by Q.-Tuan Luong for the Large Format Page, June 2002. Rev Sept 2003.
I created the LF Forum in 1997, using the "Lusenet" software and hosting provided by Philip Greenspun at greenspun.com. My understanding is that at that time the main benefit for him in providing those services for free was to showcase his expertise in DB-backed web sites. It turns out the interface was designed with excellent taste, and was effective and simple. It was probably no little factor in helping the forum grow. The forum would not have existed without Philip Greenspun, not only for the services, but also the inspiration. It was very close in spirit to the original photo.net, although it could be argued that the original photo.net was instrumental in helping Philip's tremendous web services career, while I didn't expect much in return.
My vision was to provide a place where large format enthusiasts could come and exchange freely ideas and help. No commercial activity would ever take place there. No financial contribution would be required, because I considered that the most valuable contribution would be the time spent by members to contribute contents. No copyrights claims to contents would be made, and the legalities were just to be similar to those on Usenet, after which Lusenet was aptly named, based on good common sense and courtesy.
Far from me is the thought that photo.net is bad. In fact, I have been a member there since 1996. In 2001, I was one of the first dozen of members to write specifically a long feature article for the photo.net site (the photographer's guide to Vietnam) using their format, and I still contribute regularly to the Forums there. I went to great length to buy from the photo.net affiliates, even if that meant giving up on the convenience of my usual suppliers. When I had the impression that one of those transactions didn't register properly, I alerted photo.net so they could check. Although I am an active member of photo.net (or more accurately because) and have no axe to grind against their management, I could see clearly the differences between photo.net and the LF page hosted by Lusenet.
The main difference is that the LF page is a not-for-profit online community, while photo.net is a for-profit company which operate on-line communities. As such it is difficult for photo.net to have the same idealistic goals as the LF page. It if often pointed out that photo.net is in fact not that different, being a non-profitable operation, but I don't buy this argument for two reasons. First, it was not the intention of photo.net not to turn out a profit, since otherwise I suppose it would have been founded as a non-profit. Second, the economy has taught us that when companies do not turn out a profit, bad things happen to them, like closing down or being sold to whoever, and this is always a risk that we must keep in mind.
When someone contributes to the LF page, he can take pride in donating knowledge in the spirit of giving something back to the large format community, the very reason for which I created the site myself in the first place. You can note that the ratio of contributed articles over community members is maybe one order of magnitude more in the LF page than in photo.net. It was to encourage that spirit that I didn't want to even accept donations. When someone contributes to photo.net, he increases the value of a commercial venture. One of the early strategies of Philip Greenspun was to create "magnet content", somewhat controversial, but sketchy articles which would compel the readers to fill-in the gaps by adding comments. Clearly most people wouldn't let a magazine use their words or images for free, so many people would find that there is a limit to their contributions to photo.net.
A second problem is that like all companies, photo.net needs to grow to have credibility and resources. As it grows, it attracts more members and traffic, and has to cover increasing expenses to satisfy sponsors as well as members. By contrast, the LF Page just serves ascii text to a select and relatively small (in 2002, about 1500 people posted a message, with a couple hundreds posting at least weekly) community and doesn't need much to support itself, since being a non-mainstream community, it will always remain of a manageable size. Our server would enjoy a light load, and it wouldn't be difficult to develop more custom community services.
This resulted in somewhat different communities. The LF community was as cozy and supportive as it gets. We've seen members trying to set-up a network to buy and stock film locally so that one would avoid the hassle of going through airport security with film. We've seen a member being offered a free gear loan after he posted about his gear being stolen before a trip. Of course, we've also seen some flame wars, and as a moderator, I've had my share of angry emails, but none of this resulted in the incidents I have witnessed on photo.net such as retaliatory ratings, or members with excellent records of contributions threatening to leave, or worse, being banned and denied the right to have their contributions removed.
Instead, I asked the community members what they thought of the different options. This was not meant to be a vote, since one had been conducted a year and half ago, but rather a discussion meant to put a finger on key issues. This was not a discussion of a Lusenet-like system hosted by photo.net versus a Lusenet-like system hosted independently, since this was the essence of the 2000 discussion, whose result was clear enough. Rather, I wanted to know whether other alternatives were acceptable to the community. The main conclusion which emerged from the discussion was that most people preferred to stay with the current greenspun.com/ACS interface, a testimony to its excellence. I ruled out all the options which did not satisfy this requirement. The discussion also suggested that in the case of a move to photo.net, copyright issues and the individuality of the Forum had to be preserved.
In light of those conclusions, I asked Philip Greenspun the following:
(a) copyright is independently held by authors and photo.net (b) each page has a small area at the top or bottom for which I can decide the contents. (c) no specific 35mm/MF advertising (d) an option for readers of the LF forum to search only the messages of the LF Forum and not the whole photo.netThe idea I had in mind was that while the LF contents would be visible to anyone in photo.net, the rest of photo.net would be visible only to the LF members who wanted it to be so. Philip agreed to all of those points (subsequently, the notice which said that photo.net own the copyright to all the postings was amended) but said that the staff of photo.net has so much on its plate that it would take maybe a year to implement some of them. I replied I thought it might not be the best to move to an overloaded server with overloaded staff which would have no time to devote to the sub-community issues, whereas our requirements were quite simple.
I began to explore the feasibility of an independent Lusenet/ACS server. There were three problems. First, the system, even in its more portable incarnation, known as OpenACS, would need to run on its own server because of its requirements of fringe technologies usually not provided by web hosting companies. This increased the hosting fees by one order of magnitude, and would require raising funds from members. Although I would have preferred to avoid it, this was not a major obstacle. Second, one would have to install OpenACS itself on the system. Third, one would have to transfer the old archives which are too precious to loose. I had a look at the software. It appeared to me that the second task would require a couple of days to a good software developer, while the third one, while apparently the easiest, would require probably a week, due to a number of small differences between the two softwares. I am not even a good software developer myself, so the task appeared over my head. There were a number of offers (most turned out to be unsuitable after a close look) but instead of pursuing them in parallel, I chose to accept one which looked more promising because it would have an OpenACS server already set-up and running. However, this still did not solve the third problem (data transfer), and the person who made the offer clearly couldn't be expected to spend the time required to do it. In light of the correspondence I've just had with Brian, who eventually pulled it off, it seems that my assessment was correct. Anyways, around mid-March it was clear this was not going to work in the short term, and I left for a trip in April and the beginning of May. It looked like despite more frequent outages, the greenspun.com server could still go for a while, as Philip Greenspun had given no specific deadline. I accept the blame for not considering the situation with more urgency and failing to mobilize more effectively the resources of the LF community.
The fact is that I was not even that opposed to a LF Forum move to photo.net. There were plenty of opportunities to move to another system that I could have taken if it was more important for me to prevent the forum from being taken over by photo.net rather than seeing it served with a good infrastructure. I was envisioning a move to photo.net by this summer if no alternative was found.
Then apparently the person who was then in charge of the transfer was not able to pull it off before his own vacation, and the Forum was restored for a few days. The aborted move was a wake-up call for a small group of web developers among the readers of the LF Forum, Bjorn Nilsson, Josh Wand, and Jennifer Waak. They came up with the idea of rewriting Lusenet in php/mysql, two popular technologies which would have made it much more portable and eliminate the hosting problem. This was good solution for our needs, which do not require the complexity of the enterprise-class ACS. At about the same time, I received an offer for a free account with possible root access on a decent server connected to a T1 line. This was from Brian Reid, an extremely knowledgeable person who has an history of being generous in helping on the net non-profit organizations and individuals.
I had resigned myself to the move as the least worse solution for the LF Forum, but now that we had a promising solution, the situation became very different. I asked photo.net to delay the move so that we could finalize the software, evaluate all the possibilities, and get a consensus on what to do. From my point of view, there was no urgency. The greenspun.com server, while unreliable, had been somewhat working, and some insiders acknowledged that it could still be up for a while. In fact, a year later, it is still running. The data itself was not too much at risk. After being notified of the possibility of a greenspun.com shut-down, I had taken the precaution to download it. It is true that from December 2001 to May 2002 we had not made much progress in finding a new home, but now development efforts were underway, which would have made hosting a trivial issue (many $30/month accounts would now be appropriate, since we wouldn't need to the OpenACS and its associated technologies) even if we didn't have our free server. In case of a final decision to go to an independent server, the move would complicate things further by confusing the readers, some of which would be lost during the transitions, confusing the search engines, as well as introducing new technical difficulties with database updating, and identification numbers.
However, the photo.net staffers ignored my request. Apparently Brian Mottershead tried to email me but this email was lost somewhere before reaching my inbox. Brian (a staffer with long-term financial interests in photo.net) at that time appeared to sincerely believe that greenspun.com was to be shut down soon. More importantly, he wanted the material for photo.net.
Even as the move proceeded, and despite the way it was done without my consent, or that the members of the lusenet LF community, I was still willing to cooperate with photo.net enough that I asked the community to go there at least temporarily. My main concern was to avoid a split which would loose members along the way. I also didn't want to make an unnecessary fuss since I appreciated other aspects of photo.net (I remain an active contributor to other forums to this day).
Several key contributors to the LF forum objected to the takeover by photo.net, and as a result have withheld or radically reduced their participation on the LF Forum as it was moved to photo.net.
After Brian had repetedly tried to cast doubt about the legitimacy of moving over the material previously served by Philip Greenspun to an independent server, a conversation with Philip Greenspun made it clear that Philip never considered that he had any licence about the posts, but rather that they belonged to the LF community. Although the greenspun.com server that is running Lusenet is still up more than a year after the move, Philip was worried about it back then. It was only out of concern for the future of the LF Forum that Philip asked for the transfer, and he understood very well my concerns about commercialization, having himself urged that photo.net be transformed into a not-for-profit entity. Philip made the data transfer easier by providing us with a direct database listing.
Meanwhile, although the volunteer efforts had an encouraging start, the initial progress was slowed down by disagreements amongst the volunteers developpers on how to code. In the fall, Brian decided that the forum had taken root on photo.net, and reversed his initial offer to hold a vote concerning the location of the forum, which put a damp on the enthusiasm of the developers.
Eventually Bjorn was left alone with the responsability to finish the software, but he was seriously injured in a car accident. Since he could not access the internet while recovering at the hospital, nobody could contact him, and efforts were stalled for many months. What allowed the software to be completed is that Tom Wesbrook volunteered to finish the development just at the time when Bjorn got back in touch with us, at the begining of the summer 2003. This made it possible for Bjorn to transfer graciously his code to Tom. Within the summer, Tom had finished the software, and transfered it to the largeformatphotography.info server.
Comments about the first version of this article on photo.net
A more recent discussion between Brian and advocates of the independent LF Forum (originally on Bjorn's prototype LF Forum).