Astrolabe Curating, Linking, and Computing Astronomy’s Dark Data P. Bryan Heidorn.pdf

该【Astrolabe Curating, Linking, and Computing Astronomy’s Dark Data P. Bryan Heidorn 】是由【dt83088549】上传分享，文档一共【12】页，该文档可以免费在线阅读，需要了解更多关于【Astrolabe Curating, Linking, and Computing Astronomy’s Dark Data P. Bryan Heidorn 】的内容，可以使用淘豆网的站内搜索功能，选择自己适合的文档，以下文字是截取该文章内的部分文字，如需要获得完整电子版，请下载此文档到您的设备，方便您编辑和打印。:..TheAstrophysicalJournalSupplementSeries,236:3(12pp),2018Mayhttps:///-4365/aab77e?:Curating,Linking,putingAstronomy’,,andJulieSteffen21UniversityofArizonaSchoolofInformation,HarvillBuilding4thFloor,1103East2ndStreet,Tucson,AZ85721,******@,******@,950NCherryAve,Tucson,AZ85719,USA;julie.******@Received2017October19;revised2018February25;accepted2018February26;published2018May11AbstractWhereappropriaterepositoriesarenotavailabletosupportallrelevantastronomicaldataproducts,datacanfallintodarkness:,,munitymendationsforlocatingandensuringthelong-termcurationofdarkorotherwiseat-,muchastronomicaldarkdatacurrentlyexistthatarenotcuratedelsewhere,,,theprojectaimsesstoresearchdatabyleveragingexistingcyberinfrastructureresourcesandpromotingmunity,:(seeSection5),:whileastronomers100yearsagowouldbeEarth-basedinstrumentsincludeplanningfordatamanage-intimatelyfamiliarwithatelescopeitself,spendingmanyhoursment,andthisef?cientproject-,modernastronomersrelyHowever,scientistsroutinely?ndnewwaystoanalyze,merge,heavilyondataoutput,binationsoftelescopeandmanipulatedata,creatingnewdatasetsthatdonotmeettheandarchivaldata(McCray2004,2014).?elds,patibility,,,individuals,esunusablebecauseofandtheNSF-–2020pointsoutthatanNSF--disciplinarydatasharingandAsasolution,thedesignoftheAstrolabesystemsystemispreservation(NationalAcademies2016).Whereappropriatebasedoninputfromthisgroupofexperts,aimedatdevelopingrepositoriesarenotavailabletosupportallrelevantastronom-newresourcesformanagementofpreviouslyuncuratedicaldataproducts,datacanfallintodarkness:unseenandastronomicaldata,withanemphasisonsmall,old,,butnewerdatasetsarealsorecognizedthatsomenewerandlargerdatasetsarealsooftenuncurated()andcouldremain“dark”uncuratedorlacklong-(Heidorn2008).describedinthesectionsbelow,workshopparticipantsprovideThispaperprovidesadescriptionofAstrolabe,acyberin-insightintothedataandcyberinfrastructureneedsofthemunity,includingdatacollection,analysis,mendationsforlocatingandensuringthelong--riskdataandintegratedonexistingastronomydatasystems,aswellasborrowingcomputing,putationsystemsbuiltforother(OpenScienceCollaboration2012).Accordingtoparticipantssciences,eleratingnewinaseriesofworkshopsincludingastronomers,cyberinfras-sciencethroughimprovedcurationofbotholderandnewerputationalscientists,,wedecidedtobuild1:..TheAstrophysicalJournalSupplementSeries,236:3(12pp),2018MayHeidorn,Stahlman,&SteffenAstrolabeontheCyVersecyberinfrastructure,3whichwasmunity,(AAS);throughthispartnership,-valuedatasets(munity),theprojectaimsesstoresearchdatabyleveragingexistingcyberinfrastructureresources,andtopromotescienti?cdis-coverybymakingpotentiallyusefuldatabroadlyavailabletomunity,:(1)%ofprojectsmanagementrequirements,identi?edbyexpertparticipantsinareceivedbetween350,000,000toabout830,%ofseriesofworkshops,essfulcurationofdarkprojectsreceivedbetweenabout830,000,000to5,-datainastronomy;and(2)todescribethedevelopmentoffundedprojectsrequireasophisticateddatamanagementplan,meaningtheLongTailofresearchfundinglikelytranslatestoaLongTailofdarkAstrolabeinresponsetotheseidenti?(uncurated),followedbyadescriptionoftheAstrolabewell-,plans,,Section8LongTail—.,thelowerannualfundingpartofthisconcludesthepaperwithasummaryanddirectionsfordistribution—;curationofalldata,includingthosefromsmallerprojects,-,muchofthedatafromsmallKeepingtrackofdarkdatainastronomyisimportantforprojectsendupwithoutlong-?,??ciently,scientistswillLongTaileconomicstheory(Anderson2004)notesthatnotbeableto?ndandusethedata,renderingthosedataoodsarefrequentlybasedonasmallnumberofessentially“dark.”AccordingtoHeidorn(2008),darkdatahigh--and-mortarrequiresuniquecurationstrategiestocaptureandmanagedatastores,thegoalistokeephigh-,aslow-volumeitemstakeupspaceandLongTailaretypically:heterogeneous;hand-generated;-sellingpopularmoviesstockedonBlockbus-createdthroughuniqueprocedures;curatedbyindividuals;tershelvesareexamplesofhigh-volumeitems,whilethemanyoftenarchivedinpersonalorinstitutionalrepositories;notmoreless-popular?lmsanddocumentariesarelow-volumemaintained;obscuredorprotected;seldomreused;anditems,?-volumeitemsthroughthelostovertime,inaprocessreferredtointhe?eldofecologyas,withoutincurringthecostsofphysicalshelfspaceina“datadecay”().Withoutinitialgoodbrick-and-,-curatedareaundertherightpartofthecurve—thelow-volumeitems—,smallprojectsizeandtheassociatedlackofrepresentsigni?,aswellastime(legacy),LongTailthinkinginsciencewasproposedby(),whoperformedanexplorationofNSF-fundedresearchtheworkshopspresentedherelargelyagreedwiththisgrantsacrossdisciplines,showingthatthelargest20%%ofthetotalHeidorn(2008),anotionthatledtoandguidedoftheNSFprogramsin2007,Figure1showsthedistributionthetwoworkshopsheldin2015and2016().Inthedisciplineofastronomy,(2015)explainsthatdatainNSFin2007,(2008)notedthatthistop20%,manyastronomersimplementsmallresearchprojects,oftenwith3ializedinstrumentsfordatacollection,orrelyonderivative4Alogtransformofthisgraphwouldyieldanearlystraightline,andthusand/:..TheAstrophysicalJournalSupplementSeries,236:3(12pp),2018MayHeidorn,Stahlman,&Steffenavailable,,lackofdatamanagementskills,andbecausethedemonstrated,forexample,byWhiteetal.(2009),wheretheutilityofdataforotherresearchersmaynotoutweighthenumberofpapersbasedonarchivaldatafromtheHubbleessSpaceTelescopeexceededthenumberofnon-archival(;Borgman2015).“bigdata”?eld,“little”or“small”science—inwell-curatedrepositories,asdescribedabove,manyotherwhichistheprimaryfocusoftheLongTailtheory—isdataproducts—includingdataderivedfromanalysesoftypicallyhypothesis-drivenandledbyonlyoneprincipalmissiondata,synthesisofdatafrommultiplemissionsorinvestigatorratherthanalargeteam().instruments,andsupportingpublishedresearch—oftenremainDevelopmentsinastronomicalresearchtowardteam-esscollaborationsandrapidgenerationofdatainastronomydonotandpreservationthroughoutthelifecyclesofthesedatanecessarilyprecludetheongoingpresenceof“smallscience”in().astronomy,wherescientistsmaygenerateandholdinnovativeTheNationalVirtualObservatory(NVO)()andevolvedintoaseriesofVirtualObservatory(VO)-lifecycleofarchivedrawandderivedastronomicaldata,putation,andsoftware,aswellasprojectproposalsandknowledgeproduction,particularlywhenthisresearchispublications(;;).Furthermore,theNVOwasexpectedtorequireproposerstoprovideplansfordatamanagement,andcollaboratewithacademicresearchlibrariesforlong-termuration().(NSF2015),aarenowactivelyusedthroughouttheinternationalastronomicalmunity,representinganopportunityforlinkingdataintheinstrumentconstructioninUnitedStates,recentlypublishedaLongTailtocorrespondingliteratureandbroadlyfacilitatingesstodata,,includingstorage,preservation,discoverability,andreuse,withtheAstrophysicsDataSystem(ADS)isarobustopen-accessafocusondataandpublicationsassociatedwithfederallytool,indexingnearlyallastronomicalpublicationsandfundedscienti?