Can anyone offer some leads on the use of the term 'domain' within computing and information circles? Such as 'domain scientist' or 'domain knowledge'... I've used the term for years as an in vivo category drawn from my fieldwork, but have little sense of its historical genealogy. It was already well in place by 2003 when I started fieldwork in the worlds of cyberinfrastructure, and I have some suspicion it may have come from knowledge management/business ... -- David Ribes Associate Professor Department of Human Centered Design and Engineering (HCDE) University of Washington http://davidribes.com
This is a great question and I am also curious about the answer. I always assumed it was an import from logic, e.g., range and domain. I have written a bit about the further appropriation from computer science by the DoD in defining the "cyber domain" which first appears in 2000 and then takes off. As late as 1999 military writers would talk loosely about an "information arena" or "commons" something like that, but "domain" takes off later. I read this as largely a bureaucratic bid for legitimacy, defining "cyber" as coequal with land, sea, air, and space, with the USAF attempting to lay claim to three of these (air, space, cyber). Legal discussions of domain are much older, usually used to distinguish territorial waters from high seas, which is interesting in that the "cyber domain" would in that sense be just the opposite of a "cyber commons", but if you are a global hegemon perhaps you can pretend to dominate a global commons. I have a suspicion that the logical computer science usage of domain thus made it really attractive to the military to play off of the meaning "to dominate" not just an arena or bounded conceptual area. Jon On Tue, Oct 18, 2016 at 1:19 PM, David Ribes <dribes@gmail.com> wrote:
Can anyone offer some leads on the use of the term 'domain' within computing and information circles? Such as 'domain scientist' or 'domain knowledge'...
I've used the term for years as an in vivo category drawn from my fieldwork, but have little sense of its historical genealogy. It was already well in place by 2003 when I started fieldwork in the worlds of cyberinfrastructure, and I have some suspicion it may have come from knowledge management/business ...
-- David Ribes Associate Professor Department of Human Centered Design and Engineering (HCDE) University of Washington http://davidribes.com
_______________________________________________ This email is relayed from members at sigcis.org, the email discussion list of SHOT SIGCIS. Opinions expressed here are those of the member posting and are not reviewed, edited, or endorsed by SIGCIS. The list archives are at http://lists.sigcis.org/pipermail/members-sigcis.org/ and you can change your subscription options at http://lists.sigcis.org/ listinfo.cgi/members-sigcis.org
Others will have deeper thoughts on this, but my guess is that its use in expert systems AI is at least part of the story of how "domain XYZ" caught on more broadly. (Generic AI tools) + (a knowledge-base and heuristics consisting of the "domain knowledge" of "domain scientists" or "domain experts") = (an expert system). Googling "Feigenbaum 'domain knowledge'" turns up a paper from the late 70s <http://ijcai.org/Proceedings/77-2/Papers/092.pdf> in which Ed Feigenbaum uses the term in this way. Best, Evan On Tue, Oct 18, 2016 at 1:28 PM, Jon Lindsay <jonrlindsay@gmail.com> wrote:
This is a great question and I am also curious about the answer. I always assumed it was an import from logic, e.g., range and domain. I have written a bit about the further appropriation from computer science by the DoD in defining the "cyber domain" which first appears in 2000 and then takes off. As late as 1999 military writers would talk loosely about an "information arena" or "commons" something like that, but "domain" takes off later. I read this as largely a bureaucratic bid for legitimacy, defining "cyber" as coequal with land, sea, air, and space, with the USAF attempting to lay claim to three of these (air, space, cyber). Legal discussions of domain are much older, usually used to distinguish territorial waters from high seas, which is interesting in that the "cyber domain" would in that sense be just the opposite of a "cyber commons", but if you are a global hegemon perhaps you can pretend to dominate a global commons. I have a suspicion that the logical computer science usage of domain thus made it really attractive to the military to play off of the meaning "to dominate" not just an arena or bounded conceptual area. Jon
On Tue, Oct 18, 2016 at 1:19 PM, David Ribes <dribes@gmail.com> wrote:
Can anyone offer some leads on the use of the term 'domain' within computing and information circles? Such as 'domain scientist' or 'domain knowledge'...
I've used the term for years as an in vivo category drawn from my fieldwork, but have little sense of its historical genealogy. It was already well in place by 2003 when I started fieldwork in the worlds of cyberinfrastructure, and I have some suspicion it may have come from knowledge management/business ...
-- David Ribes Associate Professor Department of Human Centered Design and Engineering (HCDE) University of Washington http://davidribes.com
_______________________________________________ This email is relayed from members at sigcis.org, the email discussion list of SHOT SIGCIS. Opinions expressed here are those of the member posting and are not reviewed, edited, or endorsed by SIGCIS. The list archives are at http://lists.sigcis.org/pipermail/members-sigcis.org/ and you can change your subscription options at http://lists.sigcis.org/listinfo.cgi/members-sigcis.org
_______________________________________________ This email is relayed from members at sigcis.org, the email discussion list of SHOT SIGCIS. Opinions expressed here are those of the member posting and are not reviewed, edited, or endorsed by SIGCIS. The list archives are at http://lists.sigcis.org/pipermail/members-sigcis.org/ and you can change your subscription options at http://lists.sigcis.org/ listinfo.cgi/members-sigcis.org
-- Evan Hepler-Smith Environmental Fellow Harvard University Center for the Environment 339.203.1096 evanheplersmith.com
David —your question about the history of “domain knowledge” stirred some memories and I did a quick search through my primary source documents. The word “domain” occurs regularly from the mid-1960s on. The earliest uses are what you might expect — narrow mathematical definitions ("what is the domain of this function") or broadly professional ("what ought to be the proper domain of the computer sciences”) — but in terms of the specific usage that you are interested in (i.e., the idea of their being different “problem domains” to which computational techniques can be applied), the real origins seems to have been the late 1960s. For example, "The expert programmer should be familiar with such modeling techniques and the necessary assumptions and domains of their applicability…” (Bruce Arden, "The Role of Programming in a Ph.D. Computer Science Program,” ACM Communications 1969) or "algorithms which are inherently known and available within a specific problem domain.” (Jean Sammett, "Programming Languages: History and Future”, also ACM Communications 1969). The term “problem domain” seems to have originated with expert systems in the early 1980s. See J. McDermott "Domain Knowledge and the Design Process” (1981) or J. Ramanathan and C.J. Shubra "Modeling of problem domains for driving program development systems” (1981). But leaving aside the specific words (domain scientist, domain knowledge), the larger debate about which knowledge/expertise was most essential to the successful application of computing power goes back to at least the late 1950s, when a discussion about whether it was best to train computer people in business or business people in computing became a hot topic in the electronic data processing literature. They do not use the “domain” language, but the question of overlapping (or possibly incommensurate) disciplinary territory is clearly an issue… -Nathan --- Nathan Ensmenger Associate Professor of Informatics School of Informatics and Computing Indiana University, Bloomington homes.soic.indiana.edu/nensmeng/
On Oct 18, 2016, at 1:19 PM, David Ribes <dribes@gmail.com> wrote:
Can anyone offer some leads on the use of the term 'domain' within computing and information circles? Such as 'domain scientist' or 'domain knowledge'...
I've used the term for years as an in vivo category drawn from my fieldwork, but have little sense of its historical genealogy. It was already well in place by 2003 when I started fieldwork in the worlds of cyberinfrastructure, and I have some suspicion it may have come from knowledge management/business ...
-- David Ribes Associate Professor Department of Human Centered Design and Engineering (HCDE) University of Washington http://davidribes.com
_______________________________________________ This email is relayed from members at sigcis.org, the email discussion list of SHOT SIGCIS. Opinions expressed here are those of the member posting and are not reviewed, edited, or endorsed by SIGCIS. The list archives are at http://lists.sigcis.org/pipermail/members-sigcis.org/ and you can change your subscription options at http://lists.sigcis.org/listinfo.cgi/members-sigcis.org
Jon, Evan, Nathan -- I just wanted to say thanks much for these insights on 'domain'. You've provided a nice imaginarium of the various linked but local usages of the term. I've been following up on these threads and scouring the literature for anyone who has done a synthetic analysis. Nothing yet. I did find a blip in Suchman 1987, where she is discussing the AI/cog.sci approach to 'slicing up the domains' for knowledge capture, but her thrust is mostly to critique the effort to thereafter define and capture a domain for 'common sense knowledge' (a project the ethnomethodologists/phenomenologists think is a no-go). As a hint to why I ask: In my fieldwork today, in short, I'm seeing the non/domain distinction being deployed very heavily in data science circles (far more so than I saw in cyberinfrastructure), and I'm thinking it's a key organizational principle. It seems to refer to something akin to general/specific, where, mostly, computation/IT/modeling/visualization is placed on the general side and 'domain' is a specific worldly knowledge or engagement... Here is a tidy example (first sentence!): http://data.uw.edu/incubator/ Nathan, I'm generally not seeing anyone today referring to computer science as 'having a domain' or efforts to define the relevant domains of CS as with your 1960s references. I'm sure there are exceptions, but today domain seems to be 'the other' to CS... Comments welcome, and thanks again, david. -- David Ribes Associate Professor Department of Human Centered Design and Engineering (HCDE) University of Washington http://davidribes.com On Tue, Oct 18, 2016 at 10:49 AM, Nathan Ensmenger < nathan.ensmenger@gmail.com> wrote:
David —your question about the history of “domain knowledge” stirred some memories and I did a quick search through my primary source documents. The word “domain” occurs regularly from the mid-1960s on. The earliest uses are what you might expect — narrow mathematical definitions ("what is the domain of this function") or broadly professional ("what ought to be the proper domain of the computer sciences”) — but in terms of the specific usage that you are interested in (i.e., the idea of their being different “problem domains” to which computational techniques can be applied), the real origins seems to have been the late 1960s.
For example, "The expert programmer should be familiar with such modeling techniques and the necessary assumptions and domains of their applicability…” (Bruce Arden, "The Role of Programming in a Ph.D. Computer Science Program,” ACM Communications 1969) or "algorithms which are inherently known and available within a specific problem domain.” (Jean Sammett, "Programming Languages: History and Future”, also ACM Communications 1969). The term “problem domain” seems to have originated with expert systems in the early 1980s. See J. McDermott "Domain Knowledge and the Design Process” (1981) or J. Ramanathan and C.J. Shubra "Modeling of problem domains for driving program development systems” (1981).
But leaving aside the specific words (domain scientist, domain knowledge), the larger debate about which knowledge/expertise was most essential to the successful application of computing power goes back to at least the late 1950s, when a discussion about whether it was best to train computer people in business or business people in computing became a hot topic in the electronic data processing literature. They do not use the “domain” language, but the question of overlapping (or possibly incommensurate) disciplinary territory is clearly an issue…
-Nathan
--- Nathan Ensmenger Associate Professor of Informatics School of Informatics and Computing Indiana University, Bloomington homes.soic.indiana.edu/nensmeng/
On Oct 18, 2016, at 1:19 PM, David Ribes <dribes@gmail.com> wrote:
Can anyone offer some leads on the use of the term 'domain' within computing and information circles? Such as 'domain scientist' or 'domain knowledge'...
I've used the term for years as an in vivo category drawn from my fieldwork, but have little sense of its historical genealogy. It was already well in place by 2003 when I started fieldwork in the worlds of cyberinfrastructure, and I have some suspicion it may have come from knowledge management/business ...
-- David Ribes Associate Professor Department of Human Centered Design and Engineering (HCDE) University of Washington http://davidribes.com
_______________________________________________ This email is relayed from members at sigcis.org, the email discussion list of SHOT SIGCIS. Opinions expressed here are those of the member posting and are not reviewed, edited, or endorsed by SIGCIS. The list archives are at http://lists.sigcis.org/pipermail/members-sigcis.org/ and you can change your subscription options at http://lists.sigcis.org/listin fo.cgi/members-sigcis.org
participants (4)
-
David Ribes -
Evan Hepler-Smith -
Jon Lindsay -
Nathan Ensmenger