Arne Hessenbruch: Compromises and opportunities: history of recent science and technology on the web [A first draft of this paper was submitted on 31 March 2002 to Jeff Hughes & Thomas Söderqvist, Writing Recent Science: The Historiography of Contemporary Science, Technology and Medicine. The paper contains an overview of the historiographical issues in the making of this website, and deals in particular with our need to get collaboration going using our software "modules", i.e. interviews done on the web. It has been adapted slightly here. Do contribute to our discussion group on this topic!]

1. What can digital storage and the internet bring to history of contemporary science?

2. Print vs. digital copy: fixity and trustworthiness (no problem)

3. The root problem: vastness and specialization of contemporary science. Possible solution: collaboration between scientists and historians.

4. Collaboration using the web

5. Problem One: Why would scientists want to collaborate?

6. Problem Two: The integrity of the historian

7. Conclusion

1. What can digital storage and the internet bring to history of contemporary science?

The problem staring historians of contemporary science in the face is the sheer size of information and the difficulty of coping with the esoteric technicalities within each scientific specialization. [1] Recent science is a huge and very diverse terrain. Our project experiments with the internet in order to address this primary problem. [2] The main two ideas are that

1. The electronic medium is capable of storing data in a smaller space than paper along with making that data more easily searchable.
2. The internet facilitate communication and collaboration between historians and living practitioners of technoscience so as to collaborate on the writing of history. Perhaps not surprisingly, we have encountered problems during the pursuit of these two objectives that illuminate some of the important issues of contemporary science.

With regard to the management of data we come up against the generally held expectation that electronic data are not as safe as data kept on paper primarily because of the issue of obsolescence. Much data collated by scientists as late as the 1980s is held in a format that has already been rendered unreadable by subsequent technological development. If one wants archived material to be available on a website, then one has the additional difficulty that a webpage is considered less permanent than a printed work. Compare the two references given in the endnotes to this paragraph: are you confident that in 50 years time both will still be valid? In my humble opinion, the problem of obsolescence of certain IT standards and the impermanence of digital records is not fundamental. I will first rehearse this argument, because it is directed against the most commonly encountered objection to the use of digital media. I will do so comparatively briefly and then turn in the main part of this article to a description of my own ongoing experiment of collaboration with actors in my field of study: materials research.

Our second idea, the use of the web as a collaborative tool between historians and scientists is even more fraught with methodological problems, and I will argue below that historical practice would need to change to accommodate new demands; an accommodation going against the grain of the integrity of historical inquiry.

Do have your say here.

2. Print vs. digital copy: fixity and trustworthiness (no problem)

While the faith in the digital medium has been increasing gradually over the last few years, there is still more status in publishing in print than on the web, and certainly historians prefer to refer to literature in print, rather than a webpage. But it is important to remember just why it is that the printed publication is considered trustworthy and permanent. To see this, it is useful to examine the history of the book in an age when fixity and trustworthiness had not yet been established in order to illuminate the present development of the web. Adrian Johns has examined the development of conventions of handling and investing credit in textual materials in early modern England. The very first paragraph of his book makes us view many aspects of books that we normally take for granted as actually quite surprising:

Pick up a modern book. This one will do: the one you are looking at right now. What sort of object is this? There are certain features about it of which you can be reasonably confident. Its professed author does indeed exist and did indeed write it. It contains information believed to be accurate, and it professes to impart knowledge to readers like you. It is produced with its author's consent, and it is indeed the edition it claims to be. If the dust jacket announces that it is the product of a given organization - in this case the University of Chicago Press - then this too may be believed. Perhaps you may even say to yourself that that fact vouches for the quality of its content. You may safely assume that the book you now hold will have been printed in many copies, and a copy of the same book bought in Australia, say, will be identical in all relevant respects to one bought in the United States or in Great Britain. [3]

Johns describes how we came to place trust in all the above aspects of books over a long period of time and he argues that trust in books developed as a new kind of civility and sociability emerged. The convention of authorial copyright is the best known convention to emerge in Johns' account, but it is in fact just one small part of a quite sweeping change. For that reason Johns' book is a voluminous, and one lesson we might take from that fact is that the establishment of textual fixity is generally a complex and long drawn out process.

The internet also cannot achieve such fixity in a short span of time but requires the development of new conventions, or new forms of civility and sociability as Johns has it, to achieve a similar kind of status as a trustworthy and stable conveyor of information. The address of our website (http://hrst.mit.edu) contains the reference to the Massachusetts Institute of Technology, which, you may say to yourself, vouches for the quality of its content. There is of course a difference between a University of Chicago Press book and an MIT website. The University of Chicago Press will have asked referees to assess prospective authors whose submissions will have been further evaluated by readers deemed independent; drafts will have been criticized and amended; and maybe several proof-readings will have taken place. By contrast, an MIT website has much less of an infrastructure in checking and rechecking uploaded material. You may of course trust that website on the history of recent science under the aegis of MIT will not be downright flippant, and I have been careful to state the credentials of staff and advisors on the site, ranging from PhDs to a Nobel Prize. But the infrastructure (sociability) remains less elaborate than that of the University of Chicago Press and so readers may be forgiven for placing less trust in our webpages than they do in a University of Chicago book. The same difference also explains that much less credit accrues to the author of a website than to the author of a book published by a reputable university press. But with the development of the same kind of infrastructure, I don't see why a website should not become just as credible as a printed text. In fact, one of the stated aims of the Sloan Foundation, one of our funders, is that one day PhDs may just as well be given for a website as for a printed linear narrative.

A further fixity in the book is currently lacking on the web that hinders publication of historical work there: the reference to printed material is unproblematic whereas the reference to a webpage is illegitimate in much academic work, because the permanence of the webpage is not assured. Again, it is just a question of an infrastructure propping up the fixity: The reason that we have faith in a textual reference is that books are printed in at least dozens of copies and distributed amongst just as many libraries across the world, so that even if a dozen libraries were to burn down, we would still be able to locate a copy in one of the surviving ones. Furthermore, we allocate resources to keep books in controlled environments (temperature, humidity) and have an infrastructure of searching, the most prominent of which is the library catalogue. The bottom line here is that the print medium also would not exhibit the fixity that we are accustomed to had it not been for the constant maintenance of librarians and building services.

Digital media require the same kind of maintenance, predicated on the continuous allocation of resources. Of course digital data have to be backed up (in multiple copies) just as multiple copies of books are deposited in a range of libraries, so that several physical copies of the digital data exist. And of course digital storage media need care and maintenance, and need to be kept away from humidity and extreme temperatures, as books do. No difference there. But digital media have the additional problem of obsolescence: On our site, for example, we have many pages kept in html format and we are constantly adding material such as text, sound (in Real streaming audio format, the back-ups kept as mp3s), image (usually jpgs, sometimes with back-ups kept in higher resolutions, and video (in Real streaming video format, with back-ups kept as avis) to our databases in various formats. We cannot rely on these formats being usable in technology of the future - in fact we can predict with some confidence that the formats currently in use will not be the formats of the future, and consequently the material stored in these formats will become unreadable with future technology. However, it is technologically comparatively easy to transform data in one format into another. The problem mostly lies in the labour-intensive nature: if we amass a great deal of data it might well take a lot of work to migrate it into new formats. In other words: the infrastructure needed to maintain digital archives must include provision for routine migration of data into new formats. It is not essentially different from care of paper archives except that in the conventional case one maintains a physical object, e.g. a book, whereas in the digital case the physical object is just a vessel to be discarded and only the digital contents of the vessel are kept.

We also know how to deal with copyright in printed material whereas in the case of the internet it is not yet so clear. Because the multiplication of digital material can be done with such ease, it presents different legal problems from those posed by the photocopier and the book. Many publishers currently investigate how profit could be made with publications on the world-wide web. MIT Press has experimented with the use of the internet as advertising by publishing their books simultaneously in a print format (books for sale) and on the web (accessible for free), and they have evidence to show that the web-publication actually enhanced sales of the printed version - an outcome that is perhaps surprising. Others are experimenting with encoded data files sold to users with an in-built blocking mechanism disabling the copying of such files. [4] On our site, we host contributions from many individuals, and we have also had to establish procedures to deal with copyright issues. All contributors are required to click through (the internet equivalent of the legal signature) a statement containing the following:

You agree to conduct yourself on the Program Site in a manner that will not interfere with any other Contributors' use and enjoyment of the Program Site or their ability to participate in a respectful and polite discourse about ideas. You will not post or transmit material that is libellous, defamatory, obscene, racist, fraudulent, harmful, abusive, threatening or hateful, that contains nudity or pornography, that violates the copyright and other intellectual property rights of others, or that is in violation of applicable law. [5]

Such procedures constitute an element of civility or sociability strengthening general trust that the web-authors do indeed exist, that they did indeed author the material posted, and that that material is of a certain quality.

To repeat, fixity in web-publications does not come overnight but requires an infrastructure and a set of behaviours along with a set of punitive practices to deter those who don't play along, just as was the case for the book; hence the management of digital data is not fundamentally different from that of print media: both require an infrastructure and resources for routine maintenance. The management of digital media is basically unproblematic.

Do have your say here.

3. The root problem: the vastness and specialization of contemporary science. Possible solution: collaboration between scientists and historians.

This is not the case with the use of the internet as a collaborative tool in the history of recent science. For this to work, the changes required in our current sociability are somewhat painful. But the problems themselves are interesting and may well give an insight into the avenues that history of science has to pursue in order to survive as an academic field.

Our project actually consists of five sub-projects, devoted to large and small topics: the Apollo Guidance Computer, Bioinformatics, Materials Research, Molecular Evolution, and Physics of Scales. I can only speak authoritatively for the sub-project on materials research, and the nature of the topic of investigation differentiates the historiographical issues from, say, physics of scales, which is comparatively small and well-defined. Materials research is a very large field and highly interdisciplinary and in that respect it is an excellent topic if the aim is to learn about problems in historiography of recent science, since contemporary science is precisely characterized by inter-disciplinarity and vastness. I will sketch the history of materials research and point out some of the salient points about publicity on materials research before turning to history of contemporary materials research and to the intervention that it might lead to in the public representation of the field.

Since about 1960, there have been institutions with the name "materials", and we [6] confine ourselves to the period after 1960. This is not to say that materials have not been investigated before: the fatigue of metals was certainly an important area of investigation during the railway boom of the 19^th century and one can successfully find research on materials in just about any age. [7] What happened around 1960? There are two (conflicting) explanations: Bernadette Bensaude-Vincent has argued that in the wake of the Sputnik crisis military funding created new inter-disciplinary laboratories, many of which achieved permanence on university campuses. In other words, the discipline of materials science and engineering was a top-down creation by funding agencies, generating the infrastructure within which a new identity was (and continues to be) forged. [8] Bernhardt Wuensch, Professor of Ceramics at the Department of Materials Science and Engineering of MIT, has a different perspective [9] : a materials generic approach emerged around 1960 because researchers in such diverse fields as metallurgy, organic chemistry, semiconductor physics, and ceramics were developing conceptual tools highlighting relationship between structure and properties [10] ; tools that could be applied in all fields. A number of instruments, such as x-ray diffraction and the electron microscope, also gained widespread usage in all fields. In other words, the above diverse disciplines began to merge because the internal development was conducive to increased collaboration and sharing. When ARPA and other funding agencies supported this development they were merely enhancing the natural development.

Indeed the first textbooks of materials science and engineering began by treating families of materials (metals, ceramics, polymers) separately, before addressing properties. [11] More recently, textbooks have paid less emphasis on the specifics of families of materials and focused more on general (materials generic) structure-property relations. A recent textbook by Allen & Thomas [12] proceeds from very general statements about symmetry and chemical bonding and proceeds to cover in turn three states (non-crystalline, crystalline, and liquid-crystalline) each of which apply to all families of materials. They then cover processing and assembly, a topic at the very applied end of the pure/applied spectrum, and finally microstructure including imperfections.

The materials generic approach thus spans several sub-disciplines that one would ordinarily have located either within physics, chemistry, or engineering. In the 1990s the manipulation of biological materials on the nanoscale has been accreted, so that materials science truly ranges across the whole spectrum of the natural sciences. [13]

Because it is a discipline of such great diversity, materials researchers are particularly observant of resources becoming available to them in neighbouring fields. The annual meetings organized by the Materials Research Society (MRS) reflect this aspect of the field: they are giant affairs with some 4,000 attendants organized into some 25 symposia of sub-fields that are considered hot. Spring Meetings are held every year in San Francisco and Fall Meetings every year in Boston. Different topics within the super-discipline of materials research are covered in Boston and San Francisco; semiconductors are emphasised at San Francisco with its proximity to Silicon Valley. These meetings are certainly markets and trading zones [14] where researchers strut their stuff and keep their eyes peeled for new resources, in terms of ideas, methods, tools, and prospective collaborators. I attended the 2001 Fall Meeting in Boston with a poster advertising our history of materials research, and I keenly attended a number of sessions. Even though all the symposia started off with a paper giving an overview of the topic to be covered, I found myself incapable of following the goings-on, sometimes after less than a minute. My first degree is in physics, and in the course of our project I have been reading materials science and engineering textbooks and attended the elementary materials-related course offered at MIT; which is all to say that I'm not a complete neophyte and yet I had the disappointing experience of being excluded by incomprehension of the technicalities. I was relieved to hear from Gopal Rao, an attendant to the conference with a PhD in Materials Research that even he was soon lost and had very much the same kind of experience as I had. The specialization within materials research is so great that no-one comprehends the entire field in any great detail, and yet the researchers actively search for resources and work hard to find out what might be of relevance to them.

One way of talking about this is by using the notion of the public sphere. The first to analyse the nature of esoteric knowledge and popularization was Ludwik Fleck. [15] He conceived of two concentric circles, the inner one of which represented esoteric science and the outer the "exoteric", which is the sphere of public discourse, where lay people discuss science in a popularised format. Fleck was primarily interested in explaining the development of a fact, in the sense of something that was not doubted, that everyone agreed upon. He argued, with the help of a case study on the Wasserman test for syphilis, that facts were not generated in the esoteric sphere alone. Another way of putting it is that he argued against the notion that pure science creates knowledge that is henceforth propagated outwards in a popularised form, that is as attenuated or vulgarised (as the French term for popularisation has it). Rather, esoteric knowledge is first produced in a tentative form not accepted by everyone; it is not yet a fact. It is then communicated to the exoteric sphere in some form or other that makes sense to the non-experts. But this reformulation actually redefines the esoteric knowledge and gives it another level of meaning. There is also a feedback mechanism back into the esoteric sphere. One analogous and obvious understanding of Fleck's feedback mechanism is the reformulation of research results to government bodies who will decide over future funding and thus literally determine whether some finding will be substantiated or not.

Steven Shapin has analysed science and the public sphere with an especial focus on the negotiation of who counts as a competent practitioner of science and who counts as lay, and points to a historical development where he finds a gulf opening up between the public and esoteric science in the period of 1860-1880. In terms of Fleck's images of two concentric circles, one might illustrate Shapin's analysis as the constant change of the boundary between inner and outer circle, a change that has both seen a blurred boundary becoming more defined and also a shifting of that boundary.

Shapin shows little interest in communication of science as a way to reach scientific colleagues, the kind of communication that patently occurs at the annual MRS Meetings (Ulrike Felt does: she points to the enabling of new and daring cross-connections as a motivation for science popularisation. [16] ). I think popularisation of materials science occurs in both cadences: 1) Overtly the simplified presentation in one symposium of an annual MRS Meeting provides resources for interdisciplinary work - overviews to an eager public of non-experts, who are nonetheless experts in neighbouring symposia. (Fleck's two concentric circles really have to be replaced by a kind of Venn diagram of expertise with many small circles that together cover the entire field of materials research. 2) Implicitly status, expertise, disciplinary boundaries, access to funding and educational structures are being negotiated. Historians and sociologists of science have mostly paid attention to the implicit negotiations and less to the overt communication, and they tend to read the introductions to scientific publications with greater care than the "main" part. In this sense, historians and sociologists are akin to occult scientists of the early modern period who specifically looked for implicit and meaningful messages in texts that on the surface were quite mundane. [17] As I will argue, both cadences of communication are important for a web-based history of materials research.

Do have your say here.

4. Collaboration using the web

Our project, History of Recent Science and Technology on the Web, is conceived as an experiment in the writing the history of recent science using the tools of the web. We have a team that develops software for the purpose, consisting of Babak Ashrafi, Daniel Tsai, James Voelkel, and Paul Warner. The software has two prongs. One relates to the gathering of information, one relates to its management and display. The software for gathering information consists of a set of modules that function as interviews, discussion groups, collaborative timelines, and the like. The aim is to have individuals contributing to the writing of history regardless of their location. For example, the interview module works in the following way. A set of questions can be posted on a webpage. A visitor to that webpage can reply to the questions and these replies are almost automatically posted on the same page. The replies can include text and images. The interview can be set so that any visitor will be able to post replies as long as s/he has registered on the site which includes identifying oneself and clicking through the legal statements mentioned above. The interview can also be set so that only a group of specified individuals are empowered to reply. One can make further variations upon this theme, so that a select group is empowered to reply and all registered users have the right to add comments to the replies. These comments could be displayed in a smaller font. The software is very flexible so that many permutations are conceivable and it is really up to historians to decide which module is best suited to specific historical tasks.

Perhaps not surprisingly, this kind of collaborative concept is not easily carried out and will not necessarily lead to good history (whatever that means). The devil is in the detail and most of the following will address the problems. But let me first finish the description of the software, turning to the way the data is displayed. It might be well to emphasize at the outset that webpages look different for each user visiting the site, since the format of webpages (currently primarily HyperText Markup Language, html) leaves some of the display to the user's computer that adapts the data to the specificities of the user's screen. [18] There are design issues for the author of webpages, in that currently webpage users are more likely to look for small bits of information on the web and not inclined to read long and complicated narratives. Posting a long narrative on the web will thus currently make sense only if it can be printed out. This should shape the output of the historian but that advice will very likely not be heeded by contributors to a website unschooled in the writing of history. The topic is clearly a can of worms but I will not address it since it is not important for my concerns here.

One of the great strengths of the web format is the searchability of words and the ability to connect relevant information with the help of the hyperlink. All contributions to the website are deposited in a database and allow for efficient searching. One outcome hoped for is that with the growth of the database a search will reveal unexpected cross-links between fields as diverse as bioinformatics and materials research. Our software team is putting some effort into the issue of automatic hyper-linking. We have built on software by the Perseus Project [19] , located at Tufts University and headed by Greg Crane, that can automatically tag words (turning them into hyperlinks) in a new document posted on the website (and thus added to the database). If one were to click on that word, one might for instance get a menu containing a list of the webpages on which that same word is used and with a further click one can be taken to any of those pages. Perseus initially developed their software for philological purposes, especially ancient Greek texts, where it makes sense to tag every single word. But in history, some kind of human intervention is required; for example, a list of all pages using the word "and" is not very interesting. The pragmatic solution could be that an historian provides a master list of words, and if the system finds any of those words in a newly submitted text, it is tagged. To conclude, displaying information on a collaborative history site one has to adapt to the nature of the medium and the long, one-dimensional narrative will lose out. In return searchability and the easy link-up with relevant material will improve.

Do have your say here.

5. Problem One: Why would scientists want to collaborate?

Unfortunately, that was not the case. My experience at the MRS Fall meeting was particularly stark in this respect: the organizing institution, the Materials Research Society, was very supportive but individual attendants at the conference showed no interest whatsoever. I had booked a presence with a poster describing our project and inviting people to contribute to the website. With the help of Babak Ashrafi I also had a laptop hooked up to the internet, so that conference visitors could browse the site and respond to an interview set up for the purposes on the spot - we also handed out flyers containing help on how to do the interview from home. The MRS advertised our poster session very generously in the conference program, the website, and on posters throughout the conference building. But out of the approximately 4,000 attendees, a big fat zero responded. We learned two important lessons from this experience, the first disappointing (at first), the other encouraging:

1. In order to involve living materials researchers we have to present them with something that is geared more to their interests. Our own historical questions will not do.
2. The Materials Research Society staff was very interested in what we do and actively support us, as long as we don't ask them for money.

I decided to approach the MRS more formally in order to discuss whether we might be able to do something historical that would be of mutual benefit, and the MRS staff, fronted by the editor of the MRS Bulletin, Betsy Fleischer, was very open. It seemed appropriate to first of all define our interests and to pinpoint any potential tensions. The explicit interests of the MRS lay in creating texts of interest to their members. They had had a historical section in the Bulletin that had been scrapped as an expenditure consolidation measure. They were very interested in having historical material especially if we could pay for it. When I suggested that the MRS had an interest in cementing the identity of the field, the suggestion was accepted but I detected a little surprise and hesitation - perhaps it was an unusually raw or alien way of putting it. The point seemed to me obvious because of the interdisciplinary nature of the field and the great diversity of academic departments and private laboratories in which materials research is done. It is also implicit in discussions about the structure of textbooks and the conflicting views of Bernadette Bensaude-Vincent and Bernhardt Wuensch mentioned above (materials research created from above through military funding or out of an internal rapprochement in conceptual and experimental tools). The 2001 President of the MRS had explicitly stated identity as a primary concern of the society in the December 2001 issue of the Bulletin:

I am conflicted about the role and identity of the MRS. [...] What is the nature of MRS? I suspect being a member of the Materials Research Society is not the same as being a member of the American Physical Society (APS) or the American Chemical Society, for example. About half of our members view MRS as their second home; they vacation with us for the meetings, but don't join if they attend neither the Spring nor Fall meeting. [...]

Sometimes I think of materials scientists as the mutts [22] of the physical science world. After all, our field is a mélange of science (e.g. physics, chemistry, crystallography) and engineering (chemical, electrical, mechanical). Its existence at the intersection of many pure disciplines (see what I meant about mutts?) is what attracted us to the field in the first place. However, I'd like to point out that the very nature of our field has repercussions for its survival. [...] A strong MRS will provide a strong identity for our field. If we lose our identity, our students and our funding will be backscattered to the departments that spawned our field. In fact many smaller materials departments have already been absorbed by their neighbors. [23]

Our interests may be stated in a similarly raw fashion: to promote our careers by helping the project to succeed, for example by devising a sensible way to utilize the modules. The MRS agreed to brainstorm to find ways of using the modules to create material that may also be used in the MRS Bulletin and on the MRS website. [24] Our first plan is to examine what might be termed the accretion of the field over the last 20 years. Just as biomaterials have become a part of materials research over the past decade, lesser sub-disciplines have also emerged and been brought into the fold. The MRS Bulletin is a particularly useful tool for tracking this, since each (monthly) issue focuses on hot topics. [25] We are now planning to select some 20 topics for an historical investigation. For each topic, we will take advantage of the MRS network of people and reach out to the broader materials science community. We will likely pick out one prominent researcher and interview him or her in the time-honoured face-to-face manner. This interview will be conducted jointly with the MRS staff, and we will prepare questions together. This process will in itself certainly be very illuminating for us and perhaps also for Betsy Fleischer and her colleagues. We will then target half a dozen other people in that field and set up an interview module on our website that only they will be empowered to post responses to. Finally, we will advertise to all MRS members and elicit their comments. [26] We, the historians, need the MRS at several stages of this process. We obviously need their network and we need to operate under a joint aegis so as to gain clout. But we also need to adapt to the kinds of questions that will get materials researchers interested, and they may well turn out to be very different from the ones with which historians of science regale each other. We bring them funding, entertainment material, and maybe also strategies for negotiating disciplinary boundaries.

Do have your say here.

6. Problem Two: The integrity of the historian

Is there a problem here? Am I compromising the integrity of the historian by sharing the decision on what historical data ought to be gathered? Am I losing my grip on the interpretation itself by opening the door to professional scientists who, in the self-understanding of professional historians of science, are prone to Whiggism and self-interested accounts? There are certainly some who do (did?): for example, as a graduate student I wrote a history of Siemens in the UK [27] , a history commissioned by that company, in my judgement with the aim of generating a corporate identity at a time of expansion and wholesale incorporation of thousands of employees from companies that Siemens had just bought up. Each employee received a copy of the final product. My supervisor at the time, Simon Schaffer, called my remuneration "blood money".

Allow me to get at this issue by first giving a history of interestedness in the natural sciences themselves, especially in physics. According to Sam Schweber, a physicist turned historian who has continued as a faculty member of the Department of Physics, Brandeis University, observes the following general development in the second half of the 20th century: early on, physicists were interested in questions of a philosophical or even theological nature, such as the Grand Unified Theory. Status was to be found in high-energy physics, related of course to military applications but also to such ethereal questions. By contrast, solid-state physics was low status. It took the basics of physics for granted; one might describe it as a field that accepted the basic quantum mechanical tool, the Schrödinger equation, and sought to apply it to as many real-life situations as possible. Sam Schweber:

I would argue that quantum mechanics differs from everything else in that it gives a highly accurate and stable foundational theory on the atomic level that can't be tampered with. Schrödinger called it a closed theory. You also know the basic entities populating that realm. There is a confluence of theory and ontologies. You can compute the anomalous magnetic moment to the precision of one part in 10 to the 12. You never need that accuracy in materials science and what I am trying to say is that the foundational theory is secure. That is the reason that pure and applied no longer exists. Nobody doubts the accuracy, veracity, efficacy, and efficiency of the foundational theory. There is nothing these people do that probes the foundational theory. [28]

The fact that many applications of solid-state physics opened up, especially with regard to semiconductors, did not enhance its status - perhaps rather the opposite. Gradually, a shift has taken place in which solid-state physics has improved its position in terms of proportion of faculty, the number of students, and status. Simultaneously a new way of funding research has gained ground. Solid-state physicists, whose research is of immediate interest to many private enterprises, have developed joint projects with industry, and that has to an increasing extent replaced the government funding that in the 1950s constituted the entire budget. This development has not taken place without tension, and one of the most eloquent critics of the process is Paul Forman who describes the development in terms of betrayal. [29] As a physics student in the 1980s, I sensed the same tensions: the solid-state physics sub-division actually paid students a salary when writing their final theses there. At the time I considered it a travesty: the role of the physicist as an objective inquirer driven only by curiosity was sullied by research in part determined by industrial demands. I wanted to be a physicist searching for the ultimate truth, not an engineer.

Solid-state physics is a field that overlaps clearly with materials research. Many MRS members doing semi-conductor research are physicists. If anything, materials research is even more integrated into industrial application than solid-state physics is. Many departments, such as the one at MIT, are called Materials Science & Engineering. The self-understanding of the field is explicitly one of attention to demand, and the claim is at the same time that the activity counts as science. In fact, the contention is, the boundary between science and technology is not clear, and the terms "pure" and "applied" simply don't make sense. From my perspective, one of the most interesting aspects of the history of materials research is its insistence on the irrelevance of pure and applied, and hence its contribution to the erosion of a boundary that seemed pretty clear in the 1950s.

Legend: The tetrahedron of materials science and engineering, in which the corners of Structure and Property (pure-ish) have equal weight with Performance and Processing (applied-ish), showing the conscious search for other concepts to describe the field than the polarity of pure and applied. From a 1990 government report [reference].

And yet, it makes me uneasy when the same boundary is blurred for me, as an historian. Collaborating with the MRS in the way described above goes beyond being paid a salary for writing an in-house history, it goes beyond the compromise of the journalist getting on with informants at, say, the mayor's office. [30] In both those cases, the writer is constrained by having to maintain social relations; for example, the journalist can write something very critical of the mayor's office only once before the source of information runs dry. The constraint lies only in the way in which the mayor's office is portrayed. In our project the compromise is at a much more fundamental level: it is in asking the scientific actors which questions to ask.

But one of the themes that have emerged from history of science is that the objective observer is not a self-evident category, that all observers are located in a specific institutional and social setting, and that objectivity is a judgement that people make on the basis of such constellations. For example, the scientist of the 1950s worked in an environment isolated from commercial interests and could pose questions freely. This academic structure, as I noticed in the above discussion of Fleck, is supported by tax moneys and justified by the conception of pure and applied: that pure science leading to fundamental insights will enable applied science and technology. In the course of the second half of the 20^th century, as governmental support for pure science has waned, many recipients of that support have revealed their interestedness by vocally supporting the concept of pure science. Similarly, as governmental support for history of science gives way to "outside" funding, does it make sense to cling to an ideology of historical work that posits the historian as disinterested? I would think not. [31]

Do have your say here.

7. Conclusion

Adrian Johns emphasized the fact that there is no inherent fixity of the print medium and that rather new sociabilities were developed over a long period of time and that only these prop up the fixity of print and only these are the cause of our trust in print. Similarly, the digital archive and the web-based historical collaboration cannot emerge as fixed and trustworthy overnight. Science historians' sociability would have to change. We would have to jettison attitudes that currently seem to prop up the identity of the historian: to accept whiggism back in through the back door, that we are in no way superior to the whiggish scientist. Historians of recent science would have to learn new technologies and new ways of interacting with the historical sources: it would no longer be a question of identifying sources but rather of collaborating with historical actors to determine what will remain as a historical source in the future. Scientists' sociability would also have to change in the process. The habit of interacting with the web is new and not yet stabilized (or will it keep changing?). There is no habit of collaborating with trained historians at all. In other words, a web-based project on the recent history of science and technology will only be able to succeed after a long period of time, in which historians and scientists learn new ways of doing things and renegotiate their roles.

Our project may find itself at a dead end soon, but it may also point a way to the future. It may well give an insight into the avenues that history of science has to pursue in order to survive as an academic field. One of them is learning new tools, and another is the collaborative project engaging many individuals, only some of whom have a historical training. This possible future is one in which history of science resembles more closely the larger projects of the natural sciences. But perhaps the most interesting and yet anxiety-provoking aspect is the deliberate search for outside interest (or perhaps marketability) that would take history of science out of its cosy but closed-in present situation where historians talk only to historians.

Do have your say here.