Dear Sir, I have recently sent my observations on the article mentioned below to the editor of Electronic Library. Subsequently, the author has sent the reply to the Editor which was forwarded to me by the editor. I hope that the discussion will enlighten us about the two OSS. The copy of the discussion has been pasted without any modifications. Regards, RS Giri Open Source Integrated Library Management Systems: Comparative Analysis of Koha and NewGenLib/ Manisha Singh and Gareema Sanaman, A comment Rabishankar Giri * * Indira Gandhi Institute of Technology, GGS Indraprastha University, Kashmere Gate, Delhi -110006, India. E-mail: rsgiri11@gmail.com Singh and Sanaman’s recent article (2012) on the comparative evaluation of two Open Source ILMS viz. Koha and NewGenLib makes space for debate. Response: There is no need for debate as the article was written in 2010 and was accepted in 2011. Mr. Giri has published his latest studies in the time period of 2011-2012. So as the technology changes every day, new improvised versions of software are developed and implemented. There may be slight difference in updated and lower versions. The two open source software have been compared previously. A white paper by Haravu (2009) compares both these software, but have been missed by the authors. In addition to this prominent work, there are a number of many other papers on NewGenLib which also if reviewed would have given a different perspective. Response: our article on comparative evaluation of two Open Source ILMS viz. Koha and NewGenLib includes more additional features, characteristics and provides specific information about both the software. And it is not possible to review entire literature and include in the paper while writing the study. Major important documents and relevant articles were reviewed by both the authors before writing review of literature and preparation of checklist. Some errors, perhaps due to oversight, that have crept in need to be corrected. On page. 813, the author has mentioned that NewGenLib was developed in 2007, but the fact is that NewGenLib version 1 is released in 2005 and on Jan. 9, 2008, it was declared as OSS under GNU GPL 3 (Wikepedia). Response: I would like to mention that Wikipedia is not an authentic source of information. NewGenlib is the product of a 4-year collaboration between the Kesavan Institute of Information and Knowledge Management (KIIKM) and Verus Solutions Pvt. Limited (VSPL), both based in Hyderabad, India. Toward the end of 2007 a business decision was taken to release NewGenLib under the GNU General Public License (GPL). “NewGenLib was developed in 2007” mentioned in the article is used to inform about its development as open source software for the first time. So, if needed the information about the release of first version of software (2008) can be added for clarifying the doubts. In the methodology (Sec. 5, pg. 814) , views/ opinions of professionals of two libraries viz. Delhi Public Library (DPL) using Koha, and Central Library, University of Delhi (CLUD) using NewGenLib have been taken into account. As per our best of knowledge, Central Library, University of Delhi is using Troodon, a proprietary software for Library Management, for more than one year (http://dulsopac.du.ac.in/) and DPL is using KOHA for more than three years. It may be that CLUD did some experiment with NewGenLib for sometime but recent interaction with the professionals from DU Library reveals that they did not experiment with NewGenLib in recent times. Even if we assume that Central Library, University of Delhi done a pilot study on NewGenLib, scaling opinion in Likert’s system will differ substantially because DPL is using the system for at least three years and its professionals got habituated with the process and naturally everything in their software seems to be easierwhereas professionals attached with initial pilot phase of experimentation of NewGenLib in CLUD will feel less habituated. Thus, it attracts a debate in the use of five point scaling system while giving weightage to the opinions where two systems are not at par. It is also to be noted that even DPL is not using all modules of Koha regularly, except acquisition, cataloguing, Circulation, Administration. Thus, the evaluation process based on the two sites, i.e., the Delhi Public Library that has not explored all features and modules of Koha and Delhi University Library that used NewGenLib for testing purpose only. Response: I want to inform the person that article is not only written on the basis of professional personal views and opinions. Article has been written after the thorough study of the software through important sources such as software manuals, relevant articles and software websites. I would also like to add to the knowledge that while I was conducting my study I have personally visited Central Library, DU where the work over library data was under progress using NewGenLib. Later if it has been discontinued, that’s not my responsibility to give reason behind the personal issues of the library. And again I would like to add more for DPL using Koha as I already had mentioned that not only one specific source has been used for writing the article. Visiting DPL and also CRL was the part of the study for gathering more and more information about the study. I have not used rating scale to compare the libraries using the software. It is used as an evaluation tool for the individual software at its own basis irrespective of the libraries. In the software evaluation section (Sec. 4, pg. 814), the authors say, “reference tools which compare the software features and functionality as a whole; reference statistical data available; verification through checklists which is a comprehensive list of questions or list of items to be noted, checked, or remembered and enables to clarify and standardize the evaluation and comparison process and directs the evaluator to the right things.” And “The software packages were evaluated and compared using the checklist which was prepared after thorough study and review of related literature, software reference tools and reference statistical data available” The author has neither referred to the standard reference tools, reference statistical data and checklists available nor provided comprehensive list of questionnaires. Response: I have not used the word Questionnaire anywhere in my study. And Checklist has been prepared personally by the efforts of both the authors after work hand and studies from the software manuals, documentation available, consulting sources and software websites. Checklist has not been copied from anywhere and it is presented in the form of the tables given in our article. And the entire points used in the checklist can be cross-checked while visiting to websites of the software used and their documentation and also by reading the articles used in the review. Again on same page the authors said, “On the basis of qualitative and quantitative results obtained from various sources and their satisfaction in meeting the requirements through the software, all the selected major and minor components or elements were rated for the comparison and various items were signed for their availability (√) and non-availability (×) in the software.” But authors did not mentioned the various sources from where they have derived the quantitative and qualitative results. If, it is from CUDL and DPL, CUDL is not using NewGenLib regularly, so there is hardly any question of satisfaction level of professionals at different levels of implementation and the other one DPL has partially exploited the Koha. Response: I have already mentioned in above responses related to various sources and have clearly written that data from Central Library, Delhi University and DPL, was only one of the source to know more about the software. On page 815, the authors mentioned “Further Koha is more compatible with international metadata and interoperability standards in comparison to NewGenLib which supports very few of them which are later discussed in the paper under formats and standards implementation.” Although the author’s mention that standards implementation are later discussed in the paper, it is found that they make general statements, e.g., in Table 1, under interoperability they give a score of 4 and 3 to Koha and Newgenlib respectively without justifying and/or specifying what interoperability standards are being supported or not supported. But, it is found that NewGenLib supports major international standards for better interoperability (Giri, 2012, 2011) and also compliant to Zotero which has been overlooked by the authors. Response: Kindly Read the article carefully, as the Table VI clearly explains the standards supported by the software. “Koha is more user friendly in terms of installation whereas NewGenLib.” This finding is in stark contrast with the findings reported by Giri from different workshop/training participants (Giri, 2012). It is to be noted that most Koha experts advises its installation on linux platform which is comparatively difficult than Windows platform. Whereas, most of NewgenLib installation is on Windows platform as found from the NewGenLib forum discussion. Response: Findings may vary from personal experience and individual point of view. Similarly, under Granularity, they provide scores to the two software without spelling out what they mean by granularity. In the absence of specific criteria on which the scores for granularity are based, score card is not convincing. If granularity means breaking main module into different smaller parts/submodules for user convenience, our three years experience of using NewGenLib tells that it takes into account very minute details as that of Koha. In several areas, like Budget Management, NewGenLib stands ahead of Koha as it has a well defined budget management system including carry forwarding unspent budget of previous financial year or in patron management, upgrading patron from one course to another by forcing change in patron id. If granularity means as author mentioned “Koha has granularity which describes various types of access permissions given to selected people who use the site and help an administrator to maintain control over who is allowed to edit and/or publish on a site whereas it is missing in NewGenLib software”, then the author has put less attention in NewGenLib and the statement is grossly inaccurate. NewGenLib has different level of permission in very minute level, who can use any module or sub module/specific function in module through Patron category's privileged matrices. Response: I have thoroughly studied the concept and have mentioned specifically about granularity. There is possibility of difference in opinion because of work hand over different versions. Since the article was written in 2010 and with sources available at that time. On page 815, the authors have stated that ”Koha is highly flexible as consultations and library can be used at terminals without any hard disk or specialized hardware which is not possible in NewGenLib.” This statement is inaccurate as NewGenLib can also be used with dumb terminals like Koha. In Page no. 816/817, under the Table II, the authors have pointed out that Koha is using Zebra search engine and mentioned no search engine for NewGenLib. Also it has been mentioned in 6.4 database features “NewGenLib database is searchable through CQL using bath and Dublin core profiles where in Koha has sophisticated zebra search engine”. The authors has provided very much inaccurate information and it is to be noted that NewGenLib is using Lucene / Solr for indexing/searching (Giri, 2011, 2012). Zebra is open source Index data software which has hardly any community around it whereas Lucene Solr is most active open source indexing software project from Apache. Many other successful open source projects like Vufind is using Lucene Solr project. Koha community is also shifting to Lucene/Solr (http://wiki.koha-community.org/wiki/Switch_to_Solr_RFC, http://lists.katipo.co.nz/pipermail/koha/2010-November/026306.html), Biblibre(one of Koha Vendor) has argued for using Lucene/Solr for Koha. (http://drupal.biblibre.com/en/blog/entry/solr-developments-for-koha). Response: I agree with the person with statement made above about the search engine. But I would like to clear that I have taken Zebra Search Engine as a feature of ILMS. It can be edited and corrected by stating Search Engine as a general feature in Table II and than giving the name of both search engines used independently through explanation. S.No. General Feature Koha NewGenLib 10 Search Engine ✔ ✔ Total Features 23 27 Explanation: There are various features common in both Koha and NewGenLib…………………………….Whereas in NewGenLib there is non-availability of Z39.50 client and server for data interchange; Z39.50 Copy Cataloguing, Engine, RSS feed for new acquisition/ search updates, LDAP Authentication and vice-versa are in Koha. Search Engine used in both software are different as Koha uses Zebra as search engine whereas NewGenlib uses Lucene and Solr (NewGenLib 2012 c) soon Koha will also have Solr (ACKU 2012). Reference can be added in online version as: ACKU, Koha online catalogue, ACKU, viewed 1 December 2012, http://178.79.186.94/. NewGenLib 2012c, Technology used, NewGenlib, viewed 1 December 2012, < http://www.slideshare.net/newgenlib/newgenlib-30-presentation>. Similarly, in the same table, under database backup, Koha is marked as ‘Easy’ and NewGenLib as ‘More easy’. Such general statements without substantiation gives biased views. Response: Explanation is clearly given below the table. Statement to be edited for database searchable feature: NewGenLib database is searchable through CQL using bath and Dublin core profiles and with Lucene and Solr (NewGenLib 2012 c) whereas Koha has sophisticated zebra search engine. Similarly, the author has mentioned there is no provision for binding management in Koha which is very much incorrect. Response: This is part of Circulation module. Please kindly read more about it as Information given about binding management is correct. Haravu, L J. (2009), “Comparison of two open source integrated Library Systems (ILS):Koha (version. 3.0) and NewGenLib (version. 2.2 beta)” , Available at www.verussolutions.biz/files/Whitepaper2.doc (accessed 20 August 2012) Under Toolkit, against NewGenLib there is J2SDK. This is not a toolkit but an application development environment. Version 3.0.4 of NewGenLib uses a few open source kits like JMARC, Jasper (for report generation). Response: Article was written over Version 3.0 in 2010 and toolkit here refers to the environment to run the program, which is correctly mentioned and is of true knowledge. In the active developments status, the author has given Koha score 5 for newgenlib 1 , but in General features of ILMS author pointed “Koha is also less frequently upgraded and is lacking in modifying, improving and enhancing its features quality from technology view point”. The statements and points given are contradictory and it is to be noted that new features of Koha is released every six months and every month they release bugfix also (Koha Wiki). In case of NewGenlib they release major updates with bugfixes in every three months (Giri, 2012). So, the score card is confusing. However, it is convincing that Koha has wider user as well as developer base across the globe and thus reporting bugs as well as bugfixes are expected to be quick in comparison to NewGenlib where Verus Solution is the sole developer. Response: There is difference in the Characteristics and General features of both software. The person has misunderstood. In Table I Koha have active development status with frequent upgrades which means that Koha is developing its features and characteristics within its old technology and Table II it is defined in term of latest technology development. On page 818 of the paper, the author’s write, “NewGenLib user interface supports few international and local Indian languages.” . But, it has also mentioned that Koha (40 languages) have multilingual support through I18N/L10N whereas NewGenLib (51 languages) through only I18N - Look and feel standard. A correct statement would be that NewGenLib supports all Unicode 4.0 complaint languages. NewGenLib also allows the use of Google transliterate functionality to enable data entry in any Unicode language in Roman script. Corrections can then be made to the entered transliterated text. Response: As NewGenLib is developed in India, so it supports few Indian Languages. And the article is written with objectivity. And Statement made “Koha (40 languages) have multilingual support through I18N/L10N whereas NewGenLib (51 languages) through only I18N - Look and feel standard” is in its correct form. On page 819 of the paper, the authors stated that “Koha, data backup created cannot be used platform independently i.e. backup taken on windows cannot be used or run on Linux and vice versa”. This is an incorrect statement because if backup is taken as .sql file how it will depend on operating system. Response: The information provided in the Article under Table IV for data backup is correct. There are three main components of Koha that must be considered when backing up and restoring: the database, the Koha file system, and the operating system customizations for the system. Refer: http://www.kohadocs.org/Backups and Restoring.html On page 820, under Acquisition Modules function, the sub-functions including background functions (e.g., Acquisitions administration) are given scores without defining what the parameter means. Similarly, on page 821, Table 6, the article uses scores for functions without defining these sufficiently clearly. The author has also mentoned that in Koha “Printing of Catalogue module in AACR2 format is not possible”. This statement is also not the correct one . Response: It is very clear from the explanation given below Table V for Acquisition. And the analysis was done thoroughly on the basis of functionality of software before giving them scores. In Koha “Printing of Catalogue module in AACR2 format is not possible is correct. Kindly read Manual of Koha for more clarification. On page 822, under Circulation functions, the authors depart from the method used to evaluate the two software for Acquistions function, which was to assign a score of 1 to 5 for each sub-function. In evaluating the Circulation function, the authors, use just Yes or No marks. They revert back to the rating of sub-functions in the Serial module (Table 8). Once again the author’s assign scores (1 to 5) for the OPAC/WebOPAC functionality (Table 9, page 824). In other words, the paper has not been consistent in evaluating the functional modules. Response: It is already mentioned that various methods have been used to evaluate the different functionalities as per the availability of Data. Conclusion The paper would have been useful if it had compared the view/opinions of professionals of libraries using most of the functionalities of the two open ILMS for at least a while. Reference Giri, Rabishankar and Sengar, DS. (2011), “Use of open source software in the learning resource centre of Indira Gandhi Institute of Technology: a case study”, Annals of Library and Information Studies, Vol. 58 No.1, pp. 41-48., available at: http://nopr.niscair.res.in/handle/123456789/11555 (accessed 28 November 2011). Haravu, L J. (2009), “Comparison of two open source integrated Library Systems (ILS):Koha (version. 3.0) and NewGenLib (version. 2.2 beta)” , Available at www.verussolutions.biz/files/Whitepaper2.doc (accessed 20 August 2012) Giri,Rabishankar(2012),"NewGenLib 3: An Integrated Open Source Library Management System that Makes your Library Visible in Web", Library Hi Tech News, Vol. 29 Iss: 10 (Date online 15/10/2012) Downloaded on: 16-11-2012 -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.