 I'm going to get us started and we'll just roll a little slowly here for the first minute or so because I still see folks coming in, which is fantastic. So welcome. We're here to chat today about OSF Discovery. It's some really cool new features that we've launched in the last few weeks and related to that, how you or the researchers that you support in your community, how you take advantage of these cool discovery features that are in the OSF and beyond. So we'll talk a little bit about those as well. This is great, all these introductions, it's really fantastic. It's in Vietnam, that's probably in Zambia from really coming from all over. It's really fantastic. Thank you for all of those. So just very quickly about myself and I put my info in here if you want to learn more about who I am and what I'm about. But I'm the product manager here at Center for Open Science and support the OSF that we're going to talk about today and our other infrastructure priorities. And I'll tell you a little bit about what the OSF is, what our focus is with that infrastructure here shortly, and how we make our priorities and our decisions as we build new things. Because folks like you are a very big part of our direction and I just facilitate what the kinds of needs and exciting priorities that you guys have in your own research communities. So my info is there if you want to reach out to me. I'm happy to chat with you about the things that we're covering today, but anything you want to talk about my team is happy to talk chat with you. So what we're going to look at today in particular is what this OSF search that we've just launched is capable of. I'm going to give you some of the basics and outline of what those are about, but then also I'll jump in there live and show you some examples. So we'll talk a little bit about those and then how yourself or the researchers that you work with how to leverage the tools that are in the OSF to take advantage of that discovery that's in within the OSF as well as other indexes and discovery. There will be a couple more webinars that are similar to this next month that are tilting the frame a little bit to be more targeted toward program officers that work with nonprofits and funders and how discovery is unique to their needs as well as institutional support like librarians and data management support. So those are coming next month. You're going to see some of the same stuff that you would see here if you want to come and join us for those, but also some unique content as well. So very quickly just tell you a little bit about the OSF if you have not heard of us before. So the OSF is a free open source online research platform and it sits among many tools that sound that description would fit. The unique concept behind the OSF is that the goal is to support researchers transparently in their work across the entire research life cycle. So there are workflows for research planning and documenting what you intend to do with your analyses and then for data active data management and collaboration and work with your research teams and then sharing your outputs in the number of ways data archiving and sharing pre-prints and other methods and then finally being able to report those and make those visible and available for your institutions and your communities and other kinds of communities in multiple ways. And so that really is we stand alone and sort of tackling all trying to provide tools across all of those different phases of the research life cycle and then we integrate tools that do particular parts really well too. So citation management and active data storage and data archiving we have integrations into dataverse and OneDrive and Google Drive and Zotero and Mendeley for example. So we know when others are doing parts of this really well we want to integrate those into the OSF as well. And so just to give you an idea where our priorities come from when we are making decisions like this new search feature so I'm going to tell you about is really a confluence of a number of things. So we're looking at a number of these are really big broad buckets here and then lots of specific actors within here but we have those integrators and interoperable opportunities like I just mentioned. We have members and I'm going to tell you a little bit about those shortly but we have preprints and institutions and registries archives that run their own content using our OSF infrastructure. So they are one of our stakeholders obviously and then the data contributors users that are adding and making sharing their data on OSF and other content and then consumers those that are reading their content and downloading it and try to reproduce it and discuss it. And when you sort of look at the dialogue between all of these different groups that's really where we're trying to make decisions and priorities on how we continue to improve the OSF rather than us taking a guess at what you guys need and putting that out there and just seeing what happens. That really is not our approach at all this is a constant conversation across these groups and then building the OSF toward where there those needs are in these broad communities. So you'll see some examples here in the search project that we've just launched and I didn't include here some screenshots of the old search but if you have used the OSF prior to a few weeks ago search had a number of barriers and I'm actually going to ask you to use the chat here while I'm covering this. So tell me a little bit about some of the barriers just in a sentence or two in the chat what are the frustrations or barriers or complaints that you have on just research discovery that not on the OSF in particular although if you have criticism of OSF search I can take it please do share it but just generally you know in a sentence and a phrase what is it that's you know causes frustration for you when you're trying to discover research content and then I'll come back around to some of those if there's some some really good ones in there we might discuss those later but we hear a lot of that from OSF users things that were problematic when they were trying to use OSF search for discovering content that they wanted to build you know their research on top of or to reproduce replicate or even determining if they could find their own content you know when they have published or shared it or hope they did on the OSF or elsewhere and so we were taking in years of data about the search experience the discovery experience on OSF and we had several interfaces all of which really didn't make it as easy as it could be for researchers to take advantage of critical features for research discovery and so we stepped back and redesigned completely our search approach and we're really excited about some of the things that are we're capable of now that that we're really the OSF couldn't support before and really you're not going to find some of these in many platforms so just a few examples here so obviously the being able to start with searching for a phrase you know you would hope that any search is going to have something like that and OSF search is going to will take those in and find not just that exact term but if there aren't many matches for that you know that exact word that you're searching for it will try to find similar terms for you as well much the same as you know Google or other search engines will do we included some of the basic wild cards there's there's all world wild cards out there but we included some of the really commonly used and requested ones like quotations to have an exact phrase or to a minus to exclude something from your results and then some OSF in particular context to like the content type within the OSF so just one workflow you can isolate just those and then some really neat stuff that you're not going to find in a lot of other places but are enabled here on the OSF so you can filter by the funder that supported the research that's reported by the researcher on their content but could be really really useful to be able to filter down by that content see as you're pursuing a grant from one of the NIH institutes you can find a content on on OSF that's supported by that exact funder so it could be very useful filter by an institution that has affiliates on that content so that is based on our OSF institutions features but that's all available as well um you can filter by the the providers those those community service providers that we host on the OSF you can filter by just some of those particular providers and then even connected resources so not even the um phrases and metadata that's on one particular source but things they connected to and indicated they have relationships to so like connected data and code in other resources all of that you can filter by and in any combination so you may find a particular OSF you know registration that has connected data and funded by the NSF like that's a combination that you could take advantage of sorry about the barking dog in the background move forward here and I'm just gonna give you a little um outline tour here and then we'll we'll jump in and actually look at the search features themselves um I'm just following some of the chat items some interesting additions in there lots about metadata um yeah good points um so yeah this is what the search interface looks like here you can see our um if a search term is entered we actually give you lots of feedback on what your um current searches or filters are are giving you back um so this case that's very very obviously broad search term which is science um and I'm getting many many many results so after ten thousand we don't give you a specific count um we'll just tell you it's more than ten thousand but once you get under that number by adding um some of the filters are having a more specific search term we'll tell you exactly how many results um you have and that will continue to update as you add um additional filters to your query um there's those OSF specific object types on the top uh selector there and then all of our our filters for more specific findings um so let me actually jump over to we'll cut some zoom stuff in my way this is our that same OSF search interface here this is all live um so I was just going to take one example here but I will put this link in the chat and you're welcome uh beyond welcome to play around with this as well um so there is the link in the chat um so just for one example if I wanted to to look at some research about hearing loss um you know I'm gonna get I have 322 results um that's a lot better than the you know thousands and thousands that I start with in the OSF corpus but still that's too many um for me to to really get into and look at each one of these results and find what I I need um I do have ways to to get some previews of what's available without even you know opening all of that content but still that's too many um so I can narrow this down to just OSF projects used for um often used for active data management collaboration and linking and sharing and papers of their of their data um I'm at down to 116 so I'm getting there um but what I really would be useful um is if I could see just data you know things that were labeled as data by the users um and now I'm really getting down to um you know some some useful numbers here there's only two of them were labeled as uh data sets and I could also narrow down that further I only need stuff where what's useful to me is what's been funded by the NIH and once I've got all of those uh filters to my search I'm down to one result and I can read a little bit more about what I can expect to find in here and then open this content and look into what's been provided there's some files that they have in here they have a DOI we'll talk about those in a minute but um yeah so any combination of all of these uh filters is what um what you have available to you in this OSF search and you can always kind of reset and um and try these you know again with different filters if you're not finding quite what you want by just removing the the breadcrumbs there um so yeah that is our uh the OSF discovery features that we're really really pleased to share and this is just another look at one of those um search card previews uh in this case this is a pre-registration and you can see um this context field so in this case I entered in my search many labs is what I searched for which is the name of a of a whole collection of projects that were replicating other work um and because that was a search term um I have this context field that tells me the title is where that search term appears um and so I get a little snippet of that if it appeared in the um like the abstract instead the context uh context would say abstract and then snag a piece of that abstract that it would preview for me just like you would expect in uh the Google searches um it does the exact same thing and italicizes those uh my search term in there will work the same way for other uh search terms and uh and filters um so very very neat useful feature there uh to see where my results where my search term occurring um I have my dates and contributors information and then those funder and license and uh resource type uh information that we looked at in the in the filters here all of those we give you a preview of uh so that you can you know really determine if this is something you want to dig further into and can open up this uh registration and see all of the details um and then in with registrations we also have all of those connected resources so the you have data and code and materials papers other supplements that you've already created you can create that relationship to with the pre-registration and you could filter by those uh in that osf search so if you only wanted pre-registrations that had included a relationship to open data you could do that uh and you would only get back pre-registrations who have this badge this relationship um likewise if you wanted only things that included all five of them you could do that as well um so you're getting much much more precise uh information uh returns in your search oops we're already into that um so what does that mean uh for those of you that are producing research sharing research or you're supporting researchers that are doing it um and a couple of the comments in the chat were you know discovery would be great if the metadata could be more consistent um and that is of course uh a terrific point um and you know we you're the term that you're hearing a lot these days is fair fair metadata fair data um which of course we take to heart uh as well um and earlier this year january um we released the number of metadata features some of which we just talked about but all of them with the goal of making uh or enabling research that is added to osf to be fair uh so the being findable there's a lot of identifiers that we take advantage of i'll talk a little bit about those in a minute um but metadata that's available for researchers to include um there's a it's all accessible obviously that's a pretty big part of the open science framework um but also standards um interoperable which we talked a little bit about before but i'll show you in some some other cool examples that all of you could take advantage of doesn't require any development um and reusable so having those licenses like we saw before um and transparent activity logs or other provenance information um and all of this is some of this is happening quickly in terms of uh these are not just good practices anymore these are expected uh practices and uh researchers are making these commitments uh when they get uh research grants or as the institutions are sometimes uh even um um uh are responsible for that compliance um so there's a lot of eyeballs on these needs of course right now um and we want to support that and um you know some of these things are in above the surface in terms of um they're increasingly expected and and visible um but that's not really where um the osf wants to stop we want to include all of these other pieces that can make that research enable that research to be fair so um good metadata obviously important and some insisting standards there but these other relationships and good practices um we really want to include uh the ability for researchers to take advantage of those as well and include those relationships um so in the just in one of those osf projects like we saw before you just can see right there examples of um data that's been you know submitted included in that project space and then there's metadata as well all in one package um so that's your you know as a consumer as a reader or as a funder who's determining if this work is following the practices that the researchers committed to I can get all of that without having to go searching around in different places um and then that also has its own metadata page for even more detailed metadata like the resource types and the funders um so that as much as the researcher is comfortable with uh documenting was sharing about their research we're going to enable and capture that um and we use uh whenever possible standards rather than just free text um so like with the um resource types here we're using the data site resource type taxonomy which is 1415 um types of research objects so um in this case this was labeled as a collection they chose that from a drop down list it wasn't something they just typed in there which um is certainly going to make it much easier to discover collections because you're not having to sort through every kind of spelling of the word collection to find what uh someone identified their their object type as and likewise the funding um support information is coming from the cross ref funder registry I start to type in my funder and it includes that funder name and identifier in uh on my object rather than again a free text and who knows what happens when I uh when they leave me to my devices to just add free text funder names so we take that out of their hands and make it super easy for them to just select their funder and then that's all powering that search that we saw earlier um so just a sense of um some of the other pieces of that metadata we're we're leveraging and we hope that uh that researchers are comfortable leveraging um most of what we have is using the the data site metadata schema there's you know there's a lot of value and consistency across um not just our workflows but in a lot of repositories that especially data repositories they are using that same data site metadata schema so it helps the user and then administrators and others become more more comfortable because they're seeing that same those same standards and we use controlled lists for for many of those things where appropriate like the funders and resource types mentioned as well as uh subjects disciplines and others and the affiliations um we use raw IDs for those um and the other PIDs that we include obviously as object um identifiers with DOIs the researchers can connect their ORCID IDs to their OSF profiles we have those raw IDs for the institutions and the crossref funder registry for those funders so all of that gets packaged in those same metadata containers um and are sent to data site when we meant to DOI for that uh content um and then we enable those relationships we talked about before so you can connect if you have early uh preregistrations you can connect your resulting data and papers and analytic code all of that to the OSF preregistration whether that is stored on the OSF or somewhere else uh it doesn't matter we'll enable you to connect it through uh identifiers um so that you you know you're not forced to just keep using the same platform if you have a good data repository or something um please do use those and then come and add that relationship to your content and then we have an application profile just for metadata um the OSF map which outlines all of these standards so you know the the big part of um accessibility and and consistency is that we can document everything that we're doing um with our metadata and so there is a public OSF map specifically for that and you can even add all of these kinds of details you can add to all the way down to the file level as well um if you have unique metadata that you want to include on your files and all of this is documented in our our help center um so hope you you know if you want to take advantage of those and get feedback on them please let us know um but you know what's critical here is that we realize the discovery doesn't just start and stop on the OSF um so we make sure that the metadata that's um we're enabling the researchers to to add the OSF that actually has a lot of value uh beyond the OSF so um zoomed in here and just one this is one object in the OSF we have a a data management and sharing plan that was required as part of an NIH grant that we submitted earlier this year and we have that plan submitted on the OSF and in the metadata here um we have uh Nicole Nikihoo is our chief product officer here at COS uh she is the key contributor she has her ORCID ID connected to her profile um she has some of those data site uh fields that we mentioned before um the resource type is an output management plan she even included uh the language that the resource is in um so that's all going to data site um and then has the crossref funder ID for the NIH um and even details on the award title um but that funder uh name is coming from the crossref funder registry and then has a license of course and so now we send all that to data site when she minted the uh the DOI for that object and so we have the uh license and resource type um information in here and uh Niki's ORCID ID and her affiliation with the Center for Open Science that all gets captured by data site um the the host repository is here and then that funder and funding specific opportunity are also uh pulled in um so all of this now is in the data site uh corpus and because Niki set up her uh auto sync with between data site and ORCID uh she doesn't have to do anything for this opportunity to now be present on her ORCID record it just automatically syncs to her ORCID record we have documentation up um on help guides on how to set up that sync um but it can be super useful as a researcher uh to not have to come and update all of this uh your ORCID record manually and that's not what they want you to do either they really want these relationships set up so that it's um adding to your record automatically and so this is one example of how valuable that can be but um setting up that auto sync between data site and ORCID you'll get a lot of your content added to your ORCID record without ever having to to think about it um and you can always come back and hide those or remove them if you if you really needed to but um they'll be added to your ORCID record automatically and have that metadata that key metadata that was submitted in this case to data site from OSF will be present in there as well and then I added just here just to um show you the representation in the OSF itself so outside of the OSF and in the OSF we can use those same filters we talked about before um in this case the resource type and the funder and that gets me down to um Nikki's submission here and the preview of some of the things that are in that project and just to give you a sense of all the things that are happening when in that same workflow um this is you know kind of a silly animation here but the idea is that as you or your colleagues are submitting an ORCID record or excuse me a piece of content to the OSF um you can include your ORCID IDs by syncing those with your OSF profile and so we take that information about you and the institution that you're part of and as we're minting that DOI for that piece of content that you submitted we take all of that those identifiers we bundle it up submit it to data site and when they mint that DOI the info gets back to your ORCID record has all those same identifiers in it they also go if you use some of our um if your institution uses some of our institutional tools it'll populate in their reports and in their interfaces and then it's available by way of that same ROAR institutional identifier as well so now it's appearing in many many places without even a specific other indexes grabbing those up even though many do grab OSF content um and share those in their indexes um but they're already now in the data site and in the ROAR uh they're available by querying those IDs uh and your ORCID IDs um so just by starting to include those identifiers when you have the opportunity and your funder identifier being another one um the amount of ways that your content could be discovered and the relationships that are available in there exponentially grow very very quickly um and so that search in the OSF is just one place where it becomes much easier to um if I'm looking for things that were funded by the same funder that your work was supported by or I'm looking only for data sets and your work is a data set that's labeled as such um the data sets also get picked up by the Google data set discovery is just one more example if you had those those metadata um to your your content um so you know a few minutes when you're you're preparing your data to add that extra metadata these are all of the advantages that come with that um so really really worth spending that time um there's some cool stuff that's coming pretty soon um on both the search front and the metadata front um on search those same features are going to be added to some of those community operated spaces like the institutions and preprints spaces um and then also those you know once you're adding all of those filters to your searches um will give you a a URL that you can use so you could have a very complex set of core of um of filters take the URL and just hang onto it and then every time you click that it'll populate those same filters so that you could get you know that set of results so if you only wanted content from one institution that's a particular um resource type and from a particular funder if you wanted all of those to be the presets you would just take that URL and just you know click on it it'll update for you um so really really neat nice little feature that's coming um at the end of this month and then on the metadata front uh we you know we we try to really have key metadata items available in the osf but as a as a general est repository that's available to any discipline um there's only so specific we can get um with the kind of metadata the the options that we have for metadata to be populated um but there are tools that specialize in that having uh being to build community taxonomies for specific disciplines or organizations um CDAR is a a tool in a group at stanford who developed such a tool um and one piece that they have is an embeddable editor where you know if um COS were to develop so kind of specialized uh psychology metadata taxonomy then we could build it in CDAR we would have it enabled in the osf and you would fill out your general you know metadata still but then I could pull up the very specialized uh form by way of this the CDAR integration and this can have a much much different um metadata taxonomy than what we generally capture can be much more specific um and uh you know we can work with organizations and with disciplines subjects to integrate their their taxonomies into the form so that um anybody can choose from one of those taxonomies fill it out and we would populate that metadata into their osf content and then it's searchable and findable and similar to the way that our general metadata are so we are working on this project right now um and hope to to be able to make this available and some of these um metadata most some of these taxonomies available by the end of the year but this is very much in progress so if you want to come and chat with us about it please do because we're really excited about this work so yeah I mean the takeaway is if you want to take advantage of all these cool discovery features that are available not just in the osf but um outside of the osf and many many of the the largest most expansive indexes of research um you know the key is as was pointed out in some of the comments earlier is um be consistent about adding metadata to your work we're we're trying to make the um to use standards in such a way that um if you take that time it won't it won't get too messy because we're we're using you know very specific standards to identify what your resource types are and who your funder is and your institution is um so it can't get too uneven um but the more you can include typically the the better um as far as the discovery goes um but one thing I do like to to end these talks with is that you don't have to try to take on all of this all at once um because there's a lot of these pieces that are um you know more than just adding a little bit of metadata to your dataset when you finish um you know there's more you could do but you don't have to do that in order to take advantage of a lot of the things that we looked at today um there are really quick wins that um you can start to take advantage of quickly like those um getting that ORCID ID and using it not just an OSF but when you when you were given the opportunity to use it and connect it to your your work um take advantage of that because the the opportunities for that to to flow um into a lot of different indexes and workflows and help you in the future is is significant um so getting that ORCID ID and um connecting it you know for the in the OSF for example um right away you're going to have lots of of new opportunities for discovery syncing that with your uh with data site is going to help you too because that'll sync things to your record um and then the more you can tell us about your content when you submit it in the OSF the more likely that um as other researchers are looking for content like it um they're going to use those same standards that we've provided uh in order to discover it so if you take advantage of them on the the metadata capture front then um you're being in a really good position to be found by those same methods um so start to take advantage of that and then you know as you get more comfortable with um with some of those practices then you might want to start connecting using some of the relationships like connecting your your code and even pre-registering um your analysis plans ahead of time kind of like a data management plan but for your whole analysis you could do that um and it's pretty um it's pretty risk-free but you get a whole set of relationships available to you that you didn't before um so lots of ways to increase that transparency and comply with those those expectations now um from funders and policy makers um by just taking a few minutes to just follow the the steps that we've already uh provided here in the OSF and probably many of the other repositories that you take advantage of but um obviously I could speak to what we're doing here in the OSF so um that's wrapping up for me I have some resources here all the slides everything we did today um will all be um sent to you um and there's some links in here you might want to check out uh my info is here if you want to chat with me and my team and we're we're happy to talk about um your experiences with any of this as well as requests for future um improvements or or how you might feel like um your work might be better represented uh in the OSF and in OSF discovery so um really appreciate your attendance today I'm gonna pop open the um the Q&A here and I see that uh there are some questions that uh my colleague Blaine has already answered um so thank you Blaine um there's a question from Shannon is there a way to search for hedges I'm not sure I know what hedges are so um I might need some context on that one um is the data with uh search hedges oh uh huh um so yeah there's um the the easy way for that uh to occur would be um if we have a systematic review template so a workflow specifically for systematic reviews in the OSF and one of the filters that's available is which of these templates um did the user submit um so you could just look at content that was submitted using that systematic reviews uh workflow um and then you would have that narrowed down already to that point and then you could continue to filter as as needed um but that would would be the uh easy first step for that um just a couple of questions in the chat as well so um Manosha asks how how are how to get peer review um done on on OSF so um some of the providers uh on OSF some of the registries in the peer uh pre-print providers that are run by other communities on the OSF um they have moderation so it's it's peer review in the purest uh sense that or you know appear in your same discipline um is reviewing and approving or not um your content and that it would appear in their service um so that's one example um we also have registered reports workflows which is a alternate peer review method where um you would submit information about your study before you go off and write a paper um and journals that accept those um would review what you've submitted and tell you whether they're they'll publish your paper no matter what happens with the actual analysis um with the results they'll they'll publish or not um just based on the methods and descriptions that you've submitted so those are um those are options that are available to you we don't do peer review on on things that are submitted to the general like OSF um corpus except for determining if there's violations of our terms of use in which case we would remove it um but otherwise we don't review um peer review content on the OSF um Marsha is asking a data with DTA file ending SPSS compatible so um yeah it won't it won't run you know the SPSS things in the browser um but yes you could download those um those files and they should uh work as expected um once downloaded um there are several hundred uh types file types that do render um in the OSF um but we don't have sort of machine actionable um computing that's running in the browser um there are some tools that we're developing currently that is going to tackle a bit of that um so I'll have some news on that on another month or so um but that's a good question all right any other questions we're right about at time yeah thank you thank you all it's really terrific questions and feedback thank you all right I think we will wrap up then but please do reach out if you have any other requests or questions and you'll get some uh you'll get resources from the session today um probably tomorrow so keep a lookout for that and I hope to hear from you thank you