Scalar: Writing Digital Scholarship with Curtis Fletcher

Scalar: Writing Digital Scholarship with Curtis Fletcher


BRIAN CROXALL: Good
afternoon, everybody. We’re glad you’re here at
the end of the semester joining us for the
last of our talks in our future scholarly
publishing series. This year, at the
libraries, the entire series has been sponsored by
the Mellon Foundation. As always, we’re always
grateful for their support. My name is Bryan Russell I’m
one of the digital communities librarians here. And I have the pleasure of
seeing Chris Fletcher today. Curtis is Associate Director
of the Polymathic Labs at USC Sidney Harman Academy
for Polymathic Studies, which is a lot to say. He received his PhD in history
from University of California and bachelors in history of
philosophy from Berkeley. His research spans the
history of technology, the history of humanities
education, science, and technology studies
and visual studies. He specializes in
digital research and writing the humanities
with particular expertise in new models for offering
credentialing and publishing born digital on [INAUDIBLE]
humanities scholarship. He is currently [INAUDIBLE]
on a digital humanities brand, ordered by the NEH
for editorial and workflow in Scalar which is
what he is going to be talking in part about today. And I could say, just by way of
personal anecdote, last summer I had the chance to be in
a workshop at Middlebury where Curtis was there. And we were talking
with a number of faculty about the ideas they had
for digital publications. And they were a couple of us
who brought in the consultant. There was some faculty there and
we got paired up with people. And it was striking
how much Curtis was able to get it done in one day. How bad I felt about– I got with my faculty member. So Curtis knows a lot
about the subject, he’s been thinking about
it for a long time. And so we’re really
glad to have him here. And please join me
in welcoming him. CURTIS FLETCHER: Thank you. Thank you, Brian. Thank you for having me. This is my first time at Brown. And it’s wonderful. I’ve always wanted to come. There’s a way in which
Brown is actually, I think, very
integral in the kind of trajectory of
digital scholarship and digital humanities. A lot of people think that
digital humanities began back in the 1950s with
sort of literate data processing of course that’s
literate data processing. Digital humanities didn’t
really evolve or emerge until more recently. A lot of people think
that that happened sort of in the early ’90s at UVA. I think it’s probably the
case that it began here at Brown in the mid ’80s with
InterMedia and even earlier, in the mid-seventies with the
hypertext experiments that some of you may have been part of. Yeah so back in
the ’70s hypertext experiments with poetry and
other humanities classes, which all led directly up
to this sort of hypermedia you archives of the late ’80s,
which then became, I think, the kind of digital archives
of the ’90s that UVA was responsible for. So I think that Brown is
actually really ahead or sort of behind a lot of that stuff. So it’s kind of like coming
back to the mothership is what I want to say. So OK here’s what
I want to do today. I want to talk a bit about
the overall aims and kind of history of the Alliance for
Network and Digital Culture. And Scalar in particular. And then I want to talk
about the sort of affordances built into Scalar given those
overall aims and that history. And then I want to show
some sort of use cases. The sort of most compelling
examples or genres of scholarly communication
that we’ve kind of identified, that are well-suited to Scalar. And then I think I’ll
end with a brief bit about future and
current development, some of which I think
may dovetail with some of the efforts going on here in
terms of digital publications. So the Scalar team
really emerged out of a series of efforts around
a journal called Vectors. Curious if anyone
here has heard of is. More than usual. Far more than usual. So Vectors is a journal
published at USC. It is published sort
of periodically now. It’s heyday was kind
of in 2005, 2008. That’s when it was
published regularly. And the idea was to create
truly experimental kind of multimodal interactive
humanities scholarship, largely humanities scholarship. So the pieces and vectors
were built largely in flash. They were custom designed. They were interactive. They are haptic, right? They were built in a way in
which the sort of arguments in these, what are
essentially essays, kind of unfolded via the user
interaction, right? Sometimes frustratingly
so, right? So for instance,
this piece here was about the kind of protocols
of traditional knowledge and archiving. And what the user came to find
throughout their interaction was that they did not have
access to a lot of these because well, this
person is deceased. And according to
this culture, you are supposed to be able
to see them, right? If a person were viewing this
piece at a particular season, they might not be able
to see certain images or watch certain videos. It acted both as
a kind of archive and as argument driven essay. But that argument
didn’t become clear until the user interacted
with it in various ways. So the model in
Vectors was this. Scholars would apply. Scholars would come in to USC. And scholars would, for a
month, worked intensively with an information design
director, a graphic designer, a creative director. Basically, a technologist
and an artist and a humanist would get together and try
and put these pieces together, right? After a month of
intensive study, they would go their own way
and over the next six months or so, they would build
one of these pieces. And six months later
and probably $30,000 later, you would get this
beautiful, beautiful flash driven interactive
humanities essay. So a lot was learned
about what it is that scholars wanted
if they were given every affordance, every digital
affordance in the world, what they would want to build
and what they would want to do. But it certainly
wasn’t scalable, right? And it wasn’t sustainable. It wasn’t sustainable because
you get funding each summer to do this, whether it’s
from NEH or someone else. It wasn’t scalable
because you can only reach so many scholars. So the idea was how does
one go about templatizing the sorts of lessons learned
working with scholars in depth over a summer to try
to build these kind of experimental digital
humanities essays. And from there came Scalar. And in fact, there’s
an inside joke here. And I never know what
I mean when I say this, but apparently,
mathematically, Scalars are how you scale vectors. Anyone know what that means? So somebody knows. But in any case, that’s
the inside joke f from our information
design director. So from vectors,
you get Scalars. So Scalars is a way to
scale that out, right? So Scalar was born
out of two contacts. The first was the contact
that I just told you, right? What is it that scholars want
to do in the digital world if they could do anything
that they wanted. The second was the formation
of the Alliance for Network and Visual Culture, which
was a set of partnerships between digital
archives and archivists, humanity centers working
in digital modalities, and academic presses, who
were interested in publishing these sorts of work. So Scalar kind of
sits at the center of those sets of relationships
as a kind of technical hub. And the sort of
concerns and issues from each of those stakeholders
are built into the platform. It became clear pretty
early on that the only way to truly get digital scholarship
adopted at a larger scale was not just to give scholars
access to these affordances but also that we would have
to do it in lockstep together. Us, the humanities scholars,
the presses, archives, everyone. So Scalar is a way
to try and build those concerns in and move
everyone online together. And you’ll see some of what
I mean about that later on. So what were those affordances? What ultimately got
built into Scalar? Well, let me just over a couple. Of the things that we noticed
was that in the digital world, scholars wanted to
be able to attend to their sources and their
media with more detail than they could in print, right? So in other words,
it isn’t just as is they wanted to get a lot more
media onto the page, right? And in particular,
film onto the page. But they also wanted to be
able to attend to it in a more granular level and
comment on that media. So this is a late
antiquarian text. It’s part of a digital edition
that should be coming out soon. And what they’ve done is
they scanned and digitized all the pages really
high resolution and put them in a Scalar. And then on the more
graphical pages, they’ve actually annotated
each and every section to translate what’s going on. This is a montage about– it’s an essay from an
essay Chaos Control by Steve Anderson. It was actually published
in American literature. And this is a montage that he
put together of different clips from the ’50s and ’60s about
kind of punch card chaos. The way in which
computers just sort of go on the fritz in all
these movies, right? [INAUDIBLE] like desktop. OK so he created
a montage and then he annotated it so
that at the beginning of each part of the montage,
the new commentary pops up. So you get this sort of
running commentary, right? It doesn’t necessarily belong
in the body of the text. It’s something that more
properly, if it can, and in the digital world it can,
live more adjacent to the media itself. Scholars wanted to be to
directly connected to archives that they were working with. But the archivists
that we’re working and the archives that we are
working with also want this. So one of the things that
Scalar allows you to do– and this is in particular,
the omega importer– it allows you to, without
leaving the authoring interface, just grab
media from various assets, whether it’s the Getty or
Internet Archive or Critical Comments without ever leaving
and just pop that material right down into your project. And then add it
into your project. So the idea was
that by doing this, Scalar would help close the
gap between the scholarship and the archives from which that
scholarship kind of emerged. In terms of structure,
what we noticed is scholars wanted to
do a couple of things. They want to be able to create
pathways through the material, right? Sounds familiar, right? They want to create narratives. They want to create
mini narratives, right? So we built in
what we call paths. They also wanted to group
material under a common heading or as we laid it. So we built in a
tagging function. Because we didn’t actually know
exactly what scholars would ultimately want to do
with this platform, we built in what we might
call a kind of flat ontology. So we have paths and
we have had tags. And we have annotations. And we have medium,
we have pages. In Scalar, those things are
totally equivalent, right? So for those of you who use
other platforms like WordPress, you know that there are
many different elements to the WordPress site. They’re not really
equivalent, right? Posts aren’t the
same thing as pages. You can’t edit a
post the same way. You can’t put a post into a
main table of contents, right? When you have a piece
of media, it really only exists in reference
to the page that it’s on. In Scalar, everything
is equivalent, right? Which means that
you can do anything you want to anything else. So not only can you have
multiple paths that share material, but that
material can also be a path or that material
can also be a tag. It can also be an annotation. So you can annotate
media and then you could tag all those
annotations to some other page. And you can put
that page on a path. And the idea was we wanted to
build in as much flexibility as possible because. We had some sense of what
it is the scholars wanted to do given these
intensive study, given these intensive
workshops with them. But there is a way in which that
was an artificial environment. It was more like a laboratory
than anything else. In other words,
these are the ideas that scholars emerged with
while working with technologists and artists, right? So we don’t necessarily
know that that’s what they’re going to want to do. So we wanted to build in
total flexibility, right? And this is the
way that we did it. We also wanted to
remain kind of agnostic to the media that
can be imported. And we did this for
a couple of reasons. So obviously, you can import
video and you can import images and you can import audio,
that sort of thing. But we also allow you to import
anything they will play nicely with iframe. Basically anything
on the web can be imported as a
media object and put in your media library,
put on a page, tagged, and you could build any
relationship you want between it and other elements. So these would be the objects. Now Scalar does
actually natively now support stl and obf files. So some 3D objects. But this is not one of those. This is actually a 3D
object from a player outside in the world wide web
somewhere out in the wild. And they just imported
that as a media object. And this was done before we
even supported 3D objects. And the idea was that we don’t
have to kind of keep ahead. We don’t know what
it is that’s going to be the next compelling
object of analysis for scholars, right? Three years ago, it wasn’t 3D
objects and now what it is. So we wanted to build that
kind of flexibility in as well. And also say that I think it’s
important because compelling digital object of analysis– not only is that a
moving target and you don’t know what people
are going to want to use. But it’s also the
case that, I think, the things that we’re using
online will at some point become the things that we’re
critiquing and talking about. We want Scalar to be
able to do that too. So not only can you
embed a neat line map, because it does something
that Scalar won’t do, but you also embed
a neat line map because you might want to
critique how neat line maps are used. OK. And the last thing I’ll say
about affordances is this– Given that there were all these
kind of rich relationships that you build up
in Scalar, we wanted to make all of that as
legible as possible, right? And this wasn’t just
something that we had to come to terms with. This is something
that in all sort of hypertextual
scholarship, this becomes a kind of friction
point kind of thing. We see this even with
early hypertext systems. If you read reviews of
these things, it’s always, I kind of have some sense
of where I can go next but I have no sense of
the global structure of this project. And as a reader, we kind
of want to know that. Used to, what we’d do is we’d
flip between the page we’re on and the table of
contents or the index and we get a sense of
where we are, right? In the digital world, that kind
of large scale global structure also, I think, needs
to be legible , especially in a platform
like Scalar, where basically, what you’re doing is building up
relationships between elements in a project, right? So this visualization is
always accessible for a reader. You just go to that
compass right there. And so what I want
to point out is that they’re doing all media. So here’s what the reader sees. So there’s like 12 pieces
of media in this project. And this piece of
media is on this page. So again, everything you do in
Scalar is really [INAUDIBLE].. You put a piece of
media on a page, that’s the relationship
in the database, right? If I annotate a piece
of media, that’s a relationship in the database. So this media resides on this
page and it has an annotation. And this this piece of
media has two annotations and that annotation actually
has a piece of media in it. So again, in terms
of flexibility, annotations are
their own things. So you can put pages
in an annotation. You can out media
in an annotation. You can put an annotation
in an annotation. And of course, at a larger
scale, not just looking at all media
[INAUDIBLE],, all content. It’s fairly easy to get a
sense of the overall picture of the project. So this is a path
called Chapter 1. And these are all
the media objects that they have tagged
with that particular tag. But of course, this can get
pretty unwieldy pretty quickly, right? And so this is where
support kind of comes in. So the ability to
build up relationships as you’re writing in Scalar
is both an affordance and a challenge. Which is to say, when
you’re supporting scholars who are doing this
kind of scholarship, you need to make sure that they
attend to both levels, right? What’s going on at
the page and what’s going on at the global level. Because things can get pretty
illegible pretty quickly. So it might be that a
scholar wants to build a path about some topic. So they want
readers on that path are kind of going
along the topic. But they also– this
is something you have to continually insist on– they need to know that
that path shows up here. That they are
building a database as they’re building a narrative. And the question
is, does that path, are those relationships,
while good at the page level– do they muddy this
visualization? Or do they muddy some overall
global sense of the project. OK. So what is Scalar. So very specifically,
Scalar is a semantic web authoring a publishing
platform that’s meant to do a couple of things. It’s, as you’ve already heard,
meant for scholarship that is media rich or media centric. It’s meant for scholarship that
is connected to online archives and repositories. And it’s meant for scholarship
that is sort of structured in innovative ways. Now it should be said that
the vast amount of Scalar projects out there
are not created by us or supported by us at all. It is open source,
it can be downloaded. But its also hosted. That’s the version
most people use, right? It’s free. Anyone can register. We have about 15,000 users
and probably 20,000 books. These are all open
sessions in Scalar as of like, 3 PM, Pacific
Standard Time, a week ago. Not everyone is
necessarily working, but their browser is open and
they’re logged into Scalar. So you can see
pretty global usage. Given these affordances,
given these aims, and given the way that we
end up supporting scholars, how do they end up
using Scalar, right? So here’s what I like to do now. I like to identify the genres
that sort of emerge, I think, as those most well
suited for the platform, for the performances that
will be built into it. Given its historical trajectory
and its overall aims. And while doing that, , answer
a couple of questions, which is, why do scholars
use the platform? What affordances do they use? What affordances don’t they use? What do they ignore? At what point in their workflow
do they come to the platform? What kind of support
do they need? What kind of
support do we offer? Is it just training. What kind of support leads
to further development. So not just the use cases,
but the overall ecosystem in which they emerge
and to which they’re sort of outputted, right? OK. So the first compelling use
case that emerged really was, the what I’ll
call book companion. So these were instances where I
had written a print monograph, typically in film
or media studies, and their press
obviously just didn’t allow them to putt all
the media that they would like, all the
evidence that they have in the print version. And so they decided this
scalar would be a great way to create a kind
of companion that contained all of that media. So on the left hand side, you
see the next the Nicest Kids in Town by Matt Delamont. And on the right hand side, you
see Jason Mattel’s complex TV. Matt Delamont is at
ASU and Jason Mattel is actually at Middlebury. So in each of these cases, the
presses were pretty hands off. They were excited about
what the scholar was doing. They embraced it. The Universally of California
press in particular was excited about the
idea that this companion might drive sales– the print version. And by and large, that has– I’m sure you guys have
read the research– been the case. Not just with this book. There’s no research on
this book individually. So both presses were excited,
but they were pretty hands off. So in other words,
this was the scholar saying I want to get this stuff
out there and they came to us. And in terms of support, there’s
a couple different scenarios. Let’s say that there’s
three different scenarios. So one scenario is
the scholar knows what our platform does and
they know exactly what it is that they want to do. And there’s a pretty
good fit there. So our support in
some ways amounts to the affordances already built
into the platform and maybe a little bit of training. That was certainly the
case with two, right? In other words,
what they want is they want to get a
bunch of media online and they want to mark it up. And it’s pretty clear
that Scalar does that. And so without much
help from the press, these scholars grab
an account on Scalar and put all the media
up and write around it and lo and behold,
they have a [INAUDIBLE].. They do you act somewhat
differently, I will say. And so even within this
sort of book companion genre that I’m calling it,
there are some variations. So Mattel’s book acts more as
a kind of archive of the film featured in the books such that
it’s like an annotated archive. So in the main
table of contents, you can kind of get at
the media via the chapters of the book itself. But you can also just
go to this gallery. And so what he’s done is he’s
grabbed the analytical portions of the portions of
his print monograph that analyzed these
particular clips. And just put it into Scalar. So again, it acts as a kind of
annotated archive, I would say. Whereas what Matt has done
is in some ways recreated a pared down version of
his narrative, right? So the argument is there. The same argument that’s
in the monograph is here. They just have a lot more
media and a lot more evidence. OK. So the second use
case I would say is long form
argument driven work. So this is the project
by Paul Harris at LMU. And this, I think, is not
only a different use case but it’s also a kind
of different support scenario in this way. So there are
instances like I just said where are the scholar,
they know what they want and they know what
Scalar does and there’s just a really good fit there. And boom. There’s another scenario where
the scholar doesn’t necessarily know what it is that they want. And they also don’t
know what it is Scalar or the digital in
general can do for them but they really want to explore
those two things, right? And in some way, these are
the more fruitful support scenarios. These are, for us at least,
the more fruitful engagements and relationships. Because this is a scholar
who knows that all he knows– they can’t do what
they want in ink print. They want to do something else. And they just want to know
what digital can do for them and Scalar in particular. So this is a
scholar who had just all these writings on rocks. Poems on rocks, critical
essays on rocks, historical essays about
rocks, all this stuff. And just organizing it
literally on the page just was not satisfying for
him so he came to us. What can we do. What can Scalar do for me. And so pretty quickly,
the idea emerged that well, what we would do
is we would create pathways as though a reader was
walking down a stone garden and they can switch hands
here and there right. So basically, this
is what I was given. This is what the
scholar gave to me. And this is what I would like it
to do if it could do anything. So multiple paths that lead
in different directions. It ends up being a kind of
choose your own adventure scholarship, where a reader
can be on a path about poems of a particular rock and
then they can switch paths, so forth and so on. So that’s what I get and
that’s what I give back. OK, this is what you want. And then that’s how
it lives in Scalar. So again, all that
can be visualised. Which she was very happy about. It isn’t just at the
page level the reader can be taking these little
narratives as though they’re going down path of rocks. They can also see
the overall path. So this is this kind of
is the kind of scholarship that I would say is immersive. So in other words,
these are scholars who not only don’t necessarily
know what it is that they want. What they do know is they
want to immerse their readers in their evidence in a way that
you can’t do it in ink press. These are scholars who
are trying to figure out the ways in which long
form argument driven work is amenable to a
kind of experience. How does one
experience an argument. The ways in which it’s amenable
to immersion in the material. And the way that it is amenable
to, by the way, user directed experiences. OK. And the same is true here. So this is another
scholar from Middlebury. So this is a project that
actually came to Stanford Press about a year ago. So maybe you may know Stanford
Press has a Mellon Grant right now to kind of figure out
what it would take to publish born-digital monographs. And in particular, what
kind of human infrastructure you can [INAUDIBLE]. What kind of acquisitions
data would you need, what kind of
design production person would you need. And so their
acquisition [INAUDIBLE] has been kind of
plugging around, trying to find born-digital
monograph to publish. And this originally
in WordPress. And when it lived in WordPress,
it was much more of an archive. So there were sections,
there were essays, and then there was
the kind of archive that you can get
at by filtering. So an archive of poems. Just tons and tons of poems. And you could filter
them the same way you would filter on
Nordstrom Rack or Zappos. It’s database driven,
but it’s like, I want a poem that has to do
with this, this, and this. OK. And then it outputs those poems, And so the complaint was that it
was just too much of an archive and it wasn’t an argument
driven monograph. So the idea was, oh,
OK, well let’s look for another platform. And I don’t know if
it was Stanford Press or it may have been Alicia
Peaker at Middlebury. One of our power users. A wonderful person,
[INAUDIBLE] by the way– who pointed in the
direction of Scalar. And so the idea here
is that you started with three separate paths. And each of those have
their own trajectory. But the most interesting part of
this book in terms of structure is that again, in what ways
are a cumulative argument amenable to a user directed
kind of experience. So in this case, in order to
get to that archive of poetry, the reader has to
recapitulate the experience that these poets in particular
make when they write poems. Is it sentimental. Is it not. Is it referring to
someone specifically. Is it for public consumption. Is it not. So all these tacit and
well articulated decisions that these poets
make in this culture before you sit down
to even write a poem– the reader has to go through
a set of binary decisions in order to get to
the archive of poetry or the particular part
of the archive that relates to those choices. OK. So continuing with argument
driven long form works, I would say there’s another
kind of genre within that– a subgenre within that. And we differentiate
them in a sense. So the last two, the
scholar is trying to figure out how recreating
a particular experience of a set of works or an
event or some culture is amenable to scholarship
in general or argument driven scholarship. In these cases, I
think what’s going on is the scholar is
instead trying to figure out the way in which argument driven
cumulative work is well-suited or in any way suited to a more
robust exploratory structure. So in other words, it’s
always the same question we’ve been asking for 30
years, which is, can you have an argument driven
work that is hypertextually linked so much that the reader
can go anywhere they want. So in this case, this
is a set of 42 essays. This is being published by the
University of California Press too, by the way. A set of 42 essays. The equivalent of like, 900
pages, something ridiculous. Like 800 media objects. But the idea is that from any
paragraph within these essays, you can link to other
paragraphs in the essay that relate to that material You can
also navigate them spatially. And again, you can go from
a map to a given paragraph and from a given
paragraph back to a map. So a really robust kind
of exploratory structure– maybe not necessarily below,
but inside of those arguments. And same thing here. So this is a woman who
did a dissertation at NYU. It was on the ACT UP
movement in 1980s. And so what she did was as she
was doing her research, as she was writing it up, she
had pages in Scalar for people in the movement. And particularly about
protests, where they took place. The particular issues for
each of the organizations. So organizations, people,
protest, the movement overall. And then of course, the concepts
that had kind of emerged in her research. And then she tagged
them all together as they existed
in the real world and also as they existed in
her scholarly imagination. And what she gets is a vast
network analysis of that work. And what she found was
that it didn’t exactly map on to what she was thinking. So there were some
clusters in some areas that she didn’t
imagine were there. And there were some
orphans in some areas that she didn’t
imagine were there. So the point is
that the author is building a database as
they’re building a narrative. But that database isn’t
just for the readers. It’s also for the author. Or at least it can be. Those relationships can reveal
things about your reader if you’re using it as a
conceptual model in this way and not just as a sort
of navigational layer for your reader. It can reveal things
that you yourself couldn’t have gotten at
by doing a close reading. OK so short form argument
driven work– meaning essays. These tend to be a little
more traditionally structured. So this is Chaos and
Control, actually, which I showed you earlier
when I was showing annotations. This was published in an
issue of American Literature. And I think just
because they’re shorter, they tend to hew to a more
traditional structure, meaning they’re sort
of just sections. So they tend to use
paths a lot and tend to ignore tags altogether. And same thing here. This was a special
issue of Urban History that was all
published in Scalar. So it was like five
different essays by five different authors. And the same thing. The authors here tend to use
the multimedia affordances of Scalar meaning they can
put all the media they want in their particular
film and they create their little narratives. But that’s about the extent
of the kind of affordances that they use with
these shorter pieces. OK, so book companions. Argument driven work. Long form, short form. And then I will say the
other kind of use case that has really emerged is,
surprise surprise, archives. And there are a number
of different ways in which people are using
Scalar to create archives. And in fact, I think that
those different uses are really starting to blend in some
ways not just in Scalar– elsewhere, I think, also. This has been going on
for years and years. But I think with
Scalar in particular because Scalar
sits at the nexus– it is at the same time
connected directly to archives. So the work that
people do in them tends to be archive driven. But also, it’s not omega. It’s meant for narrative work. And so it’s in some ways,
meant for argument driven work. So I think there’s a way in
which some of the genres that are emerging using
Scalar right now kind of sit at the intersection
of those too. This I would consider
to be more of a kind of traditional digital archive. Again, going back
to UVA and Dickens archive, Valley of the
Shadows, [INAUDIBLE] archive. Which is to say it brings
together all the material from various archives that have
to do with a particular event, particular author. They used to be called
thematic research collections. Maybe they’re still called that. I don’t know. The idea is that you bring
all this material together. The archive is kind
of the primary object. There’s an essay kind of
framing the collection overall. But you’re really putting it
out there for scholars to use. Sometimes they’re also
called virtual humanities laboratories. The idea was that you’re
putting this all together so that people can use them. So a number of those– this was created by
Will Fenton at Fordham. Wonderful guy. Great scholar. Writes for PC Magazine. Sorry, this is the same book. So political cartoons are from
one archive and [INAUDIBLE].. So it gathers
together everything having to deal with this
particular event, right? But the archive, I think,
is the primary object. On the other hand, there
are what I would consider– sorry, the bells are
sometimes confusing me. Is that the interface
or the bell? This is one I might consider to
be more of a scholarly archive. I don’t know. I’m still trying
to tease this out. You know, there are editions
and there are collections, meaning with collections,
there’s kind of a one to one relationship between
an actual collection– a holding at an institution–
and the actual digital project. Then there are
archives that bring lots of different
collections together, right? And then there’s kind
of critical work, which is more inward looking,
that kind of marks up the text. And then there’s more
extra-textual work, which is relating that
archive or collection to its overall
historical context. And then there is
more scholarly work, which is using that
archive for a kind of historical
intervention or something. And I think those kind of
10 different elements– there used to be
critical editions and there used to be
scholarly editions. They’re all kind of getting
mixed up and blended right now. And I think that’s
what we’re seeing here. So this is different. This has an archive. So quickly, [INAUDIBLE] was
like eight different scholars. And what they did is they
grabbed about 3,000 photographs from this photographer. And it wasn’t just all of the
photographs of this archive and all the photographs of
this archive– they curated it. So this was a scholarly
curated archive. It’s these ones
from this archive and these ones
from this archive. You pull them together,
there’s 3,000. This is now our archive. This is our archive. And now we’re going to
write through that archive. So these are all
essays that write about this particular
collection. So in some ways, I think that
the scholarly intervention is more the primary
object and the archive that the curator
sits underneath. But of course, in Scalar, it
doesn’t just sit underneath. Because you’re writing
through it, right? You’re running
through that archive. So if you’re on a
given piece of media and you click what we
call the Citations tab, you’ll get that piece
of media and you’ll get all the different places
where it’s been interpreted. So the media acts
as a kind of node between those
different pathways. So it’s an archive
but it’s also– and again, not only are
those genres, I think, blending and those sort
of modalities blending. But I think it’s also
possible that when they blend, they’ll close the gap
between the archive and the scholarship itself. So again, going back
to what I said before, how Scalar emerged from a
couple of different contexts– one of which is we want to
take all the consideration of the different stakeholders. So in Scalar, when you import
media into your project, that media stays at the archive. We don’t duplicate
it onto our servers. It stays in the archive. And we did that because our
archive partners were like, we don’t want you duplicating
our carefully curated copy of record. One. Two, the digital
allows us to change the relationship between
scholarship and archive from which it emerges. So always before in print, it
was like you go to the archive. It was more like a
vampiric relationship. You go to the archive
and you take and you run. And you footnote it. But no one is ever going to go
back to that archive, right? In the digital, it’s
like, no, the reader can always be just one
click away from the archive from which this
scholarship emerged, right? And in fact, in an ideal world–
we keep talking about doing this– we just don’t have
the resources yet– is building a kind of
round trip function where the interpretive
part, the scholars part, can actually round trip
back to the archive. So you have to be
curated in some way. So the sort of
interpretive layer on top of those
assets that are used can be a referenced
from the archive itself. And I also think
that it’s possible that in a particular work like
this, it may be in 10 years– well, we always say that but
it never happens, so I’m going to say like, maybe 50 years– maybe in 50 years, it will be
just as common for scholarship to exist in an
archive as it does for it to be published by
press or somewhere else. Who knows. AUDIENCE: In some ways, that’s
the meaning of libraries. CURTIS FLETCHER: Yeah. Right, they’re always sat
side by side on the shelf, but conceptually,
this monograph is just a bunch of writing with a note
here and there about this. So yeah. That’s a really good point. [INAUDIBLE] is the new library. AUDIENCE: But this is also
aggregating content, right? It’s interesting
how archive becomes this term that has a certain
degree of value and prestige. Whereas if you were to
say, this is an aggregator, it would be like, oh that
doesn’t sound seem as– I’m just curious about how
these terms are circulating [INAUDIBLE] see them. [INAUDIBLE] frame us. CURTIS FLETCHER: So two things. One, aggregator seems to
imply that there is maybe an algorithm involved. Whereas these are curated. They were chosen. Some things were included,
some things were [INAUDIBLE].. The other things is that
I’m saying digital archive– and I’ve used this term
before with other librarians and they’re like,
well, archive means that there’s the copy
of record and there’s another copy of the scholar. But I’m using it in
a very loose term. So digital archive
is kind of what people came to call
these things in the ’90s. Before that, they were called
like hypertext archives. And then it became
digital archives. So digital Dickens. And the Rosetti archive
and so forth and so on. I’m just sort of using
the term in a loose way. AUDIENCE: Oh no, [INAUDIBLE]
but it’s also interesting. CURTIS FLETCHER: Yeah. And so that’s the
other thing too. Not just how it’s written
about my round trip back to the archive. But also, which objects
did the scholars decide to put together. Why did they curate it this
way and not another way. It’d be interesting for
the archives to know that. So again, it would be
interesting to know how they’re getting aggregated
in different contexts. I should say, by
the way, I enjoy– I know that we have a lot of
question and answer at the end, but if you have questions
as I’m moving along, I’m more than happy to
field those questions. So this is just the same work. But I just wanted to show it. So there are the
interpreted essays, but also, you can just
browse the archive. It’s [INAUDIBLE]. And just to get technical, in
Scalar, these are each paths. So you just create
a path of media and then you create
a top level path and then put it to
a certain layout. And then boom, all
your media is there. OK. So– yeah? AUDIENCE: Could that also
have been done with tags? CURTIS FLETCHER:
Yeah, absolutely. Exactly. So if I had created 10 different
tags and put those on a path or made that a tag and then
done that layout, same thing. Any parent-child relationship
will display them that way. So what we saw before
was a series of– again, I’m calling them
archives, which is to say the scholars themselves
have done one of two things. They either selectively grabbed
things from here and there and said, this is
our curated archive, which is the second case. The first case was a more
traditional digital archive where there’s like,
we want everything that has to do with this
particular author’s work, whether it’s Whitman
or someone else. Or we want all of
the material that has to do with this
particular topic or event. Valley of the Shadows or– what’s another good one– Uncle Tom’s Cabin
in American culture. These are all UVA. I don’t know why. That’s like the only thing. So same thing. It’s like everything
that has to do with it and then we put it out there and
scholars get to work with it. This is a different genre and
I think it’s an emerging one. This is what I would call
the digital collection. And I call it that
because two reasons. One, scholars are not the
agents, they are librarians. And two, it showcases a
particular collection held by a particular institution. And anecdotally, I’m seeing– I just go through
Scalar’s database and I’m looking at titles– which I do– I’m seeing more
and more of these. The Hoover Institute, Stanford
is doing a bunch of them. Newberry Library
just did a great one on an exhibit of
Shakespeare stuff. LBJ Library. Anyway, I think it’s
something that librarians are moving more and more towards. It I think they got
their feet wet with Omega in a way and building exhibits. And some of them now want
to build a collection, they actually want to
have more narrative. So this is a good example. The Voltaire is actually at USC. And so what this does is
it does a couple things. It is a thematic research
question in the sense that it’s about trying to
get this collection out there to scholars and to,
students for that matter. But it’s really meant
to animate and showcase a particular collection by
a particular institution. So it’s less scholarly
in that way, I think. They tend to have
essays but those essays are more about
framing the collection and talking about the
provenance of the collection. Where did the
collection come from, how was it acquired, and
probably something about what does this particular collection
say about, in this case, Voltaire scholarship. These letters in particular–
what historical interventions about the Enlightenment can
be made about this collection or made with this collection. Or what is the provenance
of this collection, say about the distribution
of Voltaire’s writing after his death. In this particular
case, every letter has a translation
and a transcription. But it is still kind
of hypertext document, which is to say that– up here, we have
Paris is annotated. And so if you click on Paris,
what you’re going to get is a page a on Paris. And that page is tagged
to all of the letters that talk about Paris. So you can access them here. You can also access them via
a map, that sort of thing. What is this. Oh, just a timeline. That’s me. So the last, I think,
support scenario. So there projects
where they know what they want and they know
what Scalar does and it’s easy. And then there are the more
exploratory support structures where people just have
some fuzzy idea of what it is that they
want to do and also a very fuzzy idea of what
the digital can do for them and let’s find out what
scholarship might look like. There are others there with
I think where the scholars or in this case, it was
the Voltaire project, faculty or staff know what it
is that they want and just want to know whether or not
Scalar can do it for them. Now in those instances,
I think there’s still some negotiation that goes
on a series of levels. One is well no, Scalar
can’t do exactly that. So you have a choice here. You can either make
the affordances, do what you want, you can
change what it is that you want, or you can try and get
us to develop something that will do what you want. Development is kind of always
in the picture here, right? Because after all,
these use cases are the laboratory from
which we develop out further affordances. And so there’s always this
kind of balance, both for us, meaning if we
develop this for you, is this something
others could use. Perfect example, remember
for the stone path project? So Scalar– you can
branch paths in Scalar. But it isn’t all that
legible because– I can’t get into it–
the way that paths work. Well first of all,
they’re numbered. Paths are numbered. So if you put two
paths on a path, it’s going to say
path one, path two. And if you’re doing a
choose your own adventure kind of thing, you don’t
want those numbered, right? You want those to be equivalent. So there are some
cosmetic things that would prevent it from
being like a true choose your own adventure. So when that scholar wanted
to do that, we built it. We built a small
script, basically, that allows people
to add metadata to particular pages that
says take that stuff out. Well that then got used
by the second scholar that I showed who was actually
doing that with the poetry. Because we anticipated that
that’s the exact sort of thing that more and more people
are going to want to do and we were right. OK, so there’s always
that balance for us. Then, there is a
balance of well, if we do some custom
thing because we want Scalar to do
something that it can’t do, who’s going to keep
that up, right? Who is going to maintain
that– whatever it is, jquery plug-in– who is going to make sure
that it’s always live and it’s always running. So there’s always an issue of
maintaining your dependencies. Even for projects
that aren’t published. Particular products that
are published, but even for projects like this. The other thing is
that even though they knew what they wanted,
they didn’t necessarily know how it would
work in Scalar. And so one of the
kind of challenges that we often run up
against, particularly with large projects, is there
is this kind of blockage or this melting
point where scholars will get all of their
material in there and then they will just
be kind of stymied. Because they keep hearing
this mantra in their head, which we’re responsible
for, which is you can do anything in to
everything in Scalar. And anything can
be anything else. And you can build any
connection you want and everything is a
relationship and non-linear and non-linear and non-linear. And so there’s this like
moment where they’re just like, well then, what do I do. Like, how do I even begin
to structure this thing. Again, particularly
in larger projects. I don’t know that’s necessarily
what I’m doing here. But there’s a point at
which you just have to say, yes, you can structure it
in many, many different ways and you can have
interconnections between lots of
different components and lots of different elements. But you’ll never
get there unless you decide what your main
salient entry points are. And that’s kind of
what you have to do. And that’s actually what we
saw with the poetry project too, which is like those
three main entry points. It’s like, the
theory, the poetry, I think it was the group
of people themselves. Those are three
main entry points. Those are the three main paths. And then within those paths,
you can kind of cut across. But I think with the
digital in general, because most scholars
are used to working in these kind of
linear cumulative ways, there is this initial
point at which they realize that if
they can build this out any way that they want,
they have no idea what they’re going to do. And so that’s where people like
me and many of you, I assume, come in. And you just kind of have to
break that down and show them that, what are your
main entry points and then where do
you go from there. And of course,
it’s all iterative. Their main entry
points might change, but they have to
start somewhere. Actually, that is what I
was, I think, doing here. This is what a tab
looked like and this is what a series
of path looks like, and this could be
tied to all this but you’ve got to
start somewhere. OK. Very quickly. It says I’ve been doing this
for an hour and 21 minutes. I think that’s standard. BRIAN CROXALL: [INAUDIBLE] CURTIS FLETCHER: OK, quickly. We also have an API. And we did this
because we did want to allow scholars
who wanted to create all kind of vectors
[INAUDIBLE] pieces to do that if they wanted to. We’ve had some people
use it, not a lot. So what that means is basically,
anyone can create a custom skin and use Scalar as the
back end database. So in other words,
if someone wants to utilize or take advantage
of that flat ontology– because there are a lot of
databases out there– but if want to take advantage of
that particular flat ontology and use it to power some
piece of scholarship, they can do that. This was not a flash. But the idea is that– I should have done it live–
but basically, there’s this line and you get to move the
cursor kind of frustratingly. And as you move it, it will open
up portions of this huge mural. And within there,
you can grab tags which tell you something
about that particular tag. And I don’t even know
what this screenshot is. [INAUDIBLE] Yeah. OK, so then quickly,
future development. So something that may be
of use to some of you. We are in an NEH grant right
now to build out copy editing functionality in Scalar. So I know some of
you are involved in trying to output
born digital monographs. And so I am assuming
some of you have come up against the problem
of copy editing. So all the stuff in Scalar that
has been published by a press to date– this is true just for us, it’s
true for Cheryl Ball at Vega, from what I understand– the way it is copy edited
is that it is cut and paste into Word and then copy
edited with track changes and then cut and paste it
back into the platform. So the scholar ends up
becoming the typesetter. So not the best workflow, right? So we decided that this
was one of the pain points because frankly, presses
have not taken up born digital multimodal
scholarship the way that we had hoped
some years ago. And so we think that
this is potentially one of the pain points, right? So in the next eight months,
we will be building out an actual workflow so that each
individual page and scalar can have a different state,
edit, edit review, clean. So I won’t go into
what all those are, but you can imagine
what they are. And so each page within
the WYSIWYG editor will allow you to copy, edit,
and track all those changes. And then once all
those edits and author queries have been
cleared, the page manually be set to a new state. And then once everything
is in a cleared state, you can then actually
publish it live. We have a number of
people waiting on this, but development slow
now, especially when you have a small team. OK and then the other thing. So two big things we’re
working on right now. One points towards the presses
and the other one points towards the archives. So the importer that I showed
you before many, many slides ago is let’s say, not robust. Basically, you just keyword
search, you give an archive, it hits their API and
it spits out the results and you can grab
media from there. We’re building out a much
more robust archive importer, which will basically allow
you to do a couple of things. One, create a profile for
a given type of archive. So like, if the
archive is in CONTENTdm and someone writes
a profile for that and then uploads it to this
open source [INAUDIBLE],, everyone from then on
will be able to use it and unlock all CONTENTdm
systems, so forth and so on. So these will be profiles
for given protocols that people will write. And we’ll have a bunch
already in there built in. And so basically, it
allows you to grab media from any number of archives
and create kind of playlists. It’s kind of modeled
on iTunes a little bit. So you’re able to create
playlists of your media from various archiving and
then sync those playlists with your various Scalar books. Yeah? AUDIENCE: So aren’t you
sort of limiting yourself by the things that are
publicly and freely available? CURTIS FLETCHER: Yes. AUDIENCE: I mean,
isn’t scholarship being limited in this
case to things that are publicly and freely available? CURTIS FLETCHER: If you have
your own digital assets, you can get those
into Scalar too. AUDIENCE: Right, but you may not
own the rights to those assets. CURTIS FLETCHER: Yeah. In other words,
let’s say that you’re citing some article
that is behind a paywall and you want to cite it– yes, Scalar is not going to
circumvent those permission levels, absolutely. We were in talks a
while with the Getty in particular about building
out some sort of way for Scalar to track permission levels so
that scholars could use media that they have access to. And we have a million things
that we’re working on. And we haven’t
gotten to that yet. But it’s definitely something
that we’re thinking about. The other issue is, of
course, copyrighted material. And that’s just sort of up
to the scholar and the risk that they’re willing to take
in putting the media in their. Let me put it this way.
this doesn’t prevent someone from citing a piece of media
that they don’t have access to. It just allows them to
give their readers access to the stuff that we
do all have access to. So it doesn’t preclude
them in any way from doing what they used to do. It just enables them to
do many other things. It’s not like
someone, because they don’t have access to a couple
of things, not write it overall. OK so I was just showing you
a few different views of that. And then that’s it. That’s over. And I have really
gone over time. So how long do we
have for questions? Is it just until 1:15? BRIAN CROXALL: Until about 1:30. We can take one or
two questions and then we have a but more
informal conversation and still have a
bit of food here. Harriet? AUDIENCE: Would you talk about
the relationship of Scalar with publishers? I think I may have
heard this, but maybe I didn’t understand it correctly,
that there are actually a number of publishers
that kind of have given their blessing to Scalar. So that you’re in a
partnership of sorts. CURTIS FLETCHER: So we do. So like I said, the Alliance
for Network and Visual Culture has a series of partners. Some of them are archives
and some of them are presses. Those partners can vary in
their level of commitment. So like technically, NYU
Press is one of our partners because they did Mattel’s piece. But again, we’re
sort of hands off. UC Press has been a
really great partner. So not only have they published
a number of things in Scalar to date, but we’re also actually
working with them on the copy editing functionality. They’ve been really committed
to publishing this stuff. Not only to it being open
access but to it not necessarily being housed on their servers. Like all the sorts of caveats
that presses normally have, they are not as wary of. So they’ve been a great partner. Stanford recently. Because of this Mellon Grant. So as soon as Mellon decided
they were going to publish, wanted a digital
scholarship, and they kind of went out into the wild
looking for this stuff, what they found was that a fair
amount of it was in Scalar. So we did not have a formal
relationship with them as party original, ABC. But we have ended up working
with them very closely. And they they’re really open to
publishing not just our stuff but they publish a
really beautiful piece called Enchanting the Desert. If you have a chance to look
at it, it’s really wonderful. It’s just a beautiful series
of essays with lots of media and interesting connections
between those essays. AUDIENCE: Are there
people either conceptually from the Scalar
team, how you all think of Scalar, or people
using it in interesting ways– are people using it to create
living books or documents? Or is it typically you publish
your project in your book and it stays static forever? CURTIS FLETCHER: So I would
say most people are publishing something that they feel
like they can then ignore, like they would a book– which by the way is
a liability, but we do have people who want to use
it to create a living document. So a perfect example–
we’re actually working with LAUSD right now. You guys aren’t from
the west coast– Los Angeles Unified
School District. I think it’s the biggest school
district in the whole United States. Anyway, it’s massive. It’s like 700,000 students. Their ITI– so their
instructional technology division– was recently
charged with writing a report on how to
better integrate technology in the classroom. And they want that
report to be in Scalar for a couple of reasons. One, because they are advocating
for more interactivity and so they want the form to
kind of match the argument. But also, because they want
it to be a living document. And so what they’ve done is– and one thing I didn’t
mention I probably should have– not only
can you do anything to anything else in
Scalar, but also, you can repurpose anything in Scalar. So one annotation or one page
can live on multiple paths, right? So what they’re
planning on doing is they’re going to
make recommendations. And that will be
like a document. But from then on, they’ll be
able to add to it by bringing in new people from the
district who will then take portions of it and
kind of recombine and remix, given their context and
where they’re working from. AUDIENCE: Is there a
versioning system built in or something where you could
see who is doing the changes and when they’re making them? CURTIS FLETCHER: Well,
there is a versioning system in the sense that every version
is saved in the database. So every time you
click Save on anything, that version is saved
so you can revert. But getting access
to those versions– you can get access to the
versions for individual pages. You can also get access to each
individual user’s latest edits. But the overall sense, meaning
like the actual addition– not yet. But that is part of the
copy editing functionality. So the last output for
the copy editing thing is this is going to
be published, right? This is a edition one. And then any changes
that are made from then can be edition two. So then you could see the actual
difference between the two. But there’s no way to
visualize the differences between those two versions. But there will be a way to
set one aside [INAUDIBLE].. AUDIENCE: I’ve got two Scalar
projects and three in mind and applied for funding
for one of my projects. One of the questions
you get from funders is, is this going to remain
viable in the future? Books they think of
as lasting forever, but they’re worried
about digital. So I’m just curious how many
of the projects that Scalar has now globally
are on your server and how many are on independent
institution servers. And what is the future
of the Scalar server? CURTIS FLETCHER: OK, so lots
of different questions there. So I’ll start with our
server versus other servers. So two years ago, we were
kind of the only Scalar. We had a couple of
different installations, but basically, it was just
the one Scalar installation everyone uses. It’s open source. You can download it on GitHub. If you’re
technologically minded, you can do that and mess with
all the configuration files and so forth. I can’t even do it, by the way. So a couple years ago, maybe
a few people had done that. But then Reclaim Hosting– maybe you know– basically
made it one of their one click installs. And since then, the thing
is I actually don’t know. Anecdotally, I imagine
that there is maybe 50 Scalar installs right now. More and more I’m
noticing all the time, particularly because
people email me saying, can you reset my password. And then I go to their
page and I’m like, oh actually, I can’t
reset your password. It’s actually becoming
a problem because people will install Scalar and
then the person who installs it just kind of leaves. And it’s in this no
man’s land, so we’re trying to figure that out. So a lot of different
Scalar installs. So in terms of the
sustainability of our Scalar install. So it’s housed at USC on some
Enterprise level servers. It’s backed up every
day both locally and at the Arizona
disaster relief recovery– whatever it is. So if Los Angeles is nuked,
your Scalar project is fine. So in terms of sustainability,
we have an agreement. It’s going to be on USC
servers in the long term. So there’s a couple of questions
wrapped in sustainability. There’s the actual data. That will be around in
the long term, absolutely. Now only that, but
you can actually export the data if you want. So you can export it
as RDF JSON or MXL RDF. And you can back it up and
keep it on your computer. But of course,
that data will have to be expressed in some way. It will either have to be
expressed via the application or the platform itself. The platform, because
we’re in a unique position where we actually do a
contract with presses– so in other words, there are
a lot of platforms out there, we’re one that actually
has press partners. So because of that, we actually
have a commitment to them to be around in the long term. And we have said so. So 10 years, 20 years down
the road, will we still have a team that’s
updating the platform? Our answer is we’re
very committed to that being the case. But you know you never know. But it is open source
and anyone can update it. Let’s say for whatever
reason, Los Angeles was nuked and the Scalar team went
away, you could, I think, fairly easily find
someone to update your installation of Scalar
and your publication with Duke Press 10 years down the
line, if for whatever reason, it became not operable
with Firefox 2000. BRIAN CROXALL: Please
join me in thanking Curtis for the
[INAUDIBLE] conversation.

One thought on “Scalar: Writing Digital Scholarship with Curtis Fletcher

Leave a Reply

Your email address will not be published. Required fields are marked *