Most Optics Express authors don't know and couldn't care when it was invented, but it had to start somehow. It was the first. There were no previous examples of a regularly published, peer-reviewed, open access journal in the physical sciences. It has served as a model for all the ones coming after, and this is a short account of the beginning. Figure 1 shows what Optics Express looked like when a reader linked to the first issue via Netscape (who remembers Netscape?).

 

Fig. 1 Cover of Optics Express, Vol. 1, No. 1 as recovered from Netscape.

Download Full Size | PDF

Obviously, without the Internet there would be no open access journals today. (Quick note to many readers: long ago there was a time BTI, or “Before the Internet”!) Those of us who lived through the Internet's creation, and the interruption, reorientation, and liberation it brought to our working lives, have a hard time remembering what it was like because the Internet Revolution occurred so quickly. In the middle of the revolution, in only three years of inspired effort between 1994 and 1997, and without a model to follow, OSA wrote the playbook for open-access journals. The uncertain decisions that formed Optics Express turned out to be magically correct. Most of the original features remain in place, without significant change, and their correctness is easily judged — many of the open access journals in science today have adopted the Optics Express model. Naturally, this did not happen accidentally, and it wouldn't have happened at all if not for the dedication of Michael Duncan and Martijn de Sterke, as the second and third Editors-in-Chief of the Journal (see Fig. 2), who skillfully managed the continuously increasing flood of submissions, shown in Fig. 3.

 

Fig. 2 Optics Express Past Editors-in-Chief (left to right) Martijn de Sterke, Joseph Eberly, and Michael Duncan.

Download Full Size | PDF

Even before the 1990s OSA was becoming aware that new possibilities for publication were emerging. How to exploit them was far from clear. With some background prodding by an ad hoc Publications Technology subcommittee of the Publications Council, the concept of “electronic publishing” became a topic of discussion, and then at the 1994 Leadership Conference in San Francisco it was voted as OSA's top near-term priority. However, there was confusion about exactly what “electronic publishing” meant. Suggested options included an electronically accessible Bulletin Board for sharing communications about optics. This had the attractive new feature of permitting immediate connection between an author and potential readers. But a mere bulletin board was judged too trivial and unlikely to maintain the quality standards of OSA. A proposal to create an actual journal, already expressed as early as the same Leadership Conference, gradually became accepted.

However, the character of an “electronic journal” was completely undefined. The phrase “open access” was then unknown, and it had no obvious meaning even after it was first expressed. An enormous range of conflicting alternatives were imagined for an electronic journal. Would it be serious or informal? Regular or irregular in schedule? Who would be permitted to submit to it? What kind of content would be allowed? Would peer review be mandated? Who would monitor format and length? What would “publication” mean? After publication, how would citations refer to the articles? Should OSA build a permanent archive of articles, or would a 12-month public exposure be enough? Was the journal's purpose mainly to promote OSA or mainly to disseminate new work? Would it be identified with a fixed range of topics within optics? Which existing OSA journal would it compete with, if any? And naturally, a very important question, how would it all be paid for?

These policy questions needed answers in short order, but technical questions were even more challenging because technical capabilities kept changing rapidly. It now seems inconceivable that there was a time when no one had an email address. The World Wide Web (www) wasn't invented until 1990. Mosaic was the only browser initially recommended by Optics Express because all alternatives were slow or unstable. Before long, Mosaic morphed into Netscape, and Microsoft began to get Internet Explorer under control, and then AOL captured Netscape, and Netscape's defectors at Mozilla gave us Firefox. Apple's Quicktime was initially the only reliable movie platform, and remained so for a long time. Does anybody remember the creation of the pdf and the wide astonishment at both its ease of use and its free availability from Adobe? Personal computing was quite uncommon and standardization was unknown among text editors and browsers. Between then and now, there were at least three commercial text editors in widespread use: WordPerfect, WriteNow, and Microsoft Word. The disappearance of two of them is still lamented.

Some interesting proposals were suggested and discarded. More than one of these had to do with refereeing. An exotic possibility, which I gave up reluctantly, was to appoint a secret panel of experts in optics, to be changed every year. Each expert would agree to review 20 papers in the year. Submitted papers would not be assigned among the expert reviewers, but rather simply listed on a private website open to the reviewer panel. Any panelist could choose one of the submissions for review and review it in the usual way. But if a paper stayed on the list for a month without being selected by any panelist, it would be returned to the author as not interesting. This was thought too weird for consideration. Openly signed reviews were also proposed. A restriction of reviewers to the OSA Fellows was another idea. The old standard anonymous-reviewer method was finally agreed on. Like democratic government, it's full of flaws but overall seemed better than any alternative.

To make sure that the journal, not yet called Optics Express, would actually work around the world, in 1996 I recruited test sites connecting with each other and with fake editors at OSA to run alpha tests and then beta tests of the system as it handled fake articles by fake authors submitted for fake review. Colleagues and their grad students pitched in to do the testing in a variety of international locations, including sites in Germany, England, Mexico, and Russia, in addition to those in the U.S. Their tests identified key gaps and glitches in the software. This was at first impossible for the in-house code developers to believe, because they were testing from one in-house PC to another in-house PC using a primitive Microsoft Word and Microsoft Internet Explorer and thereby could prove that everything worked perfectly! It was an ongoing nightmare for OSA staff to explain and referee the conflicts between the sites, and it frustrated the external testers. The efforts of those testing groups using different text editors and browsers were truly essential. As one could expect, no alpha test was close to successful. But the same was true of the beta tests — none of them were successful either!

The journal's design and construction were based in OSA's IT Department, partly on the original logic that Optics Express wouldn't be a “real journal,” but a useful experiment. This made it difficult to explain why a fully electronic product needed such an old-fashioned element as a front cover, but I was sure that the more Optics Express looked like a journal, the more comfortable submitters would be with it. We could even animate the cover, and did so for issue no. 1, to illustrate that not only color figures but video could be incorporated in articles free of extra charge. The thinking that justified the considerable expense of development was that many of the necessarily new electronic back-office features, such as making staff-reviewer and reviewer-editor exchanges systematically Internet compatible, could later be applied to all of OSA's “real” journals. It would save confusion and grief to have all the experimenting done off to the side by Optics Express. Staff participation and coordination was managed, seemingly effortlessly, by Deborah Herrin and Ghassan Rassam in the OSA Publications Department. It's nice to report that migration of all OSA journals into the Optics Express system occurred in 2004.

The matter of the journal's name was a dormant issue at first, but I realized a name would be necessary and was trying to find a good one. In my initial focus on the restrictions that author-formatting would impose, I realized that the TeX and LaTeX editing systems would provide easily controlled formats. As a LaTeX user myself, inspired by its elegance, I proposed OpTeX as a journal name. This was before I realized what a tiny fraction of OSA's membership even knew that TeX existed and, more important, what intense negative reactions TeX aroused among experimenter friends, who vowed never to submit to any journal based on it. So I understood that a name with TeX in it could create negative impressions, but my next idea was close: OpTiX. I still think this is ok for a journal name, but a search showed that it was already a registered company name, of an OSA corporate sponsor no less. Next came Optics Express as my third tentative suggestion, and it attracted no negative reactions, so the development process gradually adopted it as official.

Obviously publication had to commence at some point, and development had to stop. Since the beta tests continued to fail, an original plan to begin publication in January 1997 had to be shelved. However, a sense of urgency was rapidly building. We needed to show concrete results from all the time and money spent. Something had to be made to work, so an executive decision was made by Publications Director Rassam. He determined that publication would begin at a date that I remember was March 1, which then I think was changed to April 1, the start of the second quarter. But April 1 is not the date to start anything serious, so it was again shifted, to the second half of the year, beginning July 1. That decision put intense pressure on both the staff, to make the system work independent of the beta tests, and on the new Editorial Board volunteers, to find scientists and engineers who would write papers that could be published. The mandated author-generated formatting was a novel challenge for writers, but we expected that it would permit a notable reduction in the time to publication, which turned out to be true, and through the first year the average was 52 days. This was very attractive to potential Optics Express authors, and it remains so, with a current time to publication of only slightly more than 60 days, despite the enormous increase in article volume, now the norm, that none of us would have dared to predict.

It is now nearly impossible to recall how persuasive one had to be to induce a close colleague to submit a real paper to a journal of a type no one had ever seen before, and to do so pretty quickly! Moreover, referees willing to deal with a dramatically new reviewing system had to be found for the papers, if and when they would be submitted. The Editorial Board played a heroic role here too. Almost all members (David Boas, Thomas Brown, Charles Clark, Michael Duncan, Roger Falcone, Mikhail V. Fedorov, Michael Feit, Bruce Shore, and James C. Wyant) were consistently available for reviewing and were steadily trying to find paper writers. A suitable number more or less miraculously appeared and survived a very unstably linked review network, held together mostly by force of will of the staff, with Production Editor Frank Harris on the front line. As July approached, I knew that July 1 would not be possible, but Dr. Rassam was adamant. My insistence that Optics Express should look and feel like a real journal allowed me to grasp a straw and assert that since the journal would publish regularly we should choose a day of the week, perhaps Monday. This allowed publication to start in July as mandated, and Monday was chosen simply because it was the latest of the first seven days in July that year.

Fortuitously, it also meant that a weekend would immediately precede the first issue. The weekend turned out to be anything but free for the staff. Some current OSA staff members still remember the frantic agony of the hours spent late into that Sunday night, even through Monday, patching together the articles that had survived review and could be published to the Internet, with correct formatting and with a movie on the cover and, most importantly, a Monday dateline. The way it all came together is shown in Fig. 1.

In recent years, the success of Optics Express, now under the leadership of its fourth Editor-in-Chief, Andrew Weiner, has prompted the emergence of many competitors issued by other societies and commercial firms, as well as the successful launch of spinoff Express journals by OSA in the key fields of biomedical optics and optical materials. And the enthusiasm for Optics Express itself continues. Statistics from the first 10-year period show in Fig. 3 how rapidly the eventual big success was being realized. The early visionaries in OSA who supported us and assisted in the creation between 1994 and 1997 of something really new deserve thanks not just from me but from the entire optics and photonics research and engineering community.

 

Fig. 3 Record of Optics Express’ growth from the beginning.

Download Full Size | PDF