Hi Margareth, sorry for the late reply. CC'ing DSpace Technical Support list as discussed, so others can offer their ideas / learn from this.
There are a few ways the bitstream count can get up, from my experience... There are some basic reasons like * The bitstream URL has been indexed directly in a search engine like Google, and people are visiting the PDF (for example) straight from there * The item has multiple bitstreams, and a visitor could be downloading each of them from the item page without refreshing the item page and generating a new 'hit' * The item has one bitstream, but a visitor is downloading it multiple times from the item page -- sometimes people will do this accidentally, or a proxy server will cause a re-download, or a spider will get stuck and start downloading things too many times, or a visitor is viewing the PDF in the browser, and hits 'refresh', etc... And then there's one other possible tip that I discovered while i was investigating a similar issue -- if you have Apache or nginx (or similar) webserver in front of your DSpace instance, and a visitor's browser or PDF viewer "streams" the PDF instead of doing a straight download (this is particularly true of browsers with PDF viewer plugins that will display the retrieved document inline) then it can end up looking like lots and lots of GET requests (with 206 responses instead of 200). I don't *think* this results in additional DSpace statistics 'hits', but it certainly can show up in some basic HTTP log statistics when looking at downloads, so I thought it was worth mentioning just in case. Hope this helps! Anyone else had similar situations where they could track down the explanation? Cheers Kim M: k...@shepherd.nz T: @kimshepherd P: +6421883635 W: www.shepherd.nz <http://shepherd.nz> 0CCB D957 0C35 F5C1 497E CDCF FC4B ABA3 2A1A FAEC On 22 March 2018 at 08:32, <margareth.ca...@bndes.gov.br> wrote: > Classificação: Documento Controlado > Restrição de Acesso: Empresas do Sistema BNDES – Uso no âmbito interno > Unidade Gestora: AARH/DEFAC/COPED (classificação conforme OS PRESI 01/2015) > > > Dear Kim Shepherd, > > I hope this finds you well. > > I am a librarian at the Brazilian Development Bank and we use DSpace to > support our institutional repository. Actually, we are updating from 1.6 to > 6.0 version. > > As we want to explore more the searches at our digital library, we would > like to understand more about DSpace Statistics. > > So, we need to elucidate a few points: > > 1. Why the number of Bitstream views is bigger than Items views? This > means that researchers can achieve our bitstreams through search tools and > harvesting? > > 2. Why the number of items view is different when we check general > statistcs and compare it with items statistics (please see the prints > below)? > > > > > 3. Bitstream view means downloads? > > Thanks in advance for all your cooperation with this matter. > > If there's another person that I should talk with, please let me know. > > Best regards, > > > > > *Margareth Ramos do Carmo Freitas - Bibliotecária - CRB/7-5127* > > Centro de Pesquisa de Informações e Dados - COPED - Coordenação > Área de Administração e Recursos Humanos > Av. República do Chile, 100 - 1º andar / E-08 > Rio de Janeiro - RJ - Brasil - CEP 20031-917 > Telefone: +55 21 2052-7240 <+55%2021%202052-7240> > *www.bndes.gov.br* <http://www.bndes.gov.br> > *www.bndes.gov.br/bibliotecadigital* > <http://www.bndes.gov.br/bibliotecadigital> > *Acesse o Portal de Informação e Conhecimento e siga nossa comunidade* > <https://colabore.bndes.net/communities/service/html/communityview?communityUuid=2446128d-0439-4497-83df-55e739136f80> > > > -- You received this message because you are subscribed to the Google Groups "DSpace Technical Support" group. To unsubscribe from this group and stop receiving emails from it, send an email to dspace-tech+unsubscr...@googlegroups.com. To post to this group, send email to dspace-tech@googlegroups.com. Visit this group at https://groups.google.com/group/dspace-tech. For more options, visit https://groups.google.com/d/optout.
