that structures can be freely reused and cited using for instance CCBY license instead of depending on the license of the paper in which the structure is described (which often entails copyright rules applied by publishers.) Well, this was discussed in the COD AB, and the current collegial decision is to put CC0 in the COD. If you feel very strong about the issue, you can submit a "COD Improvement Proposal" to the COD AB (or to the cod-bugs@ list), and I'll forward it to the COD AB and promise to organise an on-line discussion (you would be welcome to participate in it). We need to take into account that CC-BY is more restrictive for data users than CC0, so we need to look carefully as to not jeopardise their work flows. Most (all?) of our users, however, already now cite the original works since this is the custom in the academic community, so CC-BY might be not that disruptive after all. Sincerely yours, Saulius -- Dr. Saulius Gra?ulis Vilnius University, Life Science Center, Institute of Biotechnology Saul?tekio al. 7, LT-10257 Vilnius, Lietuva (Lithuania) phone (office): (+370-5)-2234353, mobile: (+370-684)-49802, (+370-614)-36366 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -------------- next part -------------- An HTML attachment was scrubbed... URL: From zhangchunbo at whu.edu.cn Fri Apr 7 05:38:29 2023 From: zhangchunbo at whu.edu.cn (bob) Date: Fri, 7 Apr 2023 10:38:29 +0800 Subject: [Cod-bugs] overlap problem for 563 crystals Message-ID: Dear COD team, I am writing to bring to your attention an issue with the COD ID (see codID-563). There seems to be an atomic overlap problem with 563 structures, specifically with the COD-1510278 as can be seen in the attached image. Could you kindly investigate this issue and let me know if it can be rectified? I would greatly appreciate your assistance in this matter. Thank you for your attention to this matter. Best regards, Bob Zhang -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: codID-563 Type: application/octet-stream Size: 4504 bytes Desc: not available URL: -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ?????-1.png Type: image/png Size: 344366 bytes Desc: not available URL: -------------- next part -------------- An HTML attachment was scrubbed... URL: From grazulis at ibt.lt Fri Apr 7 09:50:12 2023 From: grazulis at ibt.lt (=?UTF-8?Q?Saulius_Gra=c5=beulis?=) Date: Fri, 7 Apr 2023 09:50:12 +0300 Subject: [Cod-bugs] overlap problem for 563 crystals In-Reply-To: References: Message-ID: <40963ebd-6626-88cc-d2e2-98e64129b977@ibt.lt> Dear Bob, thank you for your report. On 2023-04-07 05:38, bob wrote: > I am writing to bring to your attention an issue with the COD ID (see > codID-563). There seems to be an atomic overlap problem with 563 > structures, specifically with the COD-1510278 as can be seen in the > attached image. > Could you kindly investigate this issue and let me know if it can be > rectified? I would greatly appreciate your assistance in this matter. The issue that you are reporting is not a bug but an intended description of the structures, at least in the COD 1510278. In this entry, the two atoms, Au and Pt, are on exactly the same site (xyz = 0.0 0.0 0.0), and have occupancies 0.5, meaning that this is 50:50 ratio Pt:Au alloy; each site in such crystal is either Au or Pt with 50% probability (thus the entry reports a /substitutional alloy /[1]/)/. Whether this is indeed a correct alloy structure needs to be consulted in the original paper (https://doi.org/10.1524/zpch.1985.144.144.223/html), but from the data perspective the description is correct. Other entries in the COD might have different reasons for the atoms to be too close ? substitutional disorder, incorrectly marked positional disorder, incorrectly solved or recorded structures, etc. When we find incorrect descriptions, we do correct them manually, but this is a very time-consuming work, so we cannot promise to fix all issues quickly, and we have to triage them by importance. Thus, it is ultimately your responsibility as a COD user to interpret COD structures as was intended by the authors (as in the case of COD 1510278), and to filter our those structures that are not suitable for your investigation for one reason or another. Sincerely yours, Saulius Refs.: [1] Wikipedia. Alloy; sect. "Mechanisms" (2023). URL: https://en.wikipedia.org/wiki/Alloy [accessed 2023-04-07T09:44+03:00] -- Dr. Saulius Gra?ulis Vilnius University, Life Science Center, Institute of Biotechnology Saul?tekio al. 7, LT-10257 Vilnius, Lietuva (Lithuania) phone (office): (+370-5)-2234353, mobile: (+370-684)-49802, (+370-614)-36366 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. -------------- next part -------------- An HTML attachment was scrubbed... URL: