WEBVTT
Kind: captions
Language: en

1
00:00:00.459 --> 00:00:02.700
>> Tom Scheinfeldt: Okay, we'll get started.

2
00:00:02.700 --> 00:00:06.670
I think we're running already a little bit
behind time, so I don't want to take too much

3
00:00:06.670 --> 00:00:09.680
time away.

4
00:00:09.680 --> 00:00:13.320
I'm Tom Scheinfeldt from the University of
Connecticut, and I'm very pleased to be here

5
00:00:13.320 --> 00:00:19.810
today introducing this great panel.

6
00:00:19.810 --> 00:00:28.599
On the screen now is a Twitter conversation
that I had last evening in the airport with

7
00:00:28.599 --> 00:00:34.850
one of our panelists, and I was thinking about
the meanings of interoperability for this

8
00:00:34.850 --> 00:00:41.910
panel, and James tweeted back that, he said,
um, I've got an interoperability power point,

9
00:00:41.910 --> 00:00:47.129
if needed, but I think I'm supposed to talk
about sustainability, and I tried to reassure

10
00:00:47.129 --> 00:00:52.780
him, tweeting back saying, nope, I'm thinking
about interoperability as a component of sustainability,

11
00:00:52.780 --> 00:00:58.219
and he wrote back, that almost makes it sound
as if wasting scarce resources isn't a good

12
00:00:58.219 --> 00:01:07.030
idea, and I told him that he just gave me
my opening line, so there it is, but I think

13
00:01:07.030 --> 00:01:14.990
there's a serious truth in this joking exchange,
and I think we heard some of that in the last

14
00:01:14.990 --> 00:01:20.750
panel, that interoperability, I think, really
is our clearest path to a sustainable cultural

15
00:01:20.750 --> 00:01:27.119
heritage and public information infrastructure,
and, so, whether that interoperability is

16
00:01:27.119 --> 00:01:32.899
about federated search and discovery or linked
open data or open source software or consortial

17
00:01:32.899 --> 00:01:38.680
revenue models, the things we'll hear from
our panelists today, building the national

18
00:01:38.680 --> 00:01:43.200
digital platform is ultimately about sustaining
local investments, and I think that's where

19
00:01:43.200 --> 00:01:48.729
we get to some of the concerns that were voiced
in the question and answer in the last period,

20
00:01:48.729 --> 00:01:53.189
the critique, I think, that's been levied
against this idea of the national digital

21
00:01:53.189 --> 00:02:00.509
platform is that it runs the risk, as I think
Amy and Brewster both hinted at, of isolating

22
00:02:00.509 --> 00:02:09.450
and alienating local institutions and neglecting
their needs in pursuit of some larger national

23
00:02:09.450 --> 00:02:18.370
all encompassing vision and program, but I
think that critique will only prove correct

24
00:02:18.370 --> 00:02:24.530
if we focus too much emphasis on the first
word in the program, the national, and not

25
00:02:24.530 --> 00:02:31.400
enough emphasis on the third word, platform,
and ultimately, the national digital platform

26
00:02:31.400 --> 00:02:38.930
will succeed only if by creating an interoperable
ecosystem, we heard that word, ecosystem of

27
00:02:38.930 --> 00:02:45.120
tools, serves, data and organizations, that
is a platform, only if that platform allows

28
00:02:45.120 --> 00:02:55.560
local institutions to avoid the kind of giant
resource suck of digital services and tools,

29
00:02:55.560 --> 00:03:00.350
services and tools that could be better provided
at a distance, and instead focus their resources,

30
00:03:00.350 --> 00:03:05.060
their energy, their staff expertise on the
things that can only be done locally.

31
00:03:05.060 --> 00:03:11.150
So, I don't think it's an either/or, I think
it's allowing local institutions to do what

32
00:03:11.150 --> 00:03:17.870
they do best and allowing a national ecosystem
or shared ecosystem, and maybe national is

33
00:03:17.870 --> 00:03:22.900
the problem word there, maybe we want to talk
more about a shared infrastructure or shared

34
00:03:22.900 --> 00:03:29.810
platform or shared ecosystem, allowing that
shared ecosystem to do the work that can be

35
00:03:29.810 --> 00:03:38.760
done by us all, and, so, it really is an issue
of sustainability.

36
00:03:38.760 --> 00:03:44.200
It's about allowing local institutions to
use their resources best and not duplicate

37
00:03:44.200 --> 00:03:45.200
effort.

38
00:03:45.200 --> 00:03:49.900
For digital programs, that may be really outside
of their scope, outside of their resources,

39
00:03:49.900 --> 00:03:51.260
outside of their expertise.

40
00:03:51.260 --> 00:03:56.770
So, I think that keeping our eye on this most
important goal of the national digital platform,

41
00:03:56.770 --> 00:04:02.590
kind of local cultural sustainability, is
our real challenge here today, and I think

42
00:04:02.590 --> 00:04:07.010
that was voiced very well in the last session.

43
00:04:07.010 --> 00:04:14.870
And so, we're going to hear from each of our
panelists this morning, their responses to

44
00:04:14.870 --> 00:04:19.600
that challenge and how they're meeting those
challenges in different ways.

45
00:04:19.600 --> 00:04:21.590
So, let me introduce them quickly.

46
00:04:21.590 --> 00:04:26.050
I was told just to do names and titles, so
I'll do that.

47
00:04:26.050 --> 00:04:31.319
Dan Cohen, immediately to my right, is executive
director of the Digital Public Library of

48
00:04:31.319 --> 00:04:32.319
America.

49
00:04:32.319 --> 00:04:35.620
I think everyone here probably knows what
that does.

50
00:04:35.620 --> 00:04:41.960
Evelyn McLellan is president of Artifactual
Systems which develops open source archival

51
00:04:41.960 --> 00:04:46.310
products including Archive Matica and Atom.

52
00:04:46.310 --> 00:04:51.849
Sibyl Schaefer is digital preservation analyst
at the University of California San Diego,

53
00:04:51.849 --> 00:04:57.270
and until recently, was head of digital programs
at the Rockefeller Archive Center, and James

54
00:04:57.270 --> 00:05:03.509
Shulman is president of Art Store, an organization
that provides content, tools, services to

55
00:05:03.509 --> 00:05:08.409
over 1500 colleges, universities and other
institutions around the world.

56
00:05:08.409 --> 00:05:12.979
So, with that, I'll let this panel get started
and turn it over to Dan.

57
00:05:12.979 --> 00:05:14.379
>>Dan Cohen: Thanks so much, Tom.

58
00:05:14.379 --> 00:05:16.729
It’s god to be here with Tom.

59
00:05:16.729 --> 00:05:22.100
Last week, the New Yorker had an article on
NASA's preparation for a human landing on

60
00:05:22.100 --> 00:05:26.180
Mars, which is supposed to take place sometime
in the 2030's, hopefully I’ll still be alive

61
00:05:26.180 --> 00:05:30.169
and it's, of course, a huge undertaking, but
what the article focused on was what it would

62
00:05:30.169 --> 00:05:35.380
take to get a half dozen humans there and
back, and this is the important point, without

63
00:05:35.380 --> 00:05:36.560
killing each other.

64
00:05:36.560 --> 00:05:42.169
Now, of course, you know, two decades from
now, many things will change.

65
00:05:42.169 --> 00:05:46.349
Engineering will change over the next two
decades, but I thought what NASA astrobiologist

66
00:05:46.349 --> 00:05:51.039
Christopher McKay said was really important,
that technology is surely going to change,

67
00:05:51.039 --> 00:05:56.029
but there will be the same humans as we have
now, and that struck me as extremely profound.

68
00:05:56.029 --> 00:06:00.729
The engineering will change, the technology
will change, but the human beings will remain

69
00:06:00.729 --> 00:06:02.050
the same.

70
00:06:02.050 --> 00:06:06.360
You know, thanks to IMLS, we stand here talking
about the national digital platform, which

71
00:06:06.360 --> 00:06:07.730
is really a transformative idea.

72
00:06:07.730 --> 00:06:13.389
I mean, I think it is, in fact, the Martian
landing for libraries, archives and museums

73
00:06:13.389 --> 00:06:14.389
in the United States.

74
00:06:14.389 --> 00:06:18.830
It's a very powerful and exciting idea, and
the technology that relates to that platform

75
00:06:18.830 --> 00:06:23.900
will indeed change over time, but the humans
will remain the same, and it's a belief that

76
00:06:23.900 --> 00:06:28.340
I really do wake up with every day working
at the Digital Public Library of America which

77
00:06:28.340 --> 00:06:33.129
I always say is as much a social construction
as it is a technical one.

78
00:06:33.129 --> 00:06:34.590
We launched two years ago.

79
00:06:34.590 --> 00:06:41.539
We just celebrated our second birthday, and
we're working really at scale across the country.

80
00:06:41.539 --> 00:06:46.669
We've made tremendous progress in the engineering
and in the collaboration.

81
00:06:46.669 --> 00:06:52.979
We just surpassed 10 million openly available
items from 1600 contributing institutions,

82
00:06:52.979 --> 00:06:58.419
so we're really at a very gratifying point
here two years in, but I think what's most

83
00:06:58.419 --> 00:07:01.480
gratifying about it is in fact some of these
human factors.

84
00:07:01.480 --> 00:07:04.099
I think collaboration is incredibly hard.

85
00:07:04.099 --> 00:07:09.120
It's hard for us not to kill each other and
to be oppositional with each other.

86
00:07:09.120 --> 00:07:14.830
You know, James Neal cautioned against polygamy,
I believe, on the last panel, and what I would

87
00:07:14.830 --> 00:07:25.410
say is DPLA isn't polygamist, but we are polyamorous,
and, so, we want to, in fact, engage a community

88
00:07:25.410 --> 00:07:27.499
and have community input.

89
00:07:27.499 --> 00:07:32.469
DPLA fast in Indianapolis last week, we had
350 people there, librarians, museum professionals,

90
00:07:32.469 --> 00:07:38.229
archivists, yes, archivists were there, we
had members of the public, historians, scholars,

91
00:07:38.229 --> 00:07:42.930
so many different kinds of people there, and
I think that that community, the fact that

92
00:07:42.930 --> 00:07:48.939
this exists now in 2015 is, I think, in fact,
the most important thing that DPLA has been

93
00:07:48.939 --> 00:07:52.490
able to substantiate.

94
00:07:52.490 --> 00:07:56.090
Having said that, of course there are important
engineering challenges, and there are things

95
00:07:56.090 --> 00:07:58.909
related to the hubs that are important.

96
00:07:58.909 --> 00:08:02.669
Excuse me, related to DPLA that are important
in terms of the model.

97
00:08:02.669 --> 00:08:05.310
We do have, in fact, a decentralized model.

98
00:08:05.310 --> 00:08:10.240
We are not consolidating to use Brewster Kale’s
ominous word.

99
00:08:10.240 --> 00:08:15.409
We're instead creating, I think Tom’s word,
an ecosystem, where we have hubs across the

100
00:08:15.409 --> 00:08:21.710
United States, especially state based service
hubs, that act as little mini DPLA's in their

101
00:08:21.710 --> 00:08:23.770
state to bring online the small collections.

102
00:08:23.770 --> 00:08:28.189
We don't want to forget about the small places,
as Tom noted, but to bring them all together,

103
00:08:28.189 --> 00:08:33.640
and we've been very fortunate to get funding
from IMLS and National Endowment to the Humanities

104
00:08:33.640 --> 00:08:38.910
and the Sloan Foundation and the Knight Foundation
and others to spin up these local hubs so

105
00:08:38.910 --> 00:08:43.530
we can in fact have a collaborative decentralized
model that all works together.

106
00:08:43.530 --> 00:08:48.540
So, we're pleased with the progress, but there's
clearly much more to do, and I think, here,

107
00:08:48.540 --> 00:08:50.000
engineering is important.

108
00:08:50.000 --> 00:08:55.980
You know, as astronauts are fond of saying,
leaky spaceships are sub optimal.

109
00:08:55.980 --> 00:08:57.350
They don't say that, but I'm saying that.

110
00:08:57.350 --> 00:08:59.500
I’m sure they say that at some point.

111
00:08:59.500 --> 00:09:03.600
So, how can we make the engineering, I guess,
more safer?

112
00:09:03.600 --> 00:09:09.620
Safer for humans, more easy to use so that
people don't in fact kill each other?

113
00:09:09.620 --> 00:09:12.990
Here, I think what we're trying to do at least
here at DPLA is many of the things that have

114
00:09:12.990 --> 00:09:14.500
already been laid out on the table.

115
00:09:14.500 --> 00:09:21.240
We want to make sure that there are parallel
systems in the ecosystem, we want people to

116
00:09:21.240 --> 00:09:26.110
have local variations, but we want them to
work together and to be interoperable.

117
00:09:26.110 --> 00:09:28.160
We want to make the work flows better.

118
00:09:28.160 --> 00:09:32.280
Mark Matienza, our director of technology,
is here, and he and his team are really working

119
00:09:32.280 --> 00:09:38.100
on streamlining all the processes that relate
to digital platforms, how the data gets interchanged

120
00:09:38.100 --> 00:09:42.540
with each other, how we normalize data across
multiple institutions.

121
00:09:42.540 --> 00:09:44.700
We need to have that all work together.

122
00:09:44.700 --> 00:09:47.570
Internationally, we want that to all work
together.

123
00:09:47.570 --> 00:09:49.120
We also want things to be modular.

124
00:09:49.120 --> 00:09:53.270
A lot of our code bases are extremely modular,
where people can make changes, can work on

125
00:09:53.270 --> 00:09:57.130
little bits of it, but again, have it work
together in a very webby way.

126
00:09:57.130 --> 00:10:01.890
Again, not consolidation but collaboration
in an ecosystem.

127
00:10:01.890 --> 00:10:03.450
Others in the room are doing important work.

128
00:10:03.450 --> 00:10:07.380
I'm delighted to see so many collaborators.

129
00:10:07.380 --> 00:10:10.260
In fact, most of the people in the room I
think we are collaborating with.

130
00:10:10.260 --> 00:10:15.830
If you're not collaborating with us, dan@dpla
you can send me an e mail, and we're excited

131
00:10:15.830 --> 00:10:20.700
to work with you more in the future to share
our cultural heritage in a maximal way with

132
00:10:20.700 --> 00:10:22.960
the public.

133
00:10:22.960 --> 00:10:24.090
Thanks.

134
00:10:24.090 --> 00:10:32.870
>>Evelyn McLellan: Okay, yes, I'm Evelyn McLellan
from Artifactual Systems, and my background

135
00:10:32.870 --> 00:10:38.210
is in the open source software development
community, specifically archive managment

136
00:10:38.210 --> 00:10:43.510
so I'm going to be speaking a lot from my
own experience.

137
00:10:43.510 --> 00:10:48.390
The last few years have seen phenomenal growth
in the number and sophistication of open source

138
00:10:48.390 --> 00:10:53.320
software tools that are available for libraries
and archives, and I realized when I started

139
00:10:53.320 --> 00:10:57.260
preparing for this meeting that I actually
have a very hard time keeping a handle on

140
00:10:57.260 --> 00:11:03.730
everything that's out there, so to help me
prepare, I went to the excellent FOS for Live

141
00:11:03.730 --> 00:11:10.860
site, that free and open source software hosted
by Nerasis which is dedicated to listing available

142
00:11:10.860 --> 00:11:14.670
open source software tools and providing updates
on releases.

143
00:11:14.670 --> 00:11:20.530
That site lists a total of 223 open source
software tools for use by libraries, which

144
00:11:20.530 --> 00:11:26.350
is an astonishing number, and from my perspective,
as an Archive Matica project participant,

145
00:11:26.350 --> 00:11:31.160
there are actually a lot more open source
tools out there that we integrate into the

146
00:11:31.160 --> 00:11:36.460
Archive Matica suite that aren't necessarily
for stand alone use in libraries and archives.

147
00:11:36.460 --> 00:11:42.760
So, this is an astonishing number, but it
made me think, you know, how many institutions

148
00:11:42.760 --> 00:11:47.010
are actually successfully preserving their
digital holdings and making them available

149
00:11:47.010 --> 00:11:51.730
online, and I'm thinking libraries and archives
here.

150
00:11:51.730 --> 00:11:55.990
I may sound like I'm about to provide an answer
to this, but actually, I'm not.

151
00:11:55.990 --> 00:12:02.860
I have no idea, and I'm not sure exactly how
I would find out, and I guess a related question

152
00:12:02.860 --> 00:12:08.670
is this; for those institutions successfully
undertaking these tasks, what percentage of

153
00:12:08.670 --> 00:12:13.170
their digital holdings are they managing to
preserve and provide access to, and my belief

154
00:12:13.170 --> 00:12:17.920
based on anecdotal evidence is that there's
an awful lot of digitized content that is

155
00:12:17.920 --> 00:12:23.920
not adequately preserved and that born digital
content is being acquired at a sometimes alarming

156
00:12:23.920 --> 00:12:27.610
rate, but is neither being preserved or made
accessible.

157
00:12:27.610 --> 00:12:33.940
So, I've been asked to comment on what makes
for a successful open source project or service,

158
00:12:33.940 --> 00:12:36.740
and here, I'm going to fall back on my personal
experience a little bit.

159
00:12:36.740 --> 00:12:42.012
The first thing is functionality, which sounds
like a no brainer, the software's got to work

160
00:12:42.012 --> 00:12:43.950
and it's got to fill a need.

161
00:12:43.950 --> 00:12:48.060
Speaking as someone in software development,
I can tell you it’s actually very difficult

162
00:12:48.060 --> 00:12:54.190
to figure out what exactly it is the users
need and what the software should do.

163
00:12:54.190 --> 00:12:55.190
It's not that easy.

164
00:12:55.190 --> 00:13:01.220
The second thing is a long range plan for
supporting and continuing to develop the software,

165
00:13:01.220 --> 00:13:05.280
and I think in the past, we've seen some grant
funded projects where there's been a tool

166
00:13:05.280 --> 00:13:10.350
produced, but when the grant ends, the project
kind of goes away.

167
00:13:10.350 --> 00:13:14.870
Software requires an unbelievable amount of
TLC just to keep going.

168
00:13:14.870 --> 00:13:21.190
The third thing is community support and documentation
and the importance and enormity of this task

169
00:13:21.190 --> 00:13:23.380
cannot be understated.

170
00:13:23.380 --> 00:13:27.450
At Artifactual, I find that we're often just
a little bit behind on our documentation that

171
00:13:27.450 --> 00:13:31.290
we provide for our software, and sometimes
a lot behind.

172
00:13:31.290 --> 00:13:35.750
We put all of our end user documentation and
all of our technical documentation online,

173
00:13:35.750 --> 00:13:39.519
and it requires a great deal of effort just
to keep that up to date.

174
00:13:39.519 --> 00:13:44.730
We also run two public discussion lists, one
for Archive Matica and one for Atom, and we

175
00:13:44.730 --> 00:13:50.640
try to encourage users of these lists to answer
each other's questions, but we find that we

176
00:13:50.640 --> 00:13:53.190
end up answering quite a few of the questions.

177
00:13:53.190 --> 00:13:57.110
Users come there for technical support or
to ask questions about functionality, and

178
00:13:57.110 --> 00:14:02.260
we end up answering the questions, and particularly
when it comes to Archive Matica, which it's

179
00:14:02.260 --> 00:14:06.050
kind of a technical and esoteric field.

180
00:14:06.050 --> 00:14:10.920
The fourth thing that a successful project
requires is the implementation of standards

181
00:14:10.920 --> 00:14:15.370
so that the content being preserved and made
accessible can be moved between systems or

182
00:14:15.370 --> 00:14:19.570
integrated into link data platforms and initiatives.

183
00:14:19.570 --> 00:14:23.810
We had one institution tell us they didn't
actually care that Archive Matica was open

184
00:14:23.810 --> 00:14:28.260
source, what they cared about was the premise
and implementation and the use of the Library

185
00:14:28.260 --> 00:14:36.000
of Congress specification so, this kind of
brings me to my fifth point, which is integration

186
00:14:36.000 --> 00:14:40.830
of systems, and we've started to talk a little
bit about that.

187
00:14:40.830 --> 00:14:47.070
No single software tool does everything, and
just because, I should add, just because a

188
00:14:47.070 --> 00:14:51.120
software tool is open source doesn't necessarily
mean it's interoperable with other systems,

189
00:14:51.120 --> 00:14:52.120
open source or not.

190
00:14:52.120 --> 00:14:55.440
It actually takes a lot of effort to make
two systems talk to each other.

191
00:14:55.440 --> 00:15:00.360
We're doing a project currently with the University
of Michigan, the Bentley Library there, to

192
00:15:00.360 --> 00:15:04.610
integrate Archive Matica with Archive Space
and D Space and one of the things we're very

193
00:15:04.610 --> 00:15:11.000
heavily focused on for that project is making
sure that the tools integrate not just in

194
00:15:11.000 --> 00:15:15.660
a way that meets their processing needs, but
in a way that can be used by other institutions,

195
00:15:15.660 --> 00:15:19.780
and they're working very hard to talk to other
institutions and find out, you know, how would

196
00:15:19.780 --> 00:15:25.340
you use archive space and Archive Matica together,
what would you be looking for, and finally,

197
00:15:25.340 --> 00:15:29.420
my last point is that there are two kinds
of institutions out there, and I'm about to

198
00:15:29.420 --> 00:15:30.500
get a red flag here.

199
00:15:30.500 --> 00:15:34.360
There are two kinds of institutions out there,
and I guess this is my take away point.

200
00:15:34.360 --> 00:15:39.550
There's institutions that are big and can
afford to put developer resources into supporting

201
00:15:39.550 --> 00:15:43.980
and developing open source tools, integrating
them themselves, but there are a lot of institutions

202
00:15:43.980 --> 00:15:49.450
that just need to be consumers of these services
and tools, and we have to focus efforts on

203
00:15:49.450 --> 00:15:54.400
both the large institutions and the smaller
institutions that need these tools.

204
00:15:54.400 --> 00:15:56.080
>>Sibyl Schaefer: Hi.

205
00:15:56.080 --> 00:15:59.460
Can you put the—yes, thank you.

206
00:15:59.460 --> 00:16:04.530
As Tom mentioned, I'm currently working at
the University of California San Diego, which

207
00:16:04.530 --> 00:16:06.700
as you may know, is a very large institution.

208
00:16:06.700 --> 00:16:10.000
That's a fairly recent development.

209
00:16:10.000 --> 00:16:15.320
I was asked to join this conversation while
I was the head of digital programs of the

210
00:16:15.320 --> 00:16:20.380
Rockefeller Archive Center, which is a very
small independent archive located about 20

211
00:16:20.380 --> 00:16:25.560
miles north of Manhattan and it's from that
smaller organizational perspective that I'll

212
00:16:25.560 --> 00:16:29.100
be kind of representing today.

213
00:16:29.100 --> 00:16:35.670
As head of digital programs, I implemented
many software programs to aid in our care

214
00:16:35.670 --> 00:16:41.100
and management of archival materials, digital
archival materials in particular, and our

215
00:16:41.100 --> 00:16:43.370
work flow looks something like this.

216
00:16:43.370 --> 00:16:49.260
We would receive the materials, whether on
some sort of media in which we would image

217
00:16:49.260 --> 00:16:56.230
it or through data transfer, there was descriptive
processes that were undertaken, including

218
00:16:56.230 --> 00:17:04.429
identifying potentially restricted material
and personal identifiable information.

219
00:17:04.429 --> 00:17:10.500
The materials were then packaged for preservation,
distributed geographically for redundancy,

220
00:17:10.500 --> 00:17:15.269
and then provided access to whether that be
online or on site.

221
00:17:15.269 --> 00:17:19.179
I brought this diagram up not so much to dig
into the details but to show that there's

222
00:17:19.179 --> 00:17:24.480
over seven different systems that we use to
process records from excess to access.

223
00:17:24.480 --> 00:17:29.980
We chose many of these systems mainly because
they had the specific functionality we needed,

224
00:17:29.980 --> 00:17:34.019
and we also favored ones that were either
more user friendly or could be configured

225
00:17:34.019 --> 00:17:36.789
to be user friendly.

226
00:17:36.789 --> 00:17:41.559
Being an agency dedicated to preserving the
study of philanthropy, open source systems

227
00:17:41.559 --> 00:17:46.529
always jived more with our ethos, however,
if proprietary software had the functionality

228
00:17:46.529 --> 00:17:53.139
we needed, we weren't precluded from adopting
it, and it's worth noting that these are proprietary

229
00:17:53.139 --> 00:17:58.070
systems, and there’s a couple on the diagram
are expensive and thus not readily available

230
00:17:58.070 --> 00:18:01.190
to organizations with smaller budgets.

231
00:18:01.190 --> 00:18:05.760
The most important aspect of this diagram
is really the arrows connecting the systems.

232
00:18:05.760 --> 00:18:08.250
This is where we spent most of our time.

233
00:18:08.250 --> 00:18:13.570
Interoperability, and more specifically, the
ability to control data flow from one application

234
00:18:13.570 --> 00:18:18.919
to another is an area of high need as has
already been stated here today.

235
00:18:18.919 --> 00:18:25.749
This diagram obscures a ton of work in between
each point, some of which we were able to

236
00:18:25.749 --> 00:18:28.340
automate, some of which we were not.

237
00:18:28.340 --> 00:18:31.950
This type of convoluted work flow is not unique
to the archive center.

238
00:18:31.950 --> 00:18:37.850
One tweet from the IMLS focus Twitter conversation
the other week sums it up as there's no sensible

239
00:18:37.850 --> 00:18:44.289
way to do accessing, digitization, and preservation
redaction.

240
00:18:44.289 --> 00:18:48.840
Another aspect of this diagram I'd like to
point out is the one way work flow.

241
00:18:48.840 --> 00:18:54.240
Revisioning is so nightmarish that we haven't
yet had a use case that passed the threshold

242
00:18:54.240 --> 00:18:55.789
where it was warranted.

243
00:18:55.789 --> 00:19:00.360
This will change, however, once some of our
time based access embargoes are lifted and

244
00:19:00.360 --> 00:19:05.299
the materials need to be made accessible.

245
00:19:05.299 --> 00:19:11.019
The diagram also doesn't include our subscription
to archive it and to the web data we collect,

246
00:19:11.019 --> 00:19:14.360
mainly because it hasn't yet been figured
into the work flow, but I think it's important

247
00:19:14.360 --> 00:19:18.320
to recognize that the data is coming in from
a lot of different sources.

248
00:19:18.320 --> 00:19:22.750
So, addressing the questions that we were
given prior to this meeting when talking about

249
00:19:22.750 --> 00:19:28.340
OSS, and taking it to the next level, interoperability
is number one on my wish list.

250
00:19:28.340 --> 00:19:32.470
I think IMLS could lead the way in bringing
together the organizations that are working

251
00:19:32.470 --> 00:19:37.639
with OSS in order to learn about each other,
foster collaboration and innovation and discuss

252
00:19:37.639 --> 00:19:40.940
the next step in bringing these systems more
in sync.

253
00:19:40.940 --> 00:19:44.999
The other thing I wanted to highlight was
how I determined the division of resources

254
00:19:44.999 --> 00:19:47.400
amongst all the programs we use.

255
00:19:47.400 --> 00:19:51.409
My staff was pretty small, there were four
of us in total, but I had a great deal of

256
00:19:51.409 --> 00:19:56.429
flexibility in allocating funds for software
and software services.

257
00:19:56.429 --> 00:19:59.210
One of my goals was to build staff technical
competency.

258
00:19:59.210 --> 00:20:04.600
So, in deciding between allocating time or
allocating money, a large part of that decision

259
00:20:04.600 --> 00:20:09.750
was influenced by the ecosystem surrounding
the OSS and how its development community

260
00:20:09.750 --> 00:20:11.769
supported newbies.

261
00:20:11.769 --> 00:20:16.659
The software projects that promoted the most
support and active encouragement from my staff

262
00:20:16.659 --> 00:20:20.360
were the ones that we chose to actively work
with and adopt.

263
00:20:20.360 --> 00:20:24.279
Another recommendation is to provide support
for development of guidelines or best practices

264
00:20:24.279 --> 00:20:31.250
for how open source software makers should
communicate to their variant constituencies.

265
00:20:31.250 --> 00:20:34.279
Thanks.

266
00:20:34.279 --> 00:20:46.039
>>James Shulman: I am physiologically incapable
of speaking without pictures.

267
00:20:46.039 --> 00:20:51.379
I can do shadow puppets.

268
00:20:51.379 --> 00:20:56.900
I will start, maybe they can help us find
my presentation.

269
00:20:56.900 --> 00:21:03.650
What I want to do today is talk about, I'm
going to talk about, I can’t read that so

270
00:21:03.650 --> 00:21:05.210
I’m going to go to this.

271
00:21:05.210 --> 00:21:11.279
I’m going to talk about sort of the arc
of reforming isolationist.

272
00:21:11.279 --> 00:21:17.429
It's not really fair, since what we've been
doing at Art Store really has been about being

273
00:21:17.429 --> 00:21:21.261
networked all along, but I think what I'm
going to talk about are a few projects that

274
00:21:21.261 --> 00:21:29.419
we've done that show a path toward an increasingly
networked approach to things, and I think,

275
00:21:29.419 --> 00:21:36.889
so our original mission and our original goal
was to set out to connect content owners and

276
00:21:36.889 --> 00:21:41.369
educational users and create content that
would be useful to bring together and aggregate

277
00:21:41.369 --> 00:21:46.269
content that would be useful for people teaching
art history or other related fields using

278
00:21:46.269 --> 00:21:53.710
visual materials, so we found images from
archives and museums and individual scholars

279
00:21:53.710 --> 00:21:58.600
and photographers all over the world and brought
them together and made them searchable, and

280
00:21:58.600 --> 00:22:03.019
just to highlight one thing that Amy Garmer
said on the previous panel, the need to integrate

281
00:22:03.019 --> 00:22:10.179
the local into this networked approach is
a very important sociological and ethical

282
00:22:10.179 --> 00:22:14.640
issue, so I think all the things that we've
talked about today, about the diversity of

283
00:22:14.640 --> 00:22:19.179
the ecosystem, are so important, and I think
what Sibyl was just saying about the different

284
00:22:19.179 --> 00:22:23.940
pieces of this, some of these are proprietor
software, some are open, but the glue that

285
00:22:23.940 --> 00:22:27.880
we're talking about today and that IMLS is
providing such leadership in doing is seeing

286
00:22:27.880 --> 00:22:32.600
that that glue in that ecosystem needs to
all work together, and so I think that’s

287
00:22:32.600 --> 00:22:35.389
why we’ve heard the work ecosystem a couple
of times.

288
00:22:35.389 --> 00:22:40.490
So, what we did originally, so I'm going to
give three examples of how things that we've

289
00:22:40.490 --> 00:22:44.679
done at Art Store, even though everything
we do is non profit but some of them charge

290
00:22:44.679 --> 00:22:50.679
fees, how those have cross subsidized open
things and I think the increasing utility

291
00:22:50.679 --> 00:22:51.749
of those over time.

292
00:22:51.749 --> 00:22:57.330
So, the first example, Art Store digitally
library subscribed to by 1600 institutions,

293
00:22:57.330 --> 00:23:02.809
from research universities to small public
schools.

294
00:23:02.809 --> 00:23:07.350
So, through that channel where we were bringing
together content and making it available for

295
00:23:07.350 --> 00:23:12.990
educational users as a subscription service,
the metropolitan museum came to us about 5

296
00:23:12.990 --> 00:23:18.869
or 6 years ago and said we want to make images
available for academic publishing.

297
00:23:18.869 --> 00:23:23.629
For free, we want to provide images for low
print run publications, will you do the fulfillment

298
00:23:23.629 --> 00:23:26.789
of that for us, and we said absolutely.

299
00:23:26.789 --> 00:23:33.660
So, we use this platform, this channel that
we were building to deliver online educational

300
00:23:33.660 --> 00:23:43.080
uses for very constrained purposes, to open
it up to allow for this academic, this distribution.

301
00:23:43.080 --> 00:23:46.850
They're images that they wanted to give away
to no fee, and we give away for no fee and

302
00:23:46.850 --> 00:23:49.049
you don’t have to subscribe to Art Store.

303
00:23:49.049 --> 00:23:56.440
So, already, the platform of Art Store digital
library is cross subsidizing this open content.

304
00:23:56.440 --> 00:23:59.789
That's great, but to be honest, it's a little
bit of an orphan.

305
00:23:59.789 --> 00:24:04.289
It's not in the flow, it's not in a national
platform, so it's a really good thing, and

306
00:24:04.289 --> 00:24:09.559
I urge you all to go get an image for your
article for the Burlington Magazine, but it's

307
00:24:09.559 --> 00:24:11.360
not in the flow.

308
00:24:11.360 --> 00:24:15.739
The next stage, so anyway, the way that works
though is you get a password, and you go to

309
00:24:15.739 --> 00:24:19.399
Art Store, we give you a password, you get
those images, you can fill out a form and

310
00:24:19.399 --> 00:24:24.259
download a 20 megabyte TIF for an academic
publication.

311
00:24:24.259 --> 00:24:29.080
The next example we have is we've also built
a platform called Shared Shelf, which is cloud

312
00:24:29.080 --> 00:24:35.970
based cataloging and asset management services,
and so this is really built in response to

313
00:24:35.970 --> 00:24:40.869
the silos within institutions where one department
uses one software and the map department uses

314
00:24:40.869 --> 00:24:45.119
another and the art history uses another and
the archaeologist uses another, and individual

315
00:24:45.119 --> 00:24:49.830
faculty members are storing their collections
on iPhoto or wherever and these images aren't

316
00:24:49.830 --> 00:24:52.730
getting into the institutional flow or into
a shared flow.

317
00:24:52.730 --> 00:24:57.740
So, this is hopefully useful services for
colleges and universities, and there are about

318
00:24:57.740 --> 00:25:02.840
120 places using it, has a lot of more on
metadata and authority files and the whole

319
00:25:02.840 --> 00:25:08.700
vocabulary is built in, and so, a couple years
ago, Columbia University and the Abrial Library

320
00:25:08.700 --> 00:25:13.179
and Jim and his colleague Carolyn Ann Vabian
worked together with us and the Getty Research

321
00:25:13.179 --> 00:25:17.710
Institute and got an IMLS grant to build something
called the Built Works Registry.

322
00:25:17.710 --> 00:25:24.190
The Built Works Registry is an open, messy,
sloppy, beginnings of an authority file for

323
00:25:24.190 --> 00:25:29.809
Built Works, so that everybody in the world—let’s
skip this—everybody in the world isn’t

324
00:25:29.809 --> 00:25:34.659
trying to figure out is it called Falling
Water or Water, or is it health man house

325
00:25:34.659 --> 00:25:41.700
in mill run of bear run Pennsylvania and so
it was to build one record, one stable URI.

326
00:25:41.700 --> 00:25:46.480
This, of course, is open, but it lives symbiotically
on the shared shelf platform.

327
00:25:46.480 --> 00:25:53.690
So, this, too, this has now been launched
and it's up, it has an API, so it's a big

328
00:25:53.690 --> 00:25:57.919
step forward toward being in the flow.

329
00:25:57.919 --> 00:26:02.519
Where we fit vocabularies and authorities
in this national platform, I think, is still

330
00:26:02.519 --> 00:26:07.620
emerging, but this was a step for us where
that cross subsidization was now trying to

331
00:26:07.620 --> 00:26:11.239
get things more in the flow.

332
00:26:11.239 --> 00:26:14.429
So, this is BW, the Bell Worth Registry.

333
00:26:14.429 --> 00:26:24.610
So the newest stage for us was when DPLA launched
and we became a hub for it early on, and so

334
00:26:24.610 --> 00:26:28.261
we worked with some museums that wanted their
content that was in the Art Store digital

335
00:26:28.261 --> 00:26:30.850
library to get in this flow.

336
00:26:30.850 --> 00:26:35.571
The bridges that were built in those early
days weren't made out of steel, they were

337
00:26:35.571 --> 00:26:40.480
more rope ladders, so fortunately with a new
IMLS grant, we're going to be building out

338
00:26:40.480 --> 00:26:46.239
the infrastructure, and we're going to work
on standards, like Triple IF and work towards

339
00:26:46.239 --> 00:26:52.220
Hydra interoperability so that all the things,
so that this flow of content entity into DPLA

340
00:26:52.220 --> 00:26:57.799
is very much now moving more and more towards
the arc of fitting into this national ecosystem.

341
00:26:57.799 --> 00:27:04.369
>>Tom Scheinfeldt: Okay, thank you all.

342
00:27:04.369 --> 00:27:12.720
That was great, and I guess, Trevor, I'll
start the Q & A. One thing that stood out

343
00:27:12.720 --> 00:27:20.389
for me about all of your presentations was
not so much the focus on, I think as Dan setup

344
00:27:20.389 --> 00:27:24.980
very well, not so much the focus on technology,
but the focus on people, so we're talking

345
00:27:24.980 --> 00:27:30.840
about what's the best word, you know, should
we be using ecosystem, platform, cyber infrastructure,

346
00:27:30.840 --> 00:27:34.820
you know, which of these words is most appropriate.

347
00:27:34.820 --> 00:27:38.369
You know, I think we could also say that it's
people, right?

348
00:27:38.369 --> 00:27:46.820
It's like, you know, the national digital
platform is made of people, and it seems that,

349
00:27:46.820 --> 00:27:53.899
you know, Dan, you were talking about the
kind of organizational challenges of getting

350
00:27:53.899 --> 00:27:59.539
all of your partners together and working
together to form that kind of community.

351
00:27:59.539 --> 00:28:06.080
Evelyn, you were talking about the importance
of the community to provide documentation,

352
00:28:06.080 --> 00:28:13.779
support, kind of help functions, and, Sibyl,
you were talking about the kind of idiosyncrasies

353
00:28:13.779 --> 00:28:20.110
that all institutions have in their processes,
and, you know, in some respects, I think we

354
00:28:20.110 --> 00:28:24.859
want to support those kinds of idiosyncrasies,
and this gets back to some of the issues,

355
00:28:24.859 --> 00:28:28.129
I think, that we were talking about in the
last panel.

356
00:28:28.129 --> 00:28:34.609
I'm wondering if anybody has anything to say
about how, while we build this national digital

357
00:28:34.609 --> 00:28:41.600
platform, we can continue to support the kind
of idiosyncratic, because I think that is,

358
00:28:41.600 --> 00:28:49.580
I mean, the richness of our field is really
in the idiosyncrasies of individual institutions,

359
00:28:49.580 --> 00:28:55.990
of individual practitioners, of individual
collections, and, so, while we try to collaborate

360
00:28:55.990 --> 00:29:01.679
and work together and interoperate, how we
can, at the same time, support idiosyncrasy.

361
00:29:01.679 --> 00:29:05.590
>>Michael Edson: I've got an idea.

362
00:29:05.590 --> 00:29:08.760
>>Tom Scheinfeldt: Mike, you want to jump?

363
00:29:08.760 --> 00:29:11.230
Mike will jump in and answer my question.

364
00:29:11.230 --> 00:29:13.820
>>Michael Edson: See me during the break.

365
00:29:13.820 --> 00:29:15.029
I'm Michael Edson.

366
00:29:15.029 --> 00:29:19.919
I work at the Smithsonian, but I'm here for
Clear where I am a fellow in the Open Knowledge

367
00:29:19.919 --> 00:29:22.799
Foundation.

368
00:29:22.799 --> 00:29:28.049
Trevor and I have talked for years about the
process of edge to core.

369
00:29:28.049 --> 00:29:33.679
So, back in the old days, if you wanted something
cool to happen in your organization, you'd

370
00:29:33.679 --> 00:29:38.519
put down the gates and you'd do a retreat
and get the smart people in the room, and

371
00:29:38.519 --> 00:29:42.350
you'd define the problem, you’d architect
the solution, your staff would build it, and

372
00:29:42.350 --> 00:29:48.450
then you'd deliver it to a passive and grateful
public, and that doesn't work anymore, because

373
00:29:48.450 --> 00:29:56.359
kind of the perimeter of the innovation environment
is so freaking huge now, it dwarfs the scale

374
00:29:56.359 --> 00:30:02.159
of almost any institution we can think of,
so, when we architected the Smithsonian strategy

375
00:30:02.159 --> 00:30:06.580
back in 2009, we tried to turn that into an
asset.

376
00:30:06.580 --> 00:30:11.289
We said the most interesting things in any
given domain are happening in the border habitat,

377
00:30:11.289 --> 00:30:17.460
between institutions in the 20th century sense
and everybody else in the world.

378
00:30:17.460 --> 00:30:20.669
That's where it happens in nature, that's
where you get the most biodiversity, and those

379
00:30:20.669 --> 00:30:27.769
border habitats are where you have well defined
problems which normally come from people.

380
00:30:27.769 --> 00:30:33.870
Some subject matter expertise, some technological
prowess, and some level of platform with a

381
00:30:33.870 --> 00:30:41.309
lower case P, and so the process for an organization
like digital platform at IMLS or any of our

382
00:30:41.309 --> 00:30:48.139
institutions, if you want to make something
interesting happen, is to diversify your portfolio

383
00:30:48.139 --> 00:30:54.649
of attention away from exclusive focus on
your own stuff and more towards that border

384
00:30:54.649 --> 00:31:00.479
habitat, and you develop the process of scanning
the edge for the solutions that are cropping

385
00:31:00.479 --> 00:31:01.479
up naturally.

386
00:31:01.479 --> 00:31:06.679
If you all read Without Their Permission,
by Alexis Ohanian it talks about permission

387
00:31:06.679 --> 00:31:12.229
less innovation being faster and more potent
than industrial innovation.

388
00:31:12.229 --> 00:31:16.779
So, the process is edge to core.

389
00:31:16.779 --> 00:31:21.769
Devote some number of human hours and focus
and attention and mind share of this project

390
00:31:21.769 --> 00:31:27.239
to scanning that perimeter, relieving the
edge innovators of the burden of supporting

391
00:31:27.239 --> 00:31:32.369
their own platform of keeping HR and Payroll
and their own teams going, bringing the aspects

392
00:31:32.369 --> 00:31:37.729
that can be successful into the middle where
they can scale and then release the hounds,

393
00:31:37.729 --> 00:31:41.950
you know, encourage the rest of that ecosystem
to keep going.

394
00:31:41.950 --> 00:31:43.899
That's going to move faster than anything.

395
00:31:43.899 --> 00:31:47.460
>>Tom Scheinfeldt: So, let me throw this back
to the panel.

396
00:31:47.460 --> 00:31:53.830
We've got several people on the panel who
are supporting kind of some core institutions

397
00:31:53.830 --> 00:31:58.130
or core infrastructure of the national digital
platform, the DPLA, archive management, Archive

398
00:31:58.130 --> 00:32:02.149
Matica, Art Store, these are, I think we’d
all agree, kind of core components of the

399
00:32:02.149 --> 00:32:03.529
national digital platform.

400
00:32:03.529 --> 00:32:11.619
How, as kind of core, central components are
your institutions, your projects, doing that

401
00:32:11.619 --> 00:32:20.059
work of scanning the periphery for good work,
for inspiration, for innovation, for contribution

402
00:32:20.059 --> 00:32:24.809
and supporting that richness at the nose?

403
00:32:24.809 --> 00:32:33.789
>>James Shulman: So, I'm reminded of Josh
Greenburg's dissertation, where he wrote about

404
00:32:33.789 --> 00:32:37.940
how you need translators, and if you're living
in the borders, you need people who can speak

405
00:32:37.940 --> 00:32:41.769
both languages, and, so, obviously, that's
very practical in terms of standards.

406
00:32:41.769 --> 00:32:47.200
We all think in those terms, but just in terms
of respecting cultures and being able to translate,

407
00:32:47.200 --> 00:32:55.889
and, so, we work with a big diversity of institutions,
and as those other communities that we end

408
00:32:55.889 --> 00:33:03.309
up talking to have languages that we understand
and we can speak in, that's when we can start

409
00:33:03.309 --> 00:33:04.309
to be useful.

410
00:33:04.309 --> 00:33:08.970
So, when triple IF emerges out of Stanford
and Oxford, and there are other colleagues

411
00:33:08.970 --> 00:33:13.590
working on this, this is a standard for interoperability
of images that makes so much sense, and, so,

412
00:33:13.590 --> 00:33:17.609
for us to get in that conversation, we have
to be able to speak, we have to be able to

413
00:33:17.609 --> 00:33:22.249
listen, and so it's a translation activity
too.

414
00:33:22.249 --> 00:33:28.509
>>Dan Cohen: Yeah, I mean I’ll add I thought
Sibyl's point about the arrows is really key.

415
00:33:28.509 --> 00:33:33.580
I mean that is the borders there and how you
get things from what really should be a very

416
00:33:33.580 --> 00:33:38.460
diverse set of institutions and operations
and all those things.

417
00:33:38.460 --> 00:33:46.389
Also, at the same time, being able to co mingle
and, you know, allow for innovation between

418
00:33:46.389 --> 00:33:48.799
those institutions, I think is really key.

419
00:33:48.799 --> 00:33:52.200
So, DPLA tries to handle some of those little
arrows that are really important.

420
00:33:52.200 --> 00:33:54.389
There's a lot of behind the scenes work we
do.

421
00:33:54.389 --> 00:33:58.750
For instance, on the normalization of the
data where we understand we've got 1600 institutions

422
00:33:58.750 --> 00:34:04.759
in this range from very small historical societies
up to the Smithsonian, and so we need to combine

423
00:34:04.759 --> 00:34:09.539
all those things, so we act as that little
arrow I think between, allowing for diversity

424
00:34:09.539 --> 00:34:14.650
while also acting as the translator and then
allowing for something to happen at scale,

425
00:34:14.650 --> 00:34:20.619
which you can’t do if you’re just simply
tiny little islands spread around.

426
00:34:20.619 --> 00:34:28.730
>>Evelyn McLellan: I think, we've talked a
little bit about local idiosyncrasies versus

427
00:34:28.730 --> 00:34:33.359
what general needs are, and I think we have
to be careful when we look at local idiosyncrasies

428
00:34:33.359 --> 00:34:37.780
to figure out how idiosyncratic they really
are, because I think what one institution

429
00:34:37.780 --> 00:34:41.510
might consider a unique problem is actually
a problem that a lot of other institutions

430
00:34:41.510 --> 00:34:46.599
have and I think it's important for institutions
to get together and talk about what it is

431
00:34:46.599 --> 00:34:47.810
they're trying to do.

432
00:34:47.810 --> 00:34:53.889
For example, we worked with a couple of institutions
that needed to preserve forensic discs, and

433
00:34:53.889 --> 00:34:56.879
it turned out that their needs were actually
very similar.

434
00:34:56.879 --> 00:35:03.280
So, I would ask an institution that says we
are a special snowflake, we have these very

435
00:35:03.280 --> 00:35:10.069
specific needs, are you really that different,
and if you're that different, maybe that's

436
00:35:10.069 --> 00:35:13.430
a little bit of a problem and you should take
a look at what it is you're trying to do and

437
00:35:13.430 --> 00:35:14.900
how you’re trying to do it.

438
00:35:14.900 --> 00:35:15.900
>>Tom Scheinfeldt: Great.

439
00:35:15.900 --> 00:35:22.309
So, these hub institutions can also be facilitators
between the more local, more idiosyncratic

440
00:35:22.309 --> 00:35:23.309
institutions.

441
00:35:23.309 --> 00:35:24.309
Great.

442
00:35:24.309 --> 00:35:25.309
I think we have a question over here.

443
00:35:25.309 --> 00:35:26.650
>>Michael Mayhem: My name is Michael Mayhem.

444
00:35:26.650 --> 00:35:27.760
I'm with the New York Public Library.

445
00:35:27.760 --> 00:35:30.230
This is a question that I think would have
been good for the last panel.

446
00:35:30.230 --> 00:35:31.230
I think it’s good for you too.

447
00:35:31.230 --> 00:35:35.700
I think it’s going to come up all day, and
it's really, I'm thrilled to see us as a group

448
00:35:35.700 --> 00:35:42.570
embracing this idea of the network to solution
and the idea of building on 20 years of experimentation,

449
00:35:42.570 --> 00:35:46.049
recognizing that we're not starting from zero,
and I think what emerges from this is a vision

450
00:35:46.049 --> 00:35:50.289
for this sort of patchwork quilt of solutions,
where different kinds of institution build

451
00:35:50.289 --> 00:35:52.650
a part of the solution and we switch it together.

452
00:35:52.650 --> 00:35:58.069
So, New York Public Library institutions are
100 years old or a little more and is intensely

453
00:35:58.069 --> 00:36:02.880
local even though we are a large institution,
and so, I guess my question is, as some of

454
00:36:02.880 --> 00:36:05.900
you are, you know, representing DPLA, being
to me the stand out that I am most familiar

455
00:36:05.900 --> 00:36:10.349
with, but all of you, institutions that are
trying to build this national platform, my

456
00:36:10.349 --> 00:36:13.600
question is do we have the right institutions
already?

457
00:36:13.600 --> 00:36:16.020
Do we need new ones, or do the ones that exist
need to evolve?

458
00:36:16.020 --> 00:36:21.150
Because I think there is always going to be
some tension with, certainly the public library

459
00:36:21.150 --> 00:36:26.270
is a great example, funded by local taxpayers
with an intentionally local mission, and there's

460
00:36:26.270 --> 00:36:31.450
always this struggle and trade off between
serving our core local mission and building

461
00:36:31.450 --> 00:36:36.220
something that's going to be useful at the
core, so moving your project from the fringe

462
00:36:36.220 --> 00:36:40.819
use case you were excited about to something
that's national in scale, and, so, I guess

463
00:36:40.819 --> 00:36:45.099
as we continually wrestle with that, I think
we can make tremendous progress by taking

464
00:36:45.099 --> 00:36:50.530
experiments that have been done at that fringe
and scaling them up, but I also wonder if

465
00:36:50.530 --> 00:36:54.070
we have the right kinds of institutions, so,
you know, looking at DPLA being brand new

466
00:36:54.070 --> 00:36:59.580
and an institution that was designed to fit
in the world of the web and in this time and

467
00:36:59.580 --> 00:37:00.580
in this way.

468
00:37:00.580 --> 00:37:04.620
So, my question is do we expand some of these
institutions to be broader?

469
00:37:04.620 --> 00:37:07.470
Or is this not really problem?

470
00:37:07.470 --> 00:37:12.050
Is it possible for that patchwork quilt to
really grow into something that is holistic

471
00:37:12.050 --> 00:37:14.300
and seamless?

472
00:37:14.300 --> 00:37:24.760
>>Dan Cohen: Sure, so you know, I think it's
very important, when thinking about participation

473
00:37:24.760 --> 00:37:29.490
and the kinds of organizations that should
be involved, that we rely not purely on good

474
00:37:29.490 --> 00:37:31.900
will but on narcissism.

475
00:37:31.900 --> 00:37:36.140
I think narcissism is a very important motivator,
and, so, I always try to think in terms of

476
00:37:36.140 --> 00:37:39.940
the social psychology of this of why institutions
would want to participate with something like

477
00:37:39.940 --> 00:37:45.849
DPLA, or when are there new organizations
or technologies or platforms that might be

478
00:37:45.849 --> 00:37:51.420
needed to enable those local institutions,
say, hey, we really need that, right?

479
00:37:51.420 --> 00:37:57.010
So, I think that, thinking through that social
psychology is very important, and I think

480
00:37:57.010 --> 00:38:03.039
for DPLA, we are trying, in effect, to provide
a lot of value and a lot of services to a

481
00:38:03.039 --> 00:38:09.490
variety of institutions, where they can, again,
still maintain autonomy, all of those things,

482
00:38:09.490 --> 00:38:15.410
we're not sucking them dry and taking their
stuff and leaving, you know, their local institutions

483
00:38:15.410 --> 00:38:16.820
to not have any role.

484
00:38:16.820 --> 00:38:21.059
Clearly, they have very important roles, the
center of their communities in the case of

485
00:38:21.059 --> 00:38:25.730
public libraries, but we do allow them then
to connect into this massive infrastructure,

486
00:38:25.730 --> 00:38:30.329
right, and I think the other panelists as
well, they are enabling new functionality

487
00:38:30.329 --> 00:38:32.059
that really does add value.

488
00:38:32.059 --> 00:38:35.020
But these institutions can choose to be a
part of.

489
00:38:35.020 --> 00:38:38.370
New York Public Library I think is part of
our collaboration with public libraries and

490
00:38:38.370 --> 00:38:41.980
it's really wonderful in that it's bi directional,
right?

491
00:38:41.980 --> 00:38:47.211
You have a digital collection that we're also
providing discovery for, and recently, when

492
00:38:47.211 --> 00:38:50.880
you redesigned your digital connection space,
there's a wonderful link right on the digital

493
00:38:50.880 --> 00:38:54.990
collection space to say hey, find this at
DPLA, you can find more things like these

494
00:38:54.990 --> 00:38:59.830
items, which enables New York Public, which
is this wonderful institution and has many

495
00:38:59.830 --> 00:39:05.671
audiences in New York and around the world,
but it allows you to connect into 1599 other

496
00:39:05.671 --> 00:39:10.460
institutions and their holdings and what they
have, so that gives you tremendous value to,

497
00:39:10.460 --> 00:39:14.880
in a sense, have us as a broker for what you're
doing and to leverage that.

498
00:39:14.880 --> 00:39:19.069
So, I think continually thinking about that
kind of, you know, why should we participate,

499
00:39:19.069 --> 00:39:24.850
what are the incentives here, and again, the
narcissism part rather than, I love good will,

500
00:39:24.850 --> 00:39:29.309
it's great, but we want people to find the
value in that, and DPLA is certainly working

501
00:39:29.309 --> 00:39:30.589
really hard to articulate that value.

502
00:39:30.589 --> 00:39:37.220
>>Tom Scheinfeldt: If I could, I really liked
the metaphor of the patchwork quilt, and you

503
00:39:37.220 --> 00:39:43.420
said that there are these pieces that we need
to stitch together, and it occurs to me that

504
00:39:43.420 --> 00:39:53.329
the stitching is provided by sort of the people
on the ground in the institutions, and, you

505
00:39:53.329 --> 00:39:58.780
know, Sibyl mentioned that when you choose
these kind of open platforms over spending

506
00:39:58.780 --> 00:40:03.970
a lot of money on a proprietary platform,
what you're really doing is not really choosing

507
00:40:03.970 --> 00:40:10.599
a cost savings, you're choosing to invest
your money in your people, to get them trained

508
00:40:10.599 --> 00:40:14.690
and to get them involved in an open source
community and learn how to use those technologies,

509
00:40:14.690 --> 00:40:20.770
and I wonder, Sibyl, if you could say a little
bit about the kinds of challenges that a smaller

510
00:40:20.770 --> 00:40:28.559
institution that is involved in these larger
digital initiatives faces in training staff

511
00:40:28.559 --> 00:40:29.559
and getting staff.

512
00:40:29.559 --> 00:40:32.000
We are going to have a panel on that later
on today, but I wonder if you could say a

513
00:40:32.000 --> 00:40:33.000
little bit about that.

514
00:40:33.000 --> 00:40:36.089
>>Sibyl Schaefer: Yeah, well, it is quite
tricky.

515
00:40:36.089 --> 00:40:41.730
For awhile, we didn't even have enough money,
well, we never really had enough money to

516
00:40:41.730 --> 00:40:47.869
hire a full time developer, and, so, being
in archive, I felt that the most core background

517
00:40:47.869 --> 00:40:56.510
was really education in archival studies,
to apply that archival theory and really steep

518
00:40:56.510 --> 00:41:07.260
in archival was essential, and then to kind
of pick up the tech skills as they went along,

519
00:41:07.260 --> 00:41:12.190
and it really depended on the open source
software and what documentation was available

520
00:41:12.190 --> 00:41:19.400
and how responsive developers were and how
they responded was actually quite critical,

521
00:41:19.400 --> 00:41:27.279
because you can respond and give a helpful
answer that helps a user actually learn how

522
00:41:27.279 --> 00:41:35.140
to do something versus just kind of being
dismissive or doing it for them, and I think

523
00:41:35.140 --> 00:41:41.359
that the successful software programs really
have that kind of teaching element to them,

524
00:41:41.359 --> 00:41:47.030
so they work with the constituencies to say,
okay, this is how you file a bug report, this

525
00:41:47.030 --> 00:41:53.930
is how you develop a user story, and you kind
of work with the actual groups, so you educate

526
00:41:53.930 --> 00:41:58.420
them as they participate in the program.

527
00:41:58.420 --> 00:42:00.050
>>Tom Sheinfeldt: Mark?

528
00:42:00.050 --> 00:42:02.230
Okay, is that Amy?

529
00:42:02.230 --> 00:42:04.950
I saw long blonde hair.

530
00:42:04.950 --> 00:42:05.950
Sorry.

531
00:42:05.950 --> 00:42:06.950
Hi, Amy.

532
00:42:06.950 --> 00:42:08.950
>>Amy Garmer: Amy Garmer from the Aspen Institute.

533
00:42:08.950 --> 00:42:14.310
I just wanted to respond to the last comment
who raised this question of do we need institutions

534
00:42:14.310 --> 00:42:19.230
to transform, do we need new types of institutions,
and, you know, I think, certainly, at the

535
00:42:19.230 --> 00:42:26.049
local level, we need the institutions to transform,
and that part of that is understanding that

536
00:42:26.049 --> 00:42:32.490
the job of the institution is no longer just
to curate the world for the community, but

537
00:42:32.490 --> 00:42:38.130
they are in fact curating the community for
the rest of the world, and that's part of

538
00:42:38.130 --> 00:42:45.789
the, you know, not just incentive, but that's
part of the job that they need to be doing.

539
00:42:45.789 --> 00:42:51.510
>>Tom Sheinfeldt: Excellent.

540
00:42:51.510 --> 00:42:54.349
Other questions or comments?

541
00:42:54.349 --> 00:43:04.059
If I could throw a comment, or a question
over to Evelyn, sort of building on Sibyl's

542
00:43:04.059 --> 00:43:10.990
remarks, so if we're thinking about the staff
as being the thread that helps to stitch together

543
00:43:10.990 --> 00:43:15.020
this patchwork quilt, I know in working on
projects, open source software projects myself

544
00:43:15.020 --> 00:43:20.630
on Omecca and Zotero, you know, one of the
most dreaded questions I think that you can

545
00:43:20.630 --> 00:43:28.200
get in the developer forums or in the help
forums are I'm trying to make Omecca work

546
00:43:28.200 --> 00:43:35.390
with this other thing, that, of course, your
staff doesn't have any knowledge of or expertise

547
00:43:35.390 --> 00:43:38.700
in.

548
00:43:38.700 --> 00:43:43.410
What role can the providers for open source
software tools for libraries, archives and

549
00:43:43.410 --> 00:43:50.520
museums, do you have strategies for helping
make those connections to others?

550
00:43:50.520 --> 00:43:54.880
It's one thing to answer questions about your
own product, but to answer questions about

551
00:43:54.880 --> 00:43:58.250
how your product works with another, do you
have strategies for dealing with that kind

552
00:43:58.250 --> 00:44:01.220
of a problem?

553
00:44:01.220 --> 00:44:02.599
Not to put you on the spot or anything.

554
00:44:02.599 --> 00:44:08.760
>>Evelyn McLellan: It's hard, because in the
discussion, we get so many questions about

555
00:44:08.760 --> 00:44:15.140
why doesn't your software do this or will
it work with this, and it's, I feel like our

556
00:44:15.140 --> 00:44:20.990
software should do everything, and we often
have to answer, we would love it if our software

557
00:44:20.990 --> 00:44:26.530
could do that, but we would be looking for
funding, which is not a very satisfactory

558
00:44:26.530 --> 00:44:36.520
answer, but it's, we try to connect users
with other software packages as often as we

559
00:44:36.520 --> 00:44:43.420
can, but, often, the best thing is when the
users jump in and answer the questions that

560
00:44:43.420 --> 00:44:48.089
are asked by the first users, and we really
like to see that, because we don't know all

561
00:44:48.089 --> 00:44:54.080
the answers, we're just one small group of
people, and if we can encourage the people

562
00:44:54.080 --> 00:44:59.720
on our discussion list to talk to each other,
we've seen some great back and forth conversations,

563
00:44:59.720 --> 00:45:04.319
especially between large institutions that
have a lot of resources and large scale implementations.

564
00:45:04.319 --> 00:45:08.990
I love it when they jump in and answer a question
from a small institution and say this is what

565
00:45:08.990 --> 00:45:12.940
we did, and it worked for us, and you can
look at this tool and you can do other things,

566
00:45:12.940 --> 00:45:22.119
and I guess, I mean, this is kind of a diffuse
answer, but I guess what I like to see is

567
00:45:22.119 --> 00:45:25.500
knowledge flowing down from the larger institutions
to the smaller institutions, because I really

568
00:45:25.500 --> 00:45:29.319
think we need that, because there's some really
fantastic things going on.

569
00:45:29.319 --> 00:45:34.080
There’s a lot of great work that's being
done in digital preservation and access, and

570
00:45:34.080 --> 00:45:41.490
it's, I feel sometimes like the smaller institutions
are just shut out.

571
00:45:41.490 --> 00:45:45.740
>>Tom Scheinfeldt: Other questions from the
audience?

572
00:45:45.740 --> 00:45:50.770
>>Dan Cohen: I have just a gloss on that.

573
00:45:50.770 --> 00:45:54.700
One of the wonderful things that happened
TPLA Fest last week is that there were in

574
00:45:54.700 --> 00:46:01.059
fact directors of major university libraries
in the same room with the heads of small public

575
00:46:01.059 --> 00:46:02.059
libraries.

576
00:46:02.059 --> 00:46:08.369
There were people from big museums and small
historical societies in the same room, and,

577
00:46:08.369 --> 00:46:11.950
you know, in my career, I've just, I've found
those kinds of discussions, they're very,

578
00:46:11.950 --> 00:46:18.641
very elusive, and, you know, there are conferences
that bring together, you know, research libraries,

579
00:46:18.641 --> 00:46:20.710
in fact, association research libraries is
meeting as we talk, and that's great and those

580
00:46:20.710 --> 00:46:26.660
folks need to get together, but it's very
rare to have knowledge dissemination occur

581
00:46:26.660 --> 00:46:30.359
across some of these boundaries, and I think
IMLS and others in the room can play a real

582
00:46:30.359 --> 00:46:39.090
role in sort of facilitating that type of
conversation and I certainly hope TPLA Fest

583
00:46:39.090 --> 00:46:40.090
can do that.

584
00:46:40.090 --> 00:46:41.090
>>Karen Keviani: Hi.

585
00:46:41.090 --> 00:46:42.090
Karen Keviani from WGBH.

586
00:46:42.090 --> 00:46:44.810
I would like to speak to Evelyn's comment
about how do you get smaller institutions

587
00:46:44.810 --> 00:46:49.470
to be able to contribute to the larger, bigger
institutions.

588
00:46:49.470 --> 00:46:55.010
For example, I may have one or two developers
that I could add to a sprint or some kind

589
00:46:55.010 --> 00:47:01.240
of development work, but there needs to be
better ways to engage them into that larger

590
00:47:01.240 --> 00:47:06.800
community so that they are helping build something
that will then help them locally, and I think

591
00:47:06.800 --> 00:47:11.410
the Hydra’s working really hard to do that,
which is great, but there's still a lot of

592
00:47:11.410 --> 00:47:15.660
work that needs to happen in order to really
fully engage those smaller institutions and

593
00:47:15.660 --> 00:47:18.599
for the smaller institutions to really feel
like they're contributing.

594
00:47:18.599 --> 00:47:25.160
>> SPEAKER: Well this is, I mean, I ment to
mention it during my talk but I'm so excited

595
00:47:25.160 --> 00:47:30.740
about that Hydra and the Box Project, which
is fantastic, because Hydra's a great tool,

596
00:47:30.740 --> 00:47:34.880
and, so far, really the only institutions
that have been able to use Hydra are the large

597
00:47:34.880 --> 00:47:38.650
university libraries where they have core
developers to develop Hydra gems and they’re

598
00:47:38.650 --> 00:47:45.700
doing great things, but to see this possibility
down the road of having, literally, an easy

599
00:47:45.700 --> 00:47:51.910
to install Hydra instance, that has, you know,
a lot of functionality for lots of institutions

600
00:47:51.910 --> 00:47:56.220
to use I think it is very, very exciting,
and it's great because it's a collaboration

601
00:47:56.220 --> 00:48:02.579
between, you know, three major organizations,
DPLA, Dura Space, and Stanford.

602
00:48:02.579 --> 00:48:08.059
So, I think this is one of the most exciting
projects I've seen in a while.

603
00:48:08.059 --> 00:48:12.690
>> SPEAKER: And along with that, I'm on the
technical advisory committee for Archive Space,

604
00:48:12.690 --> 00:48:18.510
and that actual question came up with somebody
who basically said I have developers, I want

605
00:48:18.510 --> 00:48:23.130
to contribute to sprints, but I want to contribute
to sprints where we're actually interested

606
00:48:23.130 --> 00:48:27.190
in that development, and, so, I think the
number one thing in open sourced software

607
00:48:27.190 --> 00:48:33.680
they can put out there is a roadmap of what
these sprints are going to be and what functionality

608
00:48:33.680 --> 00:48:37.980
they're planning and that’s very helpful
even for small organizations too.

609
00:48:37.980 --> 00:48:41.539
They're waiting for some functionality to
be developed to actually know when they can

610
00:48:41.539 --> 00:48:47.230
expect it, because that influences decisions.

611
00:48:47.230 --> 00:48:57.040
>>Tom Scheinfeldt: I'll ask a question.

612
00:48:57.040 --> 00:48:59.839
That’s something I was wondering about this
morning.

613
00:48:59.839 --> 00:49:07.809
You know, it occurs to me that a central element
of the national digital platform, especially

614
00:49:07.809 --> 00:49:15.839
for smaller institutions, but even for individual
users at larger institutions, and something

615
00:49:15.839 --> 00:49:23.299
that's not really represented here and maybe
should be are consumer services, right, consumer

616
00:49:23.299 --> 00:49:27.640
web services, like drop box or Google docs.

617
00:49:27.640 --> 00:49:33.369
I mean, I know that, we like to, and, you
know, I've done it myself in the marketing

618
00:49:33.369 --> 00:49:40.039
for our open source tools, talk about things
being easy to use or in a box, I've used both

619
00:49:40.039 --> 00:49:46.180
of those phrases before, but what do we really
mean by easy to use?

620
00:49:46.180 --> 00:49:51.700
I mean, you know, like a good example, Omecca,
it's easy to use compared to some other enterprise

621
00:49:51.700 --> 00:49:59.910
systems, but it's not really easy to use like
drop box is easy to use, and I think especially,

622
00:49:59.910 --> 00:50:04.720
when we're talking about a range of institutions,
where I think Dropbox really is pretty easy

623
00:50:04.720 --> 00:50:06.880
to use for everybody, although not really
for my mother.

624
00:50:06.880 --> 00:50:13.030
I had to give my mother a Dropbox tutorial
recently.

625
00:50:13.030 --> 00:50:19.819
How do we maybe integrate with some of those
really truly easy to use systems and bring

626
00:50:19.819 --> 00:50:24.750
them into this conversation because I do think
those are the tools that a lot of very, very

627
00:50:24.750 --> 00:50:29.000
small institutions if they are using digital
tools and digital work flows.

628
00:50:29.000 --> 00:50:34.039
They’re making heavy use of some of these
commercial services as we all do.

629
00:50:34.039 --> 00:50:41.220
>>Dan Cohen: Well, I mean, I'll jump in and
say here's where things like API's help out

630
00:50:41.220 --> 00:50:42.240
a lot.

631
00:50:42.240 --> 00:50:49.299
So, I'm not going to give your mom an API
tutorial, but I will give a tutorial to a

632
00:50:49.299 --> 00:50:54.260
third party developer that can integrate,
for instance, DPLA and DPLA collections to

633
00:50:54.260 --> 00:50:58.500
services like Twitter, Facebook, etc., so
I think there again, that notion of bridge,

634
00:50:58.500 --> 00:51:05.230
I think Sibyl’s notion of the arrows are
very important, and I think one of the values

635
00:51:05.230 --> 00:51:10.319
that DPLA provides is even if you're a small
collection, you now have an API through DPLA,

636
00:51:10.319 --> 00:51:11.319
right?

637
00:51:11.319 --> 00:51:18.789
Your collections can be accessible, knitted
together arrows in multiple directions because

638
00:51:18.789 --> 00:51:23.910
you have participated in this national network,
and I think that's really a signature kind

639
00:51:23.910 --> 00:51:27.240
of aspect to any sort of national digital
platform.

640
00:51:27.240 --> 00:51:31.220
>>Cliff Lynch: Cliff Lynch from CNI.

641
00:51:31.220 --> 00:51:40.170
I may be just trying to get at the question
you just asked from a somewhat different perspective,

642
00:51:40.170 --> 00:51:45.030
but I'd be interested in some comments on
this.

643
00:51:45.030 --> 00:51:53.520
We've heard quite a bit of discussion already
about the barriers to participation for smaller

644
00:51:53.520 --> 00:51:58.559
institutions that have very limited IT resources.

645
00:51:58.559 --> 00:52:02.960
We've also heard quite a lot of discussion
about open source software and things of that

646
00:52:02.960 --> 00:52:04.310
nature.

647
00:52:04.310 --> 00:52:15.650
I actually wonder if the national platform
of this portfolio of resources particularly

648
00:52:15.650 --> 00:52:23.570
is viewed by smaller institutions is not going
to be, is going to be much more about services,

649
00:52:23.570 --> 00:52:30.660
that really the idea is going to be there
are services available out on the network

650
00:52:30.660 --> 00:52:39.240
that hopefully work together in some kind
of cohesive way, whereby, really, a lot of

651
00:52:39.240 --> 00:52:44.410
the smaller institutions can avoid a good
deal of the IT detail by just using these

652
00:52:44.410 --> 00:52:53.490
services, and, so, I guess I'd invite some
reflection about the balance of software and

653
00:52:53.490 --> 00:52:54.490
services.

654
00:52:54.490 --> 00:52:59.760
Certainly when I look at what's going on in
other areas of information technology, especially

655
00:52:59.760 --> 00:53:09.769
with, you know, smaller organizations, we
see a massive shift of IT from locally mounted

656
00:53:09.769 --> 00:53:10.769
services.

657
00:53:10.769 --> 00:53:16.580
I'm sorry, locally mounted systems out into
a services world.

658
00:53:16.580 --> 00:53:19.080
>>Tom Scheinfeldt: James?

659
00:53:19.080 --> 00:53:26.420
>>James Shulman: So, with our IMLS funded
project, that's very much our goal.

660
00:53:26.420 --> 00:53:33.600
So, our goal is to take this platform, this
core platform that we built, Shared Shelf,

661
00:53:33.600 --> 00:53:39.589
and to make it available as free and open
services for small institutions to use it,

662
00:53:39.589 --> 00:53:49.660
to contribute to a national flow, and I think,
over the years, with various funders, I think

663
00:53:49.660 --> 00:53:54.710
it's been a lot of progress for people to
realize, so we'll make the code for those

664
00:53:54.710 --> 00:53:59.680
components open source, but we're not working
to build an open source software there, we're

665
00:53:59.680 --> 00:54:06.869
working to build components of a free service,
and that free service will go to the small

666
00:54:06.869 --> 00:54:11.470
museum, and that's why we're working with,
you know, we have some large museums in the

667
00:54:11.470 --> 00:54:17.420
group that got the grant with us and with
DPLA, but we also have the Staten Island Museum,

668
00:54:17.420 --> 00:54:22.530
and we have the El Paso Museum of Art, and
we're not looking for some group that's going

669
00:54:22.530 --> 00:54:26.789
to have three people to integrate, and interoperate,
and make it work with their digital asset

670
00:54:26.789 --> 00:54:32.180
management, their collections management and,
you know, spit out a solar Rendex and if need

671
00:54:32.180 --> 00:54:34.599
be, just write a little script to do this.

672
00:54:34.599 --> 00:54:39.880
We really think that what we need to do is
provide an on ramp that's easy, and I think

673
00:54:39.880 --> 00:54:44.130
when we talk about a national digital platform,
or a national platform, I mean, we all think

674
00:54:44.130 --> 00:54:48.119
often of the federal highway system which
has rules, you know, here’s how you get

675
00:54:48.119 --> 00:54:49.119
on.

676
00:54:49.119 --> 00:54:54.619
You don’t have a stop sign at the end of
your ramp, and so, I very much think that

677
00:54:54.619 --> 00:54:57.940
that's going to be a key part of getting people
in the game.

678
00:54:57.940 --> 00:55:04.730
>>Evelyn McLellan: I talked briefly about
the enormous growth in the number of open

679
00:55:04.730 --> 00:55:07.410
source software tools over the last five years
or so.

680
00:55:07.410 --> 00:55:11.980
I'm hoping over the next five years, we'll
be talking about the enormous growth in hosting

681
00:55:11.980 --> 00:55:19.210
services and hubs and networks things like
DPen and, you know, expansion of services

682
00:55:19.210 --> 00:55:24.210
provided by DPLA, because it literally, for
some institutions they are not looking at

683
00:55:24.210 --> 00:55:28.930
open source tools, they're not thinking about
tools, they're thinking about their needs

684
00:55:28.930 --> 00:55:34.570
and what services they can get, and from my
perspective, I'm happy to develop open source

685
00:55:34.570 --> 00:55:39.529
tools, and I hope that they get integrated
into platforms and services that these institutions

686
00:55:39.529 --> 00:55:45.089
can use, because they're not going to download
and use Archive Matica in their own institution.

687
00:55:45.089 --> 00:55:52.450
>>Dan Cohen: One of the reasons we're so grateful
to have gotten IMLS funding along with Stanford

688
00:55:52.450 --> 00:55:58.029
and Dura Space on you know, the hubs in a
box idea, is in fact that idea that it can

689
00:55:58.029 --> 00:56:05.250
be hosted remotely for institutions that don’t
want to devote developer time, etc., or can’t

690
00:56:05.250 --> 00:56:06.250
develop themselves.

691
00:56:06.250 --> 00:56:09.540
Obviously, we’re creating open source software
and we’ve got great partners.

692
00:56:09.540 --> 00:56:14.750
I think it'll be terrific to have this out
there and to be able to be deployable in that

693
00:56:14.750 --> 00:56:20.850
way, but I also think it presents the opportunities
for service based operation as well.

694
00:56:20.850 --> 00:56:23.960
>>Jim Neal: Jim Neal, Columbia.

695
00:56:23.960 --> 00:56:30.359
The costs of the IMLS sponsored event, we
have focused, in this conversation, about

696
00:56:30.359 --> 00:56:33.490
the state of a distributed national capacity.

697
00:56:33.490 --> 00:56:38.150
I wonder if the panelists would talk a little
bit about the state of an expanded global

698
00:56:38.150 --> 00:56:39.789
capacity.

699
00:56:39.789 --> 00:56:47.910
How do we fit into a larger platform which
is global and far more diverse than US centric

700
00:56:47.910 --> 00:56:49.420
issues that we’re talking about today.

701
00:56:49.420 --> 00:56:53.029
>>Tom Scheinfeldt: James, do you want to start?

702
00:56:53.029 --> 00:57:02.190
>>James Shulman: Well, I think that these
border conversations need common languages

703
00:57:02.190 --> 00:57:04.609
and need translators, as I was saying before.

704
00:57:04.609 --> 00:57:10.519
I think Dan can speak to the work that he
does regularly and DPLA does with other national

705
00:57:10.519 --> 00:57:15.609
libraries, and I think that's sort of the
level of supra aggregation, and then there's

706
00:57:15.609 --> 00:57:24.480
the micro level of letting people in the game,
and I think those rules of invitation and

707
00:57:24.480 --> 00:57:27.980
correspondence are very much the same around
the world.

708
00:57:27.980 --> 00:57:33.150
I mean, the intellectual property rules are
different, and the languages are different,

709
00:57:33.150 --> 00:57:38.230
so those are things, those are added modules
that we have to figure out and that will slow

710
00:57:38.230 --> 00:57:45.180
down some flows, but as you say, there's certainly
no reason that those handshakes aren't extended

711
00:57:45.180 --> 00:57:48.160
at the micro institutional level.

712
00:57:48.160 --> 00:57:52.339
>> SPEAKER: So, Jim, it's happening right
now.

713
00:57:52.339 --> 00:57:54.880
I mean, it is happening.

714
00:57:54.880 --> 00:57:58.910
Right now it’s the 100th anniversary of
the Great War, the First World War.

715
00:57:58.910 --> 00:58:04.960
DPLA’s data which is completely interoperable
with Europeana and Digital Enze in New Zealand,

716
00:58:04.960 --> 00:58:10.390
and Trove in Australia is all combined right
now in large exhibition of hundreds of thousands

717
00:58:10.390 --> 00:58:15.480
of items that Europeana mounted and were able
to mount that because of the seamless interoperability.

718
00:58:15.480 --> 00:58:20.710
So, you know, these are just the first steps,
but I think that project alone, it’s called

719
00:58:20.710 --> 00:58:27.279
Europeana 1914-1918 incredibly suggestive,
and I think it would be a great role for IMLS

720
00:58:27.279 --> 00:58:31.519
and others in the room to facilitate more
of these conversations that I have regularly

721
00:58:31.519 --> 00:58:36.069
with folks around the world, my counterparts
other national and international digital library

722
00:58:36.069 --> 00:58:37.069
projects.

723
00:58:37.069 --> 00:58:41.100
I think there’s tremendous potential to,
in a sense, bootstrap into a global digital

724
00:58:41.100 --> 00:58:42.819
library through these methods.

725
00:58:42.819 --> 00:58:45.630
>>Tom Scheinfeldt: Mark?

726
00:58:45.630 --> 00:58:51.240
Oh, I’m sorry, mic in front.

727
00:58:51.240 --> 00:58:55.720
>>SPEAKER: Tom, you mentioned linked open
data in the first few sentences.

728
00:58:55.720 --> 00:58:57.869
Could you speak a little bit more about it?

729
00:58:57.869 --> 00:58:59.069
>>Tom Scheinfeldt: Me?

730
00:58:59.069 --> 00:59:00.269
>> SPEAKER: Could anyone?

731
00:59:00.269 --> 00:59:03.650
>>Tom Scheinfeldt: Why don't I open up to
the panelists.

732
00:59:03.650 --> 00:59:07.990
Anyone want to tackle linked open data?

733
00:59:07.990 --> 00:59:20.710
>>Dan Cohen: I might pass the mic to Mark
Matienzo but I will say it is an area of extreme

734
00:59:20.710 --> 00:59:23.180
interest for DPLA.

735
00:59:23.180 --> 00:59:28.390
I think, again, here's another role that we
can play, is that a lot of institutions cancel

736
00:59:28.390 --> 00:59:32.769
out linked data into their data payloads,
and so, one of the things we can do as part

737
00:59:32.769 --> 00:59:38.200
of the normalization and enhancement of the
1600 institutions’ data is in fact to add

738
00:59:38.200 --> 00:59:42.309
in linked data as part of that process.

739
00:59:42.309 --> 00:59:44.109
Their data is now super enhanced.

740
00:59:44.109 --> 00:59:50.000
It's connected in new ways across institutions
nationally, and indeed internationally through

741
00:59:50.000 --> 00:59:56.250
linked data ideas, so it's an area where we're
going to work really hard at DPLA to get going,

742
00:59:56.250 --> 01:00:01.740
and in fact, if you look at our latest metadata
application profile, 4.0, which you all should

743
01:00:01.740 --> 01:00:06.809
read immediately, there is in fact a significant
portion in that that's devoted to linked data,

744
01:00:06.809 --> 01:00:09.760
so it's definitely on our radar.

745
01:00:09.760 --> 01:00:16.190
>>Mark Matienzo: I'm Mark Matienzo the director
of technology at DPLA.

746
01:00:16.190 --> 01:00:20.809
I kind of wanted to follow up on something
that was just said before about sort of Jim's

747
01:00:20.809 --> 01:00:24.970
question about how things happen in the international
context.

748
01:00:24.970 --> 01:00:30.109
I mean, I think with representation of the
panel and those of us in the room, I mean,

749
01:00:30.109 --> 01:00:37.589
I don't want to just toot DPLA's horn or the
horn of the Hydra project, but we have made

750
01:00:37.589 --> 01:00:43.049
a lot of progress in starting to connect with
Europeana on multiple levels, so the DPLA

751
01:00:43.049 --> 01:00:49.190
metadata application profile is built on the
Europeana data model, so we are a participant

752
01:00:49.190 --> 01:00:51.620
in discussions in data modeling.

753
01:00:51.620 --> 01:00:56.730
We now have succeeded in getting Europeana
to contribute to data modeling efforts within

754
01:00:56.730 --> 01:01:01.901
the Hydra community quite significantly, which
is a very, very recent development, and they

755
01:01:01.901 --> 01:01:07.599
are showing continued interest in sort of
related and allied projects such as Blacklight

756
01:01:07.599 --> 01:01:09.340
and Spotlight.

757
01:01:09.340 --> 01:01:17.970
So, also, similarly, Archive Matica, you know,
has an international community of users, and

758
01:01:17.970 --> 01:01:23.200
so, you know, it sort of, when we think about
the international community I think we need

759
01:01:23.200 --> 01:01:24.819
to figure out what our plans are.

760
01:01:24.819 --> 01:01:30.430
I mean, from my perspective, it's specifically
around interoperability and sort of not repeating

761
01:01:30.430 --> 01:01:34.040
effort around data modeling where it's not
necessary.

762
01:01:34.040 --> 01:01:39.319
You know, going back to what Evelyn said before,
is I think this notion of idiosyncrasy is

763
01:01:39.319 --> 01:01:44.619
kind of somewhat problematic, but they're
not really idiosyncrasies.

764
01:01:44.619 --> 01:01:48.930
If we start to see this problem at multiple
places, this is a problem that exists in scale.

765
01:01:48.930 --> 01:01:53.559
The types of things that Sibyl were talking
about is really an international problem for

766
01:01:53.559 --> 01:01:57.560
a lot of archives that are working with digital
material.

767
01:01:57.560 --> 01:02:04.660
So, you know, just because we start to feel
the friction in our own institutions about

768
01:02:04.660 --> 01:02:10.599
where these work flows fail, that, you know,
that indicates a larger scale problem that

769
01:02:10.599 --> 01:02:13.380
often exists on an international level.

770
01:02:13.380 --> 01:02:20.670
>>Vicky Rice: So, Vicky Rice.

771
01:02:20.670 --> 01:02:22.130
Stanford.

772
01:02:22.130 --> 01:02:32.900
Another aspect of collaboration is the web
archiving community, and IBC is at Stanford,

773
01:02:32.900 --> 01:02:39.259
while Tom and I are here and not at that meeting
and that is truly international.

774
01:02:39.259 --> 01:02:45.980
You have the national libraries doing web
archiving, you have the smaller libraries

775
01:02:45.980 --> 01:02:50.420
doing web archiving through Archive It and
other tools, and, so, I want to make sure

776
01:02:50.420 --> 01:02:58.660
it doesn't get lost that, again, having API's,
and Dan mentioned this, and I would love Dan

777
01:02:58.660 --> 01:03:07.749
to come give a tutorial to my mother, please,
having APIs so that the tools that are being

778
01:03:07.749 --> 01:03:15.859
used to capture and preserve and disseminate
and to manage the web published content would

779
01:03:15.859 --> 01:03:21.849
be very, very important, and to make those
tools easy to use so that the people that

780
01:03:21.849 --> 01:03:27.320
Sibyl is talking about, the folks at the local
level, can get their government documents,

781
01:03:27.320 --> 01:03:33.030
the town government documents and preserve
them and make them available would be just

782
01:03:33.030 --> 01:03:34.740
huge.

783
01:03:34.740 --> 01:03:39.380
>>Speaker: I think that's a great place for
us to call it.

784
01:03:39.380 --> 01:03:46.740
We thank you so much for talking about where
we are now, this state of distributed national

785
01:03:46.740 --> 01:03:47.980
digital capacity.

786
01:03:47.980 --> 01:03:52.230
We will come back in 20 minutes or so and
talk about the gaps, which is always more

787
01:03:52.230 --> 01:04:00.180
fun, to talk about what we need still, and
that's where we're, obviously, very interested.

788
01:04:00.180 --> 01:04:01.650
So, thank you all.

789
01:04:01.650 --> 01:04:03.990
Let's give them a round of applause for their
help.

790
01:04:03.990 --> 01:04:04.090
(Applause.)