 Hi, I'm Dan Cohen, and this is my colleague, Joan Forgezi-Toriano. We're excited to be here to present our project on a different take on scholarly communication. I found the last session we went to also intriguing and I think which had some other models for open access publishing and we'll leave ample time I think at the end to talk about maybe differences in models of publishing. But where we want to start today is with a little bit about where this project came from, why we decided to try a different take on publishing. And as you can see here we talk about it as being scholarship and publication the web way. We wanted to sort of think about scholarly communication starting with the web rather than with a print model. And so a lot of what we'll talk about today is differences in web-first publishing versus traditional academic journal publishing where you submit an article to be reviewed, it goes through some process. We tried to take our inspiration in the Pressword Project from what was going out on the web, how the web reviews, aggregates, curates, validates material, and we'll show you some examples from that. So what we want to start out is Joan and I both come out of I guess digital history or digital humanities more broadly and one of the things that we've noticed over the past few years in digital humanities is that there's a lot of us who are already publishing I guess in air quotes on the web. People who have blogs, this is a very promising grad student, Ben Schmidt, who's a Princeton grad student working in digital history. He has a terrific blog called Sapping Attention and he writes very long, thorough, article-sized pieces on his blog. Recently he's been working with ship's logs and tracking ships and also doing some text mining relating to that. And so there are lots of people like Ben out there I guess in the DH community producing scholarship, stuff that looks like scholarship but isn't in a journal, and also that would have a very hard time being in a traditional print journal. So he often has videos of his visualizations on his site. He has multimedia and things that would have to be flattened out to get into a print journal. He's working in a web first way to put his scholarship up there. There are other genres. My colleague Steve Barnes, who's a Russian historian, has a group blog with other Russian historians where they have different kinds of genres than you normally see in a regular academic journal. So when new books come out they have a very rapid conversation emerging around those new scholarly works on the blog. Here's one about Gulag Boss, which is a Soviet memoir, that they were able to have a kind of real-time discussion right on the blog about. They've also done analyses of old Soviet videos that they pull from YouTube and discuss in depth. Really kind of wonderful use, pedagogical use, and also scholarly use for this kind of web first publishing. There are terrific collaborative non-digital history kinds of things going on now. Frog in a Well, which is a terrific blog, a collaborative web blog of about 20 scholars in East Asian history. It's written in four languages, Korean, Chinese, Japanese, and English, and has really deep scholarship and discussion. And web first just on a blog without the intervention of an academic journal. There are institutional sites that span many different places and disciplines like niche, which is the network in Canadian history and environment where you see geologists and climate scientists working with historians in a kind of interdisciplinary fashion to craft new forms of scholarship. Again, something you normally wouldn't see in a sort of generally siloed view of the academic and kind of Balkanized view where different disciplines like history and climate science would never kind of publish together or rarely talk to each other. This is a good example of a kind of web first place that these scholars can go and discuss relevant issues and sort of create scholarship. It includes things like podcasts and new media that, again, is hard to find in a print journal or cannot exist in a print journal. There's new kinds of publications, sort of incipient publications, Ant Spider B, which is also an environmental and digital publication, which has incredible, rich visualizations, in this case, an article on San Francisco Bay's Forgotten Past where a scholar has taken satellite imagery and walked through how things have changed in the bay over time. Full color, interactive, includes conversation. Again, organically arising out of these fields of discussion into these new forms of publication. We focused here on history and maybe digital humanities, but I would say, as Cliff Lynch pointed out in the plenary, the opening plenary, of course in the sciences there's all kinds of new forms, and indeed for far longer than in history there have been forms of web first publication, archive.org, maybe more recently, plus one, lots of stuff being pushed out onto the web. But there is this real question of how do we validate, review this kind of material. And so that's where Joan and I began this project. How do we think about material that's published openly on the web but providing a kind of layer over that, a layer on top of that that validates, that finds important material and disseminates it to communities of interest, which has always been, of course, the function of the academic journal. How can we re-envision that in a kind of web first way? Well, you look around the web and there indeed are models like this. In technology news there's sites like TechMeme that do their best to kind of algorithmically and now TechMeme more recently has also brought in a kind of human hand to pull out the most important stories of the day. So you can't keep track of the hypercaffeinated tech news every day. On the TechMeme at the end of the day you'll see the eight or ten important stories. And that's a model that's been very effective for them to sort of aggregate and curate this material. There's all human models like the browser, which is started by one person in London, to find high quality, long form writing. It's a very popular site, gets lots of visitors and again pulls things from open venues of writing, puts them up on one site, again sort of validates it as high quality work. Food Press, not often talked about in academia, although we are foodies I guess. So a site like Food Press is really interesting. It takes recipes found in WordPress.com blogs. WordPress.com is now hosting tens of millions of blogs and there's lots of foodies there and they're putting up recipes. And Food Press, which is actually produced by the WordPress folks, again pulls the best stuff, best onion dip recipes from the giant sea of all those blogs and creates a kind of vetted high quality location for finding all the posts on vegetarian food that you'd like to get. Again narrows it down, whittles it down and I think we've seen this again and again on the web and I think we're actually in a moment right now where lots of people are talking about this real need to kind of anthologize and cut down on the vast sea of material that's up there on the web for communities of interest. There are sites that do this, again, based on your particular interest like Paper Lee, Tweeted Times, a lot of these are social network based where they read through your Twitter stream and tries to pull out links that are being retweeted a lot and puts that forward into a kind of daily newspaper just for you. But of course the problem here is that we don't get that kind of validation and this finds things are important to you but of course in academia we care what is important or valid for all of us and we need that sense that we're all kind of reading the same things. One of the things that academic journals do really well is that we know that the handful of articles that come out in a particular journal, the American Historical Review, that most of us in our profession will at least leaf through that and get a sense of what's there and find important work. So thinking about that we put together this project Press Forward, pressforward.org where we, as we say in our motto, sort of bring together the best scholarship from across the web, the open web to produce vital open publications scholarly communities can gather around. Our idea is that there's high quality material out there, it just needs to be organized in some fashion and validated and then sent out to an awaiting community that doesn't want to wade through hundreds of blogs or thousands of websites. So this is our model which is a very different model than the, again, submit model that you get for an academic journal where you send in your Word document or PDF, you wait six months or a year, you get some thumbs up or thumbs down or revise and resubmit and then you wait another couple of years for your piece to come out. We try to work a little bit more rapidly but also in a kind of layered fashion and our feeling is that there's always been layered review in academia. It just hasn't been that apparent because we have this binary world of the journal where something is accepted or rejected. It can only be published in one place which is very un-web-like in the way that we link to good things from multiple different locations. So we are trying to find things out there on the open web and slowly go through a process over time of vetting that into what looks like a journal at the end. So we start again with that open web. So for instance in Digital Humanities and Joan will talk about this in greater depth in just a second. There are hundreds of blogs in Digital Humanities, hundreds of institutional places, centers like the Roy Rosenzweig Center for History and New Media where we put out new projects and new software tools and all kinds of things that are part of DH. Lots of things out there on the open web. Every day Joan and I and a wide variety of other editors will sift through this and again Joan is going to get into this in more depth in just a second to find important new material for that day. We also have a kind of daily relevant content section. We also have a kind of editor's choices where we actually go through and do another layer beyond just what's being retweeted the most. Go through and assess content for quality, new projects for their quality and we'll make editor's choices that will push out to these subscribers. And then every 90 days we actually have a journal which is the best of the best. So we might start with 15,000 blog posts at the bottom of that pyramid. We might narrow that down every day to 8 to 10 just like tech meme of sort of hey this is neat. We'll get to editor's choices which is a narrowing of that. And then to the quarterly journal it might be only 10 or 12 articles, projects, other things that we've identified on the open web that are really worthy of greater engagement and validation that will end up in the quarterly journal. And again we haven't come up with a better term for this. But we do view it as a journal and indeed it will look like a journal when we go through that. So I think I'm going to hand it off to Joan and she'll walk through the process a bit more. So Digital Humanities Now is this sort of now publication, the weekly publication. Since November 2011, so just over a year we've been working in version 2.0. And version 2.0 as Dan mentioned includes an in-house editorial team of Dan, myself and the two graduate students Jerry Waringa and Sasha Huffman. And so what we did as a group is we developed some criteria for what we thought would be useful in ways to identify content for the editor's choice category. And previous iterations of this had been Dan as a single editor in version 1.5, very time consuming job and in the version 1.0 was fully automated publishing. So we decided it was worth a try having a little bit more editorial control over this. So for the first seven months we published five days a week and since June we have published two times a week. Thinking that that is about all the tension our readers have for new content and also it allows us to be a little bit more selective in our choices. So since November 2011 we've published over 1200 posts, over 1,200 pieces. And what we publish are editor's choice, you can see that in the left column there. These are pieces that we've identified as moving the field forward in some way. They are reports of research and progress, they are comments on the shape of the field, they are really about scholarship. What we do is we select an excerpt from those pieces, we select an image, and then we provide a link back to the original. And so in the editor's choice category we're trying to sort of take the pulse on the latest work in the field. And we only have usually one or two editor's choice pieces a day, so only up to about four a week. If you really want to know the most latest work going on in digital humanities we'll pick the four things for you to pay attention to on any given week. So that means that we can identify conversations and trends that are happening in the digital humanities field. And it helps inform our selection of content for the Journal of Digital Humanities, which I'll talk about next. And sometimes what happens is there's a decent amount of conversation about a particular topic, and then we'll do something called a roundup, where we'll select excerpts from maybe three or four pieces that are all kind of on the same topic and group them together. So we just had one roundup post, we call them on December 4th, so last week, and we've already had 475 unique views of that. And what we do in that roundup is we provide an excerpt and a link back to the original for four individuals, in this case individuals, who are writing about the same topic of text mining. So we're trying to surface and redistribute good work, call attention to the good work we see out there. And then we also provide, and you'll see in the second central column, news, information for practitioners and scholars, things that everyone would like to know about, such as jobs, calls for papers and participation, new reports that come out we think would be valuable. And in these cases, we provide just a sentence or two with basic information and a link back to the original. So our goal here is to send visitors back to the original source. We're trying to give the authors and creators of this content full credit. We're just trying to help draw attention. We're trying to aggregate the attention on these scholars and on these activities. Okay, so this is a very basic model of how we do this. We aggregate content related to digital humanities. That's digital humanities on the web. You see a big funnel there. We collect content from the open web, the most open web using Google Alerts. We also have registry of self-identified practitioners of digital humanities. So people who have volunteered their name and their, you know, affiliation as working in digital humanities. And we have over a thousand folks in the registry. And then from those, from that registry, we have about 600 feeds off of people's, or either their personal or their institutional blogs or websites. We have about 600 that we subscribe to and we call that our compendium. So these are self-identified and then also we do a little bit of searching around, too, to add people when we see new projects and new work. So we follow over 2,000 people on Twitter also, which actually is a very large number given that we have over 6,000 followers. So we're trying to build up our follow list on anyone who looks like they do work in digital humanities. We're going to be following back and we're going back through our records to increase that. So all of this coming into this big funnel results in approximately 1,000 posts per week of which we publish about 20. So, you know, something on the order of four editors' choice pieces and then maybe 16 items of news. So that's actually a lot to filter through and obviously not all 1,000 posts are relevant. They're not always related to digital humanities scholarship. Many people use their blogs for many other purposes. But there is a decent amount to filter through. So we're trying out a few different methods of how to filter through this. We're using free services here. Of course, there are services you could pay for like Fever or Percolate or two. But we have customized Yahoo Pipes filter, actually, where we're playing around with keywords and different ways to do that. We also use Tweeted Times, which identifies the content that's most shared from your Twitter, among the people you follow on Twitter. And we're currently developing a prediction and recommendation system with some computer science faculty at George Mason. So we're working on a more customizable algorithmically, more obvious open algorithmic filtering that is something we intend to make available as part of the project deliverables, which I'll mention a little bit at the end. And at the moment, the most useful way of filtering through all this content that we have is we have a lot of volunteers. And this is something we began in June. We have weekly volunteers. We call them our editors at large. We have approximately 70 that have signed up so far. So every week there's about four to six people. We ask them to commit to about an hour a day of looking through the content coming in. And so you'll see that they are receiving the aggregated feeds, just like the editors-in-chief. And then they actually do the first round of filtering through. They do the first round of review of the content. And then they suggest items that they think would be valuable to redistribute. And this works very well. We did, for a number of weeks, look at everything in addition to what the editors at large were selecting. And I think in only one instance found something that had sort of slipped through the cracks. So we found that our editors at large are very good at what they do. They know the field, they're volunteering their time, and they're providing a great service for us. Because when it was just the in-house editorial board and we were publishing five days a week, that took about 15 to 20 hours a week. So that's not necessarily a sustainable or replicable model for most other fields. So with the editors at large and the scaling back to two times a week, now it takes only about four hours a week to publish. And then we have an additional hour or two for the managing the editors at large and managing the feeds and things like that. So that's a much more realistic amount of time for someone else in another field if they wanted to begin their own publication to take on. So that's how we get the content into the site. And then the main way we reach our audience is actually through Twitter. We have over 6,500 followers on Twitter. We also offer RSS feeds out. So we have a lot of subscribers to our own RSS output. And then you can also sign up to have the content emailed to you as well. And every time what we're sending out, again, is a link back to the original, sending the reader to the source and to the producer of that content rather than to us. It's not about us, it's about surfacing everyone else's work. So we've had a lot of visitors to our site. In addition, we had over 41,000 unique visitors so far in 2012, which is, I think, pretty significant if you consider the range and the reach of other kinds of print-based publications. So we're doing this using WordPress, which this is a little bit light, this image, but basically this is a view of the publishing side, the backend of WordPress, where you can see each post is identified, has a separate spot for it in the database. The interface is very easy to use. It's meant to allow you to easily publish on the web, and so we use it for that reason. And also it's open source and extensible. So this is something that anyone could use. And in developing our method, we're basically going to be developing a platform to help others do this work as well, and that's going to take the form of a plugin for WordPress. So it'll be a way to improve the already free platform. We're going to offer a free plugin to then continue that improvement. So going back to the pyramid, that is how we go from the open web to the relevant content to a field. That's the compendium and the radar following our followers and things such like that. And then how we select our weekly items, the now items. And then what happens at the end of the quarter is that we review what we published in the previous quarter, and we're on a very quick turnaround time, in fact, because we publish the best of the previous quarter by the end of the next. So we have a three month turnaround time to identify the best content, get in touch with the authors, prepare the final versions, and then release. So it's definitely a quicker turnaround than traditional publishing. So this is an example of our altmetrics spreadsheet we've got here. So we review all the editor's choice pieces from the quarter. We identify topics and conversations and trends from that period. We use traditional editorial criteria. What was used an interesting methodology, had interesting results, was the most polished, had the most interesting results. Anybody using new sources, things like that. And we also consider what will have the most lasting value. What is the content that someone is going to want to assign to a class, or something that someone is going to want to be able to come back to, if that individual scholar's blog no longer offers that particular piece, the Journal of Digital Humanities can offer a home for that. And what will we want to be able to reference in the future? That's what we have in mind when we're selecting our content. And then we also do some basic altmetrics here. We think about the number of views on a particular post, how many times it was clicked or retweeted. We also pay attention to the number of comments on pieces. This is part of the goal of sending people back to the original, is to encourage a conversation on the site of the author and the producer. So we pay attention to how many comments have been written since we featured the site. And so after we identify the content we'd like to reproduce, we've tried several different things in terms of getting the content from the original format into the journal. Sometimes for our first issue we tried an open peer review, or you might say continued open peer review, because they were already published on the open web and open for comment. And then we work with authors individually to revise their work as well. Another thing that we do is we solicit reviews of tools and projects and exhibits. Those have been the topics for the first three issues. And like I said, we publish one quarter later. So July and August and September quarter will be out by the 21st of December. So we work on that kind of turnaround time. So the next slide is an example of the interface for editing a single post. And you can see at the bottom we use a plugin called Edit Flow, which will also be improving and returning back to the community as a way to communicate with authors. The comments live here. They also can be emailed to the author, so you have different ways of communicating. And then we do the normal WordPress items like select the category and choose which pages it should appear on, et cetera. And then we end up with this front page, which is a scrollable page. So the other thing that we do is we prepare versions of this for download, so versions that are portable. And we produce EPUB using the anthologize plugin, which is very easy. It takes about a minute to do. And we produce an iBooks version and the PDF from the iBooks, from iBooks author, because we wanted to have a little bit more control over what the end products look like, and that was the best way to do that. So these are a couple pages from the second, our second issue. You can see we can include images and we can format them. We have a nice two-column page view. We can include as many images as we want. On the site themselves, you can open them up to see more detail. The other thing that we've been able to do is for the HTML version that's accessible through the website, we can host multimedia. So in our first issue, we had some podcasts and we had videos as well. And that's something that was not reproducible in the EPUB or iBooks or PDF, but still exists online. So our first issue was particularly large. We had 125 two-column pages in addition to about four hours of podcasts and two videos. Our second and third issues have been roughly 80, 85 two-column pages. So we're talking about a lot of content that was produced on the open web. It was available there and we helped to identify it and then give it a larger audience in the hopes that it would be seen as a valuable thing to do to put your work out. So the journal so far, our first issue was released in March. We've had almost 31,000 unique visitors and already 1,000 unique visitors for the month of December and we haven't published, we haven't released an issue since October. So again, this is a large number for a journal, particularly a brand new journal. So we have another prototype that is being worked on sort of within George Mason, Global Perspectives in Digital History. This topic is big enough to warrant a separate publication from Digital Humanities Now. What's interesting about Global Perspectives in Digital History, one of their challenges is that this is multilingual. And so there's, at the moment, German, French, and English and there's Italian and Spanish editors being brought into the loop as we speak. So you can see we have a Google Translate option up at the top and so one of the challenges of this is not only the translation of the material but also working with a distributed editorial board that is monitoring this work in multiple conversations. We also are helping others outside of George Mason work on this problem. We were approached by my co-vandagraph to a librarian at Florida State who was interested in having a similar publication for open access materials and he contacted us about six weeks later. He had open access now and he's running on the same model that Digital Humanities Now is. We helped him understand how to collect the feeds and how to evaluate the content and then helps obviously build the site looks very similar to Digital Humanities Now. So what we're looking to do next, in addition to continue to refine our methods with Digital Humanities Now and the Journal of Digital Humanities is aim to write up some of the best practices for others who want to do this work. We're also developing a software platform that I mentioned, the plugin for WordPress, to offer an easier method to do this. So at the moment we're doing much of our editorial review of the aggregated material within Google services and we'd like to make that something that we host and control ourselves because a free system is never a long lasting system. So what we're planning to do is build a dashboard that's viewable within WordPress. So all the reviewing, editorial conversation and publishing is all done from within WordPress. Currently we have to go back to the original in order to select the excerpt and the content we want to reproduce. If we're drawing it into WordPress right away, we could just move it from one category, basically the review category, into the publishing dashboard without having to go outside. All the content is brought in, we're saved to step and that's actually a pretty time consuming step. So right now we have a functioning dashboard and we're going to be reworking the interface and the styling later this month and in the next couple months, but we're planning to release an alpha functioning version early next year. So what you have here is a screenshot of the reader page. This is just styled by Twitter Bootstrap so it might look familiar to you, but not the long term look. So this would be the interface for reviewing the aggregated content. We'd have editors and editors at large sort of in our model would be able to access this and nominate content. Then there would be a separate page just for the editorial review. As I mentioned, there would be able to draft, move posts into draft mode in the regular WordPress dashboard right away so that would reduce the time for publishing. So we're planning to release an alpha version that we use in our own prototypes in January and then revise for beta release later in 2013 and spend 2013 and 2014 the academic year on helping others who are interested in using this platform or working on aggregating and valuing publishing on the open web in the future. So I'm going to close with a few lessons learned and I think Dan will probably supplement these as well. One thing that we were very pleased to find out is that there are volunteers out there and that they will help you. They're part of the community and are interested in what's going on. A lot of the feedback that we get from the volunteers is that they enjoy reading widely. They're basically assigned to survey the field for a week or to watch it closely for a week and they enjoy that experience and they've learned so much. We get a lot of feedback like that. Also it reduces the amount of labor that we have to put into publishing which we appreciate and if this is going to be something that other fields and communities want to do it's great to be able to say if you have enough volunteers you can do this in four hours a week. That seems like a manageable amount of time for rotating editorial responsibilities. However when you're publishing a journal editorial work review remains. You're not going to get rid of that just because something has been on the open web and received comments before. We do work with authors although we try very hard because of our compressed timeline to not encourage total revision but we're open to it if someone would like to but we're thinking that time to publication is important and this is something that we are experimenting with. It's worth trying. Another thing that I was actually a bit surprised at is that people and groups even the digital humanities field don't publish as much about their work in progress as I anticipated. I think it certainly ebbs and flows. The semester schedule certainly affects this. Summer was pretty productive not surprisingly but I think a lot of people are either hesitant or not thinking it's worth their time to write about works in progress and we'd I think like to disagree. And then the last thing is that there's a lot of work being done collaboratively as thought about as a project maybe a little bit less than the intellectual independent work of what we might call scholarship and I think that it's maybe time to think about redefining what scholarship is. Particularly in humanities this is a little bit more something we struggle with versus sole authorship versus collaborative work in the digital humanities field it's very difficult to do much of the work that is so impressive on your own and yet it requires intellectual knowledge and methods and labor. So we'd like to suggest that maybe it's time to redefine projects as scholarship and encourage people to share their experience and their work in progress because waiting three years until the project is done just so it can be announced doesn't get out to the rest of the community about the development of the practice. So that's a few lessons learned. I think that was a terrific summary. I'll just add that we're going to do several things at one time. We are open access proponents and so we're trying to think about an environment that is kind of a virtuous circle for publishing openly and so if scholars and I think this has really been true in just our first year of the Journal of Digital Humanities scholars are really excited to get into digital humanities now and to JDH just because they get a huge audience for their work and they get it fairly immediately and we're hoping that that encourages more people to publish openly but we will admit I think in a moment of self-criticism this is an open question of whether we can encourage in this kind of model more people to publish their work let's say work in progress or other kinds of new media work in a fashion that we can actually pluck out you've won the lottery you've been one out of 15,000 blog posts that make it into the Journal of Digital Humanities but we do want to encourage that and we also want to encourage different genres to be acceptable in digital humanities and this has been talked about in prior CNI meetings but there is this problem that you have to flatten your work out into straight narrative textual material to get credit for it and so we want to actually credit the material in its original environment as it lives on the web and so being HTML first being on the web first and accepting it as it is as interactive map interactive maps and video audio all these things I think is a healthy thing for us to kind of encourage in this way the other point I would make is as Joan pointed out we're also in the process of trying to think about what a leaner publishing operation would look like so we have had I think rather spirited debates at the Center for History and New Media about things like should we have a house styles for the Journal of Digital Humanities and most journals have a house style and they enforce some specific citation style and so forth and we've really carefully weighed whether it's worth us going through the real time which is our in kind labor and faculty staff etc in kind labor to polish everything up into an in house style so if you look at Journal of Digital Humanities you will see that we have left some material in a kind of I wouldn't call it informal but there's a variety of styles within the journal and we're sort of testing the waters on that because I certainly know from publishing in a traditional fashion that you can get back notes from an editor or from a peer reviewer that really come down to kind of word choice substitutions that really aren't worth anyone's time and leads to an expansion of cost from the supply side so we're thinking hard about that we're also thinking about this question of democratizing the editorial board and so moving from four people in house to 40 or 400 worldwide who have this one week exercise what we found actually is that it creates and sort of reinforces this community spirit that digital humanities now and journal digital humanities are not this sort of external thing that are published by a commercial publisher and you're lucky to get into it there's this feeling that it is sourced out of the community that there are lots of people involved with validating and making decisions about what gets in there everyone else knows that their comments on the blogs and by Twitter are being assessed in a kind of ambient way and so they feel kind of participatory and very democratic spirit is part of that and I think that that helps to reduce the cost because you're not getting that sort of grudging peer review labor that you would get from reviewing let's say for commercial journal fun though I would say we are working to enhance the demand side and Joan talked about a lot of these numbers but we think use is really important we don't want this to just sit on virtual or real shelves we want people to feel that this is important stuff and so the fact that we're getting I think a real exponential increase in RSS subscribers, twitter followers et cetera it really means that you're going to tell an author you're going to get 10,000 views of that new project you launched if we link to you and people feel that as something valid and so then we need to get into the process of how do we then explain that to promotion and tenure committees and all these other bodies that are sort of important to scholars but I think we've made some steps toward that in this first year and I hope that in the platform that we release next year and in helping that we'll test other parts of this model and see where it's weak and strong and make some strides forward for everyone to kind of have an open access system that works in this very different way maybe we'll stop there and we'd love to have a spirited discussion right here