Disparate and Uncertain Group, Lacking Both Grandeur and Consistency’*
As I write this essay, the process of making the Vance Integral Edition is drawing to a close. It is a strange feeling: the VIE has been part of my life for over five years, since August 1999. By the time you have the final volumes on your bookshelves, some six and a half years will have elapsed since the first call for volunteers on Mike Berro’s website. The feelings of those of us who have worked on the project from the outset will, therefore, be somewhat mixed. Celebration and satisfaction will be uppermost—at long last, we can finally get back to reading Jack Vance for pleasure again!—but inevitably there will be a sense of loss too. For the VIE managers, a somewhat fluid group of perhaps thirty which on occasion has resembled the cabal of magicians in ‘Rhialto the Marvellous’, the project has been a big part of our lives.
If accomplishing this monumental task has seemed, at times, an arduous duty, in essence it has been a pleasure; a pleasure we must now forego. But our rewards are immense: the satisfaction of a job well done, 44 volumes that represent Jack Vance’s life’s work, and many enduring friendships. This essay looks back over the life of the VIE, the work of its volunteers and managers, and explains just what putting the set together meant in practical terms.
The VIE has been a volunteer project. Nobody was paid for the hours they put into the labour of love this set has become. Volunteers’ commitment ranged from many hours per week for several years, to proofing as little as a single short story.
Around 250 volunteers (not counting packers) worked on the VIE, roughly a quarter of whom only undertook a single assignment. But every volunteer’s contribution has been important, and every name you see in the credits lists following this account has contributed in a real way to the set of books you see now. The resources to fund a professional staff were never available for this project; the only way it could be brought to fruition was through the unstinting efforts of Vance readers around the world who loved Jack Vance’s work enough to give up a portion of their lives to it.
The VIE would not exist without them. This essay is dedicated to every VIE volunteer.
THE EARLY DAYS
Looking back to the far-off summer of 1999 and the genesis of the VIE, what strikes me most is just how much we underestimated the task. The idea for the VIE belonged to Paul Rhoads, a painter whose only qualification for undertaking it was his enthusiasm for Jack’s work. At this initial stage he had only a nebulous plan for its realisation. The original inspiration came from the sober, dignified and serious German editions by Andreas Irle—who was to become an important VIE volunteer.
Much has changed during the course of the project and many other people have made crucial contributions to the shape and nature of the edition. But Paul’s initial idea—a classic, complete and archival edition, produced by volunteer work organized on the internet—has remained constant. In 1999 most of Jack’s work was out of print; it was difficult to form a general idea of his art and artistic development. Extant editions were mostly crumbling paperbacks adorned with lurid covers which, by their nature, also constituted a block to the discovery of Vance by a larger public. There clearly existed a danger to the very survival of his œuvre in the longer term.
Paul was aware that some of the titles had been changed, and that there were a few textual problems here or there, but his estimation of this aspect of the situation was startlingly inadequate, and we soon discovered another important reason to create the VIE: commercial editors had rarely treated Jack’s work with either respect or understanding, and most of the published texts were more or less seriously degraded.
Paul, who not only lacked all editorial experience but was a computer neophyte, contacted Mike Berro in August 1999. Mike’s ‘Vance Information Page’ was already the most popular Vance site on the internet, a virtual meeting place for far-flung Vance devotees. My own first ever search on the internet (in common with many other VIE managers) was on “Jack Vance”, and it led me to Mike’s website. Mike was instantly enthusiastic for the VIE idea. He built the VIE site: it worked as an instant rallying-call and the results were overwhelming; subscribers and volunteers cascaded in.
While the VIE has continued to draw in volunteers throughout its life, a high proportion of the eventual VIE management had gathered within the first week.
Paul, christened ‘Editor-in-Chief’ by Mike, struggled to organise this quickly growing group, which included people of many skills crucial to the project, and ideas began to be fleshed out. Norma and John Vance, of course, remained a great source of help throughout; in particular Norma’s knowledge of Jack’s working practices, and her memories of certain editorial abuses, proved invaluable in textual restoration.
While coping with the first influx of volunteers, Paul had also to define essential work processes and to assign specific managers to implement them. Even at this early stage, Paul realised that if the project were to move from inspired idea to concrete conclusion, solid foundations would need to be laid before volunteers launched into helter-skelter effort. These foundations were designed by John Foley, who continued, very quietly, to guide and support the Editor-in-Chief through years of subsequent negotiations and manoeuvres. Without John’s vision and understanding of how to design a large and complex project, the VIE could never have come to fruition.
His contribution, or the problem it addressed, is not easy to articulate. Essentially it involved ensuring that volunteer talents were harnessed, but within the necessary standards, and in harmony with the project’s goals. At times this involved reconciling contradictory drives, trends and aspirations. The VIE could only have been viable with a large—and constantly replenished—stream of volunteers, who could be recruited only via the internet. The VIE had therefore to remain ‘open’ throughout its life to continue to draw in volunteers and maintain momentum. But recruiting the volunteers was only half—and the easier half—of the story.
Success would require that the volunteer effort was used correctly. Inevitably this involved imposing standards, procedures and even deadlines upon hundreds of unpaid individuals.
The scope of this organisation can be seen by glancing at the work credits which follow this essay. Without volunteers, the VIE could not have been commercially viable. The ‘true cost’ of the project, with the imputed value of volunteer labour, would have run into many millions of dollars. To hire specialists would have pushed the costs of the volumes beyond affordability; but beyond commercial considerations, it would in any case also have created an unstable situation were paid specialists mingled with volunteers. The project, to meet financial and operational considerations, had to be entirely volunteer. If the project had necessarily a hierarchy, it also required a certain equality among the volunteers; an equality of purpose if nothing else. We were therefore obliged to compensate with quantity for the necessarily somewhat ‘amateur’—a word used here in the noblest sense—quality of our personnel. Such an approach magnified the risk of individualism, solipsism and eccentricity. Everybody would necessarily have their own conception of the VIE, how it should be achieved, and their own place in the whole. Such individuals’ conceptions would inevitably conflict with others’ at times. The solution had to strike a delicate balance between remaining open, with individuals welcomed and respected, while providing robust mechanisms whereby debate—to say nothing of occasional actions motivated by less noble passions—could be resolved. This dynamic would depend on more than good intentions, and its implementation would depend on more than simple tolerance.
John Foley is one of a large group of VIE managers who were mostly invisible to volunteers and subscribers. He had worked as a professional publisher for Bell Labs, and is a long-standing friend of Paul’s—indeed it was through Paul that he came to discover Jack’s work. In the very early days John Foley defined and promulgated a ‘Master Plan’. The ‘Master Plan’ was more than a document. It was a set of basic concepts, the limpidity of which masked their wisdom only from the obtuse or inattentive. Paul Rhoads has admitted how, initially, he was dismayed at its tedious and cumbersome requirements; but time was to change his mind. The essence of the Master Plan may be expressed in four points:
In practice none of this was simple or orderly. We were obliged by the nature of our task to entrust many people with responsibilities without proper assessment of their capacity or zeal. It should not seem strange, and certainly not reprehensible, if disagreements and even frictions occurred, sometimes pursued beyond the point of politesse. It may or may not be a testament to the character-building virtues of reading Jack Vance, but we can be proud that those who formed ‘VIE management’ at the project’s outset are, for the most part, still in management at the end. Still, over the years it was sometimes necessary to change personnel. It was also necessary to create new teams and impose new procedures—as this essay will recount.
Responsibility for this, which fell on the Editor-in-Chief, was politically delicate, and team heads had to toil under great loads of work, managing often dozens of workers, all this augmented significantly by clerical procedures imposed by the Master Plan. It is a testament to John Foley’s statesmanlike foresight that the project was provided with a structure allowing us to achieve the task while coping with the human dimension.
That most team heads, including the Editor-in-Chief, were also volunteers on other teams was certainly an element in project harmony. We were, by turns, captain and soldier, soldier and captain. The absolute need for hierarchy—lubricated by at least mutual respect and usually considerably more—and rigid review and tracking procedures, became clearer and clearer to everyone, and thus more and more accepted. We were as democratic as possible but, thanks to the Master Plan, no more anarchic than allowable.
THE WORK PROCESSES – DIGITISATION
Initially Paul articulated VIE work in an unconventional style, using ‘Olympian’ names; as each new area of work opened up, and a manager was found to carry it out, he or she was assigned a name from the pantheon of Greek and Roman gods. Paul called himself ‘Vulcan’, conceiving himself as an artisan hammering out the VIE edition. My own Olympian name was ‘Cerberus’, after the three-headed dog who guarded Hades, jocularly bestowed when I was made the VIE’s gatekeeper (first VIE contact of each new volunteer or subscriber). Though this approach helped generate esprit de corps, it became too quirky and unwieldy as the project grew, and a more prosaic and formalised nomenclature soon came into use.
The Master Plan, with its yokes of auditability, double checks and reversibility, nonetheless allowed for flexibility. If something didn’t work, we changed it, or came up with something new. More rigorous checks and balances were added as the project developed. One reason that the first half of the project took so much longer to deliver than the second was our initially inadequate quality controls. It was important, at the beginning, to generate momentum, to get jobs done and to see what worked and what didn’t. With experience, a higher proportion of the experiments was successful, and events moved forward more swiftly.
Initial ideas of what constituted the component tasks were in retrospect laughably naïve. We conceived of the work in four simple steps: Digitise, Correct, Proof, Compose. Thereafter the books need only be printed and sent out to subscribers. It would be hard work, of course, but not especially complex. Two years seemed a reasonable estimate of how long things would take. If we had known then what we know now, would we ever have begun?
In retrospect the continuation of the VIE seems more assured than it ever did at the time. Revisiting some of our original emails uncovers a series of false starts and abortive work. A very early attempt to manage digitisation and proofing through one team proved administratively burdensome and procedurally flawed. It was not until digitisation was split off under the energy and enthusiasm of John Schwab that a degree of momentum was generated. I was then able to organise the first proofing jobs, although even here my recollections are of dead ends and impossibly impractical procedures before a smooth work flow began.
John Schwab’s Olympian cognomen ‘Hercules’ proved so apt for his stupendous labours that, alone of all the nicknames, it remained in use throughout the project.
Digitisation was the business of getting Jack’s printed texts into electronic form. For a few of the later texts, Jack’s original electronic files were available (although even these were not unproblematic) but for most texts, digitisation meant either scanning or typing the texts. Neither method was to prove entirely satisfactory. Digitisation was also the first opportunity to make use of the rapidly growing number of volunteers. We knew that, to keep our growing number of volunteers eager and keen, we had to keep them occupied and focused. John Schwab put them to work turning texts into MS Word documents.
The number of volunteers who could help with digitisation was limited by circumstance. Firstly, the volunteer had to have a copy of one of the two ‘preferred editions’ of each text. ‘Preferred editions’ were those which we thought were likely to have been subject to the least editorial meddling: generally one preferred edition would be the first published, and the second an Underwood-Miller edition. Since at this stage of the project our knowledge of editorial practices and procedures was inevitably limited, we could do no more than establish rules of thumb. That initial list of preferred editions was put together by the two volunteers with the best credentials to do so: Nick Gevers, an academic from South Africa with a professional interest in Vance’s fiction, and Alun Hughes, a librarian with a wide experience of bibliographical studies.
Secondly, volunteers needed access to a scanner and OCR (optical character recognition) software, though a few chose instead to type the files directly. My own first assignment on the VIE was digitising ‘Marune’, and after laboriously and inaccurately typing the first chapter, I soon secured access to a scanner.
By these means, texts began to move through the digitisation process. The next stage, as originally conceived, was the correction of the texts to remove all identifiable editorial intervention. This process, known as ‘textual integrity’ (TI), was carried out under the management of Alun Hughes. Alun felt that it would be premature to begin TI until we had a reasonable body of texts on which to work; until we could be sure the digitised texts were free of manifest error; and that a set of textual integrity guidelines could be produced to ensure that everyone was working to a consistent standard.
As more texts were digitised, more proofers were able to begin work and absorb VIE procedures. Even for those who did not move into management, there were great benefits when the final proofing effort began in 2002. However, despite the hard work of proofers, it became apparent that digitisation introduced unacceptably high error rates, which if uncontrolled would feed through into the volumes themselves. Many of these errors proved too subtle for spell-checkers: ‘h’ was frequently misrendered as ‘b’, so ‘he’ would become ‘be’, and punctuation was particularly prone to mangling.
As a result it was necessary to produce a further refinement to the system: double-digitisation, or ‘DD’. DD worked on the principle that if a text were digitised twice—from two hardcopy sources or two different softwares—and the results compared, an error would survive only if it were identically rendered in each scanning. These two DD scans were in addition to the original digitisation, so that DD ensured a minimum of three bites at the cherry. For this we have to thank Chris Corley, who first codified systematic scanning errors or ‘scannos’, which led Paul Rhoads to see the potential benefits of a second digitisation, and Richard Chandler, who put it into practice and led by example. In common with many VIE managers, Chris and Richard also undertook laborious text restoration work—Chris on Night Lamp and Richard on The Miracle Workers.
DD became the first of the VIE technological disciplines, spawning its own terminology. The two DD scans were subjected to a comparison process known as ‘jockeying’ to identify areas of divergence, and then compared (‘monkeyed’) with the original digitised text to root out all remaining digitisation errors. The results could then be fed into the textual integrity process for verification. Jockeying fell under the control of Damien Jones, while Suan Yong led on monkeying. These copious and unglamorous jobs were of tremendous significance in producing clean electronic texts.
THE WORK PROCESSES – PROOFING
The organisation of pre-proofing (The term ‘pre-proofing’ is anachronistic: at the outset we did not conceive of two separate categories of proof-reading. For reasons of clarity the two elements of proofing are treated consecutively in this essay, although in fact they book-ended the TI and Composition phases.) was an early job for me in VIE management (Steve Sherman was later to take on this role with distinction). Essentially this involved assigning each text emerging from digitisation to a pair of volunteers to proof-read prior to the text being passed into TI. Pre-proofing was useful in catching errors which had slipped through the digitisation effort, but this was not its primary benefit: pre-proofing was also a tool for talent-spotting. We had been clear from the outset that, to maintain its vitality, the VIE would need to maintain a steady influx of committed volunteers. Inevitably, some of those involved at the beginning would fall away, and constant replenishment would be necessary. Pre-proofing also generated some useful side-effects: it allowed us to include many more volunteers in direct work, so that they became familiar with work procedures, team structures, file movement protocols and errata notation jargon.
The talent-spotting quickly bore fruit. Those with a good eye for detail, a capacity for hard work, and an inquisitive cast of mind found themselves earmarked as potential managers, candidates to undertake the delicate TI work, or star proofers for further down the line. Many senior VIE managers came to the fore in this way.
Four undoubted stars emerged: Steve Sherman, later to fill a variety of management roles; Chris Corley, later head of the whole proof-reading effort; Patrick Dusoulier, a redoubtable TI worker and head of the ‘Textport’ wing of the VIE, which prepares the texts for use by subsequent publishers; and Dave Kennedy, whose attention to detail was matched only by his capacity for hard work. These four joined me in a ‘Proofing Mentors’ group to share lessons with less experienced proofers. Other hardworking and capable proof-readers, such as Rob Friefeld, Dave Worden, Ron Chernich, Chuck King, Suan Yong, Linnéa Anglemark and Rob Gerrand also found their way into TI.
The most labour-intensive aspect of the project was post-proofing. We decided that each composed text should be read by no less than 6 and preferably by 10 volunteers. With around 150 texts, this equated to a minimum of 900 jobs. Chris Corley accepted the responsibility of organizing this vast work and created a series of post-proofing teams, each with its own manager. Each manager coined a suitable name for their team:
The teams varied in size, with an upper figure of around twenty, although not all members of a given team would participate in every proof-read. Members of the post-proofing teams—a group which made up the vast majority of VIE volunteers—were assigned a text along with their fellow team members. Their comments would be fed back to the team manager, who would then screen, collate and pass them on to Chris. He would then prepare a summary report of outstanding issues, which would be resolved by Composition or TI as appropriate.
Alongside the more conventional proofing disciplines, a series of tools developed specifically for the VIE was also deployed, known collectively as ‘techno-proofing’. This wing of the project was managed by Ron Chernich—who was simultaneously working on a painstaking textual restoration of The Dragon Masters. Two main tools were developed: the Vocabulary/Dictionary Analysis Engine (VDAE) made by Koen Vyverman, and WordPick, designed by Ian Davies. These helped identify areas of suspicion in the electronic texts which could then be subject to further scrutiny. The VDAE was a flexible tool which made it easy to check whether a particular word occurred in other texts, or whether it was inconsistently rendered in the same text. It was also invaluable in identifying invisible characters which could adversely affect the printed text. The use of such tailor-made tools, added to more pedestrian techniques such as MS Word’s own custom dictionaries, uncovered a number of scanning errors and editorial problems which had repeatedly eluded human proofers.
THE WORK PROCESSES – TEXTUAL INTEGRITY
Alun Hughes’ fascinating essay ‘Strange Animals in Questionable Poses’ sets out the results of the TI activity. The mechanism by which the TI work was carried out evolved as the scope of the task became apparent. The day-to-day TI administration was carried out initially by me, and latterly by Steve Sherman.
Each text was the responsibility of a ‘wallah’ whose job was to work through the evidence (typescript if we were lucky, previous published editions if not) and evolve a series of ‘propositions’.(‘wallah’: ‘One employed in a particular occupation or activity: a kitchen wallah; rickshaw wallahs.’ (American Heritage Dictionary).
A proposition represented a request by the wallah to change the digitised text (the ‘vtext’) to another reading, based on the evidence they had uncovered. Each proposition would be accompanied by a supporting argument to allow a sound decision to be made. The wallah was expected to work with a ‘second’, a fellow TI wallah who would act as a sounding board, devil’s advocate and general support.
Once the wallah and second had reached agreement—or agreed to disagree—on which propositions were to be put forward, the text would be sent on to the ‘Board Reviewer’ who had the final responsibility for ruling on the propositions, with support from the second. Consistency of reviewing was important, so the pool of designated reviewers remained small: Paul Rhoads, Alun Hughes, Steve Sherman, Rob Friefeld and me. It was very important that reviewers should also be practising TI wallahs to allow extensive familiarity with precedents and classes of evidence. Norma remained available for consultation on particularly intractable problems, something for which we were grateful on many occasions.
The TI pool was never large, and special mention should go to the most productive members of the group: Steve Sherman, Rob Friefeld, Chuck King and Patrick Dusoulier. I was lucky enough to carry out the TI for ‘Wyst’, ‘Marune’, ‘Mazirian the Magician’ and ‘Cugel the Clever’. For two of those I was working with copies of Jack’s own typescripts—a fascinating and rewarding experience.
THE WORK PROCESSES – COMPOSITION
The output from TI for each text was a file reflecting the changes approved by Board Review. The next stage was implementation of those changes, or ‘Imping’. This task was carried out under the aegis of Damien Jones. Each text was imped twice in parallel to minimise the chances of error. At this point, for the first time, an electronic version of the corrected text existed. A security check would be carried out—usually by Paul Rhoads—to ensure that the text had not become corrupted between initial digitisation and final corrected text; and then the file would be ready for formal composition. The security check involved a comparison of first and last versions of the digitised text to ensure, for instance, that blocks of text had not been ‘lost in transit’—which on occasion they had.
The MS Word files were then imported into specialist composition software to allow professional-standard production. At the very beginning of digitisation John Foley had foreseen that correction and composition must be strictly separated because of the potential for compositional issues to become as controversial as textual ones. To allow the composition team to define VIE compositional standards in tranquillity, compositional questions were ruled ‘out of bounds’ for all other teams. The remit of the teams towards the beginning of the work process was to deliver texts to composition in line with simple and consistent guidelines. This helped these teams remain focused on textual issues, or the specifics of their task, so that project work did not become bogged down in cross-currents. As head of the Composition team John Foley chose to work with an exceptionally small body—a choice thought to be extremely risky at the time. He appointed John Schwab as assistant team head (responsible, among other things, for public tracking) and two other composers only: Joel Anderson, a professional in this domain, as ‘Master Composer’, and Andreas Irle. Paul Rhoads served as a back-up composer but played a greater role as ‘Master Designer’. Though the size of the composition team made work burdens heavier on individuals, it allowed this team to establish and apply more easily the distinctive ‘VIE look’.
As an example of the sort of evolution which other teams also had occasionally to undergo, the Composition team initially worked with electronic compositional tools PageMaker and Quark but later changed to InDesign, a tool which became available well after the project had begun.
John Foley also developed a second key document, based on a study of Vance’s published work, entitled ‘Distinctive Treatment of Words’ or ‘DTW’, treating such matters as use of italics. Though the DTW guidelines evolved as manuscript and other evidence flowed into the project, they were used not only by the Composition team but all the various teams as a rallying point for many surprising questions that arose.
The question of typography turned out to be unexpectedly controversial. After working for several months with Andreas and John Schwab to explore available fonts, Paul had felt that no available typeface was suitable to the presentation of the prose of Jack Vance. The solution was Amiante, an interconnected set of typefaces, designed under the aegis of John Foley, by Paul, with help from Joel Anderson and the other composers. Despite an initial outcry, John Foley patiently nursed Amiante to its final form, negotiating its acceptance, and eventual popularity.
The frontispieces, reproductions of original etchings, are also by Paul Rhoads, as well as many of the graphic elements throughout the texts, which were often solicited and specified by the composers. The ‘knick-knacks’ (text separators), for example, were imagined by Joel Anderson, who also created many of them. The maps which adorn some of the volumes are based on original sketches by Jack, and in some cases (for instance Durdane) have never before been published. While Paul drew all the maps, many volunteers contributed to their conception: for example, the ‘stippled ocean effect’ suggested by Rob Friefeld is used for several, while Suan Yong helped locate the place names on Caraz, the original map being only a very rough sketch. Joel Anderson made the ‘Stark’ diagrams (for volume 44) working from Vance’s original drawings.
Of all the VIE disciplines, Composition was perhaps the one requiring the highest professional skill, and with the greatest chance of catastrophic error.
Once the books were composed, they were subject to further scrutiny by the Composition Review Team (CRT), a discipline whose functions are implicit in the title. CRT was initially managed by Robin Rouch, and frequently exceeded its brief, spotting not only compositional defects but also proofing errors and ‘TI issues’. (Robin created the teams associated with composition review on her own initiative in response to observations she had made as a proofing team manager. Such independence of thought and action lent strength and vigour to this aspect of the project.) We have CRT to thank for noticing that the last chapter of ‘Marune’ had been lost during composition! When Robin stood down after a period of exceptional service, her place at the head of CRT was taken by Marcel van Genderen who maintained her high standards. Once texts had been approved by CRT they were released into post-proofing.
A similar process, known as Correction Validation (‘Correction Validation Team’, the acronym was also used as a label for the process) was carried out after post-proofing to ensure that all proofers’ findings had been correctly incorporated. This team was also originally created by Robin Rouch. The subsequent manager of this team, Bob Luckin, was one of the managers who joined the VIE part-way through its life. One of the strengths of the project was this constant replenishment of new volunteers and managers. Even at the time of writing this essay, six months or so from the end of the project, there remains a steady stream of new volunteers. This refreshment has been essential in maintaining the project’s momentum, replacing lost volunteers and supplying new motivation and energy.
THE WORK FESTIVALS
OAKLAND WORKFEST (January 2000)
It rapidly became apparent that while work was proceeding in the right direction—albeit in fits and starts—using the internet and email, there would be advantages to a physical gathering of project managers. As a result, January 2000 saw the ‘Oakland Workfest’, the first of several VIE meetings.
The Workfest was held at the home of Jack Vance in the hills of Oakland, California. For all volunteers, the chance to meet Jack and his wife Norma, and their son John and his wife Tammy, was an experience to cherish. We all took away many wonderful memories of the Vances’ expansive hospitality and pithy nuggets of Jack’s table talk.
Meeting Jack was not the primary purpose of the Workfest, although the benefit to volunteers’ morale was a spin-off we felt no qualms about exploiting to the full. The real purpose of the meeting, which lasted a week, was to thrash out subsequent work procedures and to inventory the extensive collection of manuscript evidence held by the Vances. This proved especially helpful in planning how the TI work would proceed. It was also an invaluable opportunity to get to know in person those with whom we had had so much email contact over the past few months. Many of those who were subsequently to prove VIE management stalwarts were present, including Bob Lacovara, John Schwab, Alun Hughes, and Suan Yong.
TI CONFERENCES, CHINON AND OAKLAND (November 2000 And January 2001)
It had been envisaged from the outset of the VIE that face-to-face meetings would be infrequent at best. The cost of getting together groups of volunteers from around the world was a powerful discouragement. There were some areas of the project, however, where there was no alternative, and the first of these was the Textual Integrity process. TI was an area where consistency of approach was essential, and TI wallahs would need to master a large body of recondite lore. Alun Hughes spent much of the summer of 2000 in putting together the TI guidelines, based on his own professional and academic expertise, and the practical experience we had gained from the Oakland Workfest.
The standards for being accepted as a TI worker were high, and membership of the TI group was limited to those who had shown both enthusiasm and competence as a proof-reader, added to a curiosity about textual matters and a willingness to commit to producing work within deadlines (not at this stage a feature of any other aspect of the project). Despite the stringent qualifying requirements—which also included submitting to a test TI read-through of ‘Emphyrio’—there was still a large cadre ready and willing to work. For reasons of administrative convenience, we decided to hold two TI Conferences, one for the European wallahs, another for the American and Australian group. Alun and I would run both seminars to ensure a consistency of approach.
The European meeting was held at Paul Rhoads’ home in France, Château de St. Louand. St. Louand had the advantages of a beautiful location in the Loire Valley and ample space to accommodate travellers and economise on hotel bills! The conference explored Jack’s evolving work practices over the years and their implications for TI work, and looked at case studies from the early TI assignments: ‘Wyst’, where I had been the wallah, and ‘Madouc’, lovingly prepared by Steve Sherman. Two months later, the conference was repeated in Oakland.
Both meetings were invaluable in establishing a solid and practical work programme, and for further strengthening work relationships. At this stage of the project, many of us had still not met face to face. The strong relationships built at these conferences were essential in sustaining the VIE over the years ahead.
A series of meetings was also held to sign off the composed texts before they were released to the printers, Sfera International of Milan (who became, during the project, ‘Areagroup Media’, and then ‘Zones’), to be turned into the books you see on your shelves today. These were the so-called ‘Golden Master’ (GM) meetings, stipulated in the Master Plan as a final chance to bring in late errata, to cast a fresh eye on the work accomplished, and to finalise and sign off each volume. Initially it had been planned to publish the VIE as a 60-volume set. For various reasons this was modified to a 44-volume set (for example, the Durdane and Tschai series were gathered into single volumes). It was further planned to ready the volumes for printing in a single batch. This approach proved both unwieldy and impolitic: to do everything at once would have led to unacceptable delays in publication (which would have been unfair on subscribers who had already paid sizeable deposits), and we would have had to cope with a daunting quantity of material in the final stages. As a result we decided to divide the edition into two ‘waves’, and publish and deliver the first in 2002. The first two GM meetings dealt with ‘Wave 1’—the 22 volumes printed and shipped first. As for ‘Wave 2’, it ended up being divided first into two ‘batches’, with a Golden Master meeting for each batch. Batch 2 was eventually divided into two sub-batches because, for instance, of the special problems posed by volume 44.
The first meeting—GM1, logically enough—was held at the home of John Foley in New Jersey, USA. The role of GM1 was to correct outstanding errata, and bring together the texts with the front-matter for each volume. Any remaining compositional questions were also addressed. In addition to John, the meeting was attended by Paul Rhoads, Joel Anderson (VIE Master Composer), Marcel van Genderen (Composition Review Team) and Bob Lacovara, who designed the meeting’s networking system and the archiving protocols it gave rise to.
The output from this meeting was 22 volumes (not the first 22, but the 22 most convientient) which were all but ready for printing. They were sent to Sfera who produced a series of ‘blues’, proof copies which needed only final approval before being printed.
This final approval was given only after GM2, held again at Château de St. Louand. TI and proofing team managers spent a week reviewing the ‘blues’ to ensure that no typos remained, and that no new errors had been introduced since composition. As a result around 100 infelicities—mostly compositional—were spotted, and final corrections made. Even in these last adjustments, the core principles of auditability and double-checking were maintained.
A 22-volume ‘wave’ proved to be at the limit of practicability, and in response we decided to deal with Wave 2 as two more manageably-sized batches. Thus GM3.1 and GM3.2 were organized and held at St. Louand in the summer and autumn of 2004. For reasons of administrative convenience, the work done at GM4 was largely conducted in a ‘virtual’ manner during late 2004, with blues shipped from Milan to individual proofers, and the errata collated by sub-team heads.
March 2003 saw the final stage of Wave 1, packing the books ready for final dispatch. A group of eager volunteers, under the hands-on leadership of Patrick Dusoulier, boxed the many thousands of books according to a rigorous schedule. A similar event was necessary for Wave 2 in early 2005, and also for the Second printing.
(Packers of Wave 1 were: John Edwards, Brian Gharst, Henri Gooren, Evert-Jan de Groot, Andreas Irle, Luke McMath, Paul Rhoads, Thomas Rydbeck, Billy and Gale Webb, and Patrick Dusoulier who organized the work.
Wave 2 packers were: Nicola de Angeli, Phia Bouwmeester, Wilma Bouwmeester, Vlad Degen, Josh Freeman, Rob Friefeld, Brian Gharst, Evert Jan de Groot, Andreas Irle, Jurriaan Kalkman, Menno van der Leden, Bob Luckin, Dustin Maeno, Misi Mladoniczky, Errico Rescigno, Paul Rhoads, Thomas Rydbeck, Josh Snyder, Billy & Gale Webb. The Second Printing Packers were: Nicola de Angeli, Andreas Irle, Christa J., Jurriaan Kalkman, Paul Rhoads, Thomas Rydbeck, Vince Serano, Billy Webb.
USA-transhipment was in all cases handled by Richard Factor, John A. D. Foley, and R.C. Lacovara.)
THE INVISIBLE MANAGERS
THE VIE BOARD
To collect money and sell books, the VIE had to become a legal entity. This problem was solved for us by Ed Winskill, a lawyer from Tacoma, Washington, and his colleague, Bob Nelson, based in California. They set up a VIE not-for-profit corporation, and Ed, as well as Mike Berro, Norma and John Vance, Bob Lacovara and Paul Rhoads, served on the corporate board.
In the early months the project required a degree of working capital; this was provided by Mike Berro, at a time when it was by no means sure that the project would succeed. It was a high-risk investment, and Mike deserves credit for his vision and generosity.
Many of the hardest-working VIE managers operated largely behind the scenes. Their work was not as readily defined as more obvious disciplines such as post-proofing and textual integrity, but was essential to the successful delivery of your VIE volumes.
John Schwab was the VIE’s archivist. He maintained the primary archive of VIE texts (although Suan Yong, Mike Berro and Koen Vyverman maintained backup archives). No text could be moved between volunteers without John’s approval, and he also devised a consistent file-naming system, without which keeping track of the movement of thousands of files all around the world would have proved impossible. His occasional irascibility at slack application of approved procedures could well be understood! John’s labours as archivist were in addition to his other VIE work, including launching digitisation, several weighty TI jobs, and his work as a composer. If all this were not enough, John set up the ‘master’ assignment tracking and composition tracking pages to ensure that readily accessible information on the status of all texts and volunteers was available. Few, if any, VIE volunteers worked harder than John Schwab.
Suan Yong was another manager who carried out a tremendous amount of work behind the scenes. Suan set up and maintained the VIE internal databases, including subscriber records, and took over responsibility for the Master Tracking Chart.
The errata database, to which readers could send errors found in their VIE volumes, was also created and maintained by Suan. No job was too unglamorous for Suan, who did much work at the business end of double-digitisation. He was the project’s first ‘gate-keeper’, responsible for responding to initial contacts from potential volunteers and subscribers. His TI work on the Durdane trilogy was also exemplary: he deduced the existence of a map by Jack which was later discovered at the Mugar Library. In debates among VIE managers, Suan was never afraid to argue the less popular position, and his contributions were often invaluable in arriving at consensus.
‘The legendary locator’, Hans van der Veeke, volunteered at a later stage, but found that he was offered no work. Eventually, he sought out the Editor-in-Chief to indicate that resources were being neglected. As a result he quickly entered management as ‘Volunteer Co-ordinator’. It now became his responsibility that no other volunteers ‘fell between the cracks’. Taking his work to heart, Hans also came to maintain the volunteer pages on the website, as well as making sure that job credits were being properly allocated. In particular he gathered and collated all credit data, published it in ‘Cosmopolis’ for volunteer verification, as well as organizing the volume 44 credits list.
As the VIE progressed, its work processes became ever more complicated. By late 2001 we were in need of an extra manager just to manage the processes! Joel Riedesel stepped forward to become the VIE’s ‘Work Tsar’. His job was to ensure smooth integration of the mushrooming separate phases, and that no aspect of work for any of over 150 texts was neglected, including their assimilation into the 44-volume structure, as well as maps, tables of contents, credit lists and so on. Joel’s work proved spectacularly successful and popular, so much so that he was not allowed to leave the post. His status reports, landing in managers’ inboxes every Sunday, simultaneously reassured the reader that the VIE continued to move forward, while drawing attention to one’s own delinquencies which might be causing bottlenecks elsewhere. Joel was also an energetic proofer on the Clam Muffins team; although since the Clams were managed by his wife Robin Rouch it may be that he had little choice. Joel also took on the job of ‘Volume Post-Proofing’ (VPP), a process he vigorously advocated, using post-proofing teams to review individual volumes once the texts making them up were brought together. For instance, volume 3, ‘Gadget Stories’, comprises 21 separate texts, plus ‘front matter’, all of which were proofed independently. VPP was a chance to review all these texts, together in their volume, prior to Golden Master, to ensure such things as continuity of page numbering.
The complexity of Joel’s job is suggested by the diagram of VIE process interaction: proof that a picture is worth, in fact, several thousand words.
As the founding editor of the newsletter ‘Cosmopolis’ alone, Bob Lacovara’s achievements would deserve our admiration. But Bob was also the project’s financial planner, a difficult assignment that involved pricing the VIE sets at the minimum price consistent with viability, dealing with exchange rate fluctuations over a period of years, and ensuring that meaningful financial information was available. No less important was his role as unofficial ‘deputy Editor-in-Chief’, in which capacity he proved an invaluable moral and practical support to Paul Rhoads. Bob also earned the thanks of all VIE managers by conceiving that notes and queries to the electronic files should be entered as end-notes in that file, rather than in accompanying ‘bis’ files. This foresighted decision saved immense time and confusion.
This ‘Jack-of-all-trades’ section of the essay is also the most appropriate place for me to shoe-horn my own varied contributions to the VIE; if only to demonstrate my credentials to be writing this essay. In many ways my involvement has mirrored that of many other VIE managers (Steve Sherman and Rob Friefeld spring to mind) in taking on one small job early on, and soon becoming a gun-for-hire in all areas of the project. False modesty is a deeply unappealing characteristic, and while I may not have worked as hard as some, the range of jobs I’ve undertaken qualifies me to express my opinion with a certain degree of authority. My first job was to digitise
‘Marune’, a text I also subsequently restored in TI. I was also the first leader of pre-proofing (in which capacity I set up the ‘proofing mentors’ group), project gatekeeper, TI wallah, manager and reviewer, and co-ordinator of all the Golden Master meetings. For me, as for all VIE managers, the project has been essentially fun rather than drudge; but it’s fun we’ve all worked hard for.
The VIE from the outset was conceived as being about more than creating a few hundred durable hardback book-sets. We also hoped that future publishers of Jack’s work would use the restored texts rather than earlier editorial mangling.
Patrick Dusoulier took on an important role in this context by setting up ‘Textport’, a team for preparing electronic texts for use by e-book publishers and commercial reprints. Textport involves retrofitting VIE text files with those corrections brought to the composed files, downstream of the main correction work, via post-proofing and the Golden Master processes. As of this writing many new commercial publications have come out using VIE-restored texts. Regrettably their standards have not been high, so that in some cases new errors have been introduced. That said, Vance re-publication in English, but especially in French translation, has already been stimulated by the VIE. French translations are being redone, or corrected, from VIE texts, usually in close collaboration with Patrick, who is working also with Spanish and German publishers (in the latter case the VIE’s own Andreas Irle, who has also translated Lurulu.) Patrick’s work on Textport came after earlier VIE management as a pre-proofing mentor, and he spent much time as a diligent TI wallah, whose involvement was essential in the evolution and testing of the TI guidelines. He was ably assisted by Suan Yong and Chuck King. Chuck was active in many other areas of the VIE, including prodigious labours on double-digitisation and TI, and leading on ‘RTF-diff’, an arcane procedure designed to ensure that material was not lost as texts crossed electronic platforms.
The logistics of the VIE were complex from the outset. Hundreds of volunteers, literally from every continent, made for a challenging environment. In a project of so many defined phases and activities, there was a distinction between specific communication within groups—for instance the TI team—and more general communications affecting the whole project. This latter area was very important in building a ‘VIE spirit’ and ensuring that volunteers felt part of the whole project.
The official route for such general communication was ‘Cosmopolis’, a monthly electronic newsletter devised by Bob Lacovara, who in January 2000 became its first editor. ‘Cosmopolis’ was a mixture of news about the project from the various teams, and articles of interest to Vance readers. ‘Cosmopolis’ was an open publication with contributions solicited from volunteers, and indeed anyone with an interest in the work of Jack Vance. Having established the publication, Bob Lacovara stepped down from the editor’s chair in 2001, to become ‘Editor Emeritus’ (although with none of the negative connotations associated with the latter term in ‘The Languages of Pao’). Subsequently, during periods of editorial hiatus, Bob edited several more issues; Paul Rhoads and Debbie Cohen also served as editor for brief periods. But Derek Benson in particular, and later Dave Reitsema (also head of the Tanchinaros proofing team), did the bulk of this work.
Cosmopolis spawned in 2001 a sister publication, the Cosmopolis Literary Supplement, or CLS. The CLS was a showcase for volunteers’ own writing influenced, however indirectly, by Jack’s work. CLS was initially edited by Joel Anderson and Paul Rhoads, and latterly Till Noever. Work was presented from a number of volunteers, and received on occasion with real enthusiasm. I myself enjoyed the thrill of serialising two novels in the CLS, which certainly helped me feel an even stronger bond with the project.
Although the initial publicity for the VIE was carried out through Mike Berro’s website, the project also launched its own site to provide a central source of information about the project and access to tracking data. The final version of the website was a collaboration between Max Ventura (who designed the site), Paul Rhoads (illustration), Suan Yong (who integrated the tracking charts) and Rob Friefeld (hosting). Rob was another of the VIE’s versatile and hardworking managers. Initially a star proofer at pre-proofing, he became a most prolific TI wallah and ultimately a reviewer as well. He was among the most prolific of techno-proofers and a stalwart of composition review and verification. If this were not enough, Rob was also head of the Penwipers post-proofing team.
One of the aims of the VIE has been to promote Jack’s work to a wider audience: the 44-volume set is of course a contribution to this, but in addition two stand-alone volumes were produced early in the project. The Gift Edition (also called ‘the Gift Volume’): ‘Coup de Grace and Other Stories’ (2001) showcased a number of Jack’s shorter works, and was intended as a primer for the reader unfamiliar with his works. The volume brought seven much-loved stories together, including ‘The Moon Moth’, ‘The Murthe’ and ‘Dodkin’s Job’.
The Science Fiction Volume, produced in 2002, presented ‘The Languages of Pao’ and ‘The Dragon Masters’. Both texts were published in significantly revised forms, drawing on the work of the Textual Integrity team to produce for the first time texts as close as possible to Jack’s original intent; both had previously been subject to damaging editorial degradation. The Science Fiction Volume was therefore an important achievement for the VIE, demonstrating its ability to carry out credible textual restoration based on sound evidential principles. It also had the aim of raising the VIE’s profile in the science-fiction community, and unearthing more volunteers.
FRANKFURT BOOK FAIR
In 2000, the VIE took a place on Sfera’s stall at the prestigious Frankfurt Book Fair. The accompanying brochure (translated into six languages) set out the aims of the VIE and the principles on which it operated, and generated a good deal of favourable publicity. (This built on the excellent work of John Robinson, who worked extremely hard to publicise the VIE in the science-fiction community, helping immeasurably in the recruitment of volunteers.) Jack’s work has always been more popular in Europe than the US, with particularly strong followings in the UK, the Netherlands, France and Scandinavia, and the Frankfurt Book Fair was one way of capitalising on this.
Communications were essential to the realisation of the VIE. The project began in 1999, and even three years earlier it is unlikely the project would have been logistically possible. Day-to-day communication was via the internet, with files being passed between volunteers electronically. Volunteers therefore needed to be able to use the internet and email to be able to participate; and it was only in the latter years of the 1990s that mass email use became standard. The sheer volume of file movement and tracking required by the VIE’s work processes meant that any paper-based system would have been impossibly unwieldy. The possibilities of the internet, and the very open way in which the VIE used it, left the project exposed to destabilisation by anyone who approached it with a malign or destructive purpose. Nonetheless, without the ease of communication afforded by the Web, and the ability constantly to refresh the volunteer base, the VIE would not have been possible at all.
The VIE has been a long time in the making, and I am conscious that this essay too is not characterised by terseness or brevity. I can only hope you have showed more patience than the truculent Hurtiancz:
I began with a salute to the VIE volunteers. Their importance cannot be overstated; but the VIE is, more than anything, a tribute to the delight of reading Jack Vance. It is this delight which has drawn such unstinting efforts, from around the globe, over several years.
Our labours on the VIE are ended; but our achievement is enduring.
Bosham, England, December 2004