VI. What improvements could a funded service offer?

A. New features

There are a number of things about KIDSPHERE which are awkward -- either for the readers or for the people who try to operate the service. All of these awkward features could easily be remedied if the service were funded on a regular basis. In the previous section we made the case that KIDSPHERE has value and merits funding. Let us ask now how that money could best be spent, that is, which features of KIDSPHERE should be changed once stable funding becomes available.

First of all, one should make the service easier to manage and easier to use. On the management side it would be desirable to automate as many administrative functions as possible -- subscriptions, cancellations, address changes and bounced mail. Of these tasks, the first three can be handled by standard packages such as the majordomo mailing list manager. There would need to be a supported machine to handle these services, but the cost of such a device is not high. Bounced mail is more difficult to deal with, since by definition there has been a breakdown in some component of the mail system when mail bounces from a remote site. A portion of the bounced mail can be filtered, and invalid addresses can be stripped from the mailing list if problems at the remote site are not fixed within some reasonable time period. One needs to be careful not to cut off subscriptions from sites which are not maintained over the weekend or over holiday periods. This is typical of most school sites, and one must simply tolerate problems at these sites during such periods without penalizing subscribers whose accounts are temporarily inaccessible for this reason.

The automation of KIDSPHERE's administrative functions would have another benefit -- that of allowing the service to accommodate many more subscribers in the future. At present the mailing list's volume is limited by the time it takes to handle administrative chores and to process the actual messages that flow through the list. A reduction in administrative overhead would allow time to optimize message-processing and allow for more messages to be sent out daily.

B. New distribution mechanisms

Should this increase in traffic occur, it will only exacerbate a problem that many readers have already identified -- that of dealing with the high volume of message traffic that already exists on the mailing list. Fortunately there are a number of remedies for this problem. All involve somewhat more attention from the service provider's end, but this effort is almost certainly cost-effective from the subscribers' viewpoint. Here are a few options that could be implemented without great difficulty:

News distribution. This would involve the operation of an NNTP (Net News Transport Protocol) server to distribute KIDSPHERE along with the hundreds of other newsgroups that many sites already receive. Individual subscriptions would not be necessary at these sites, and people could read KIDSPHERE articles using local news clients. This makes things simpler from the readers' viewpoint and greatly reduces overhead at the distribution end of the service.

Headline service. This would be a service which would distribute the message headers (name of submitter, date of submission and subject) for all messages received in a given day at KIDSPHERE. Subscribers to the headline service could request the full text of any article which appeared to be of interest by sending an electronic mail message back to the headline service distribution point.

Gopher and Web distribution. A gopher distribution of KIDSPHERE already exists through ASK-ERIC. This distribution could be made directly from the originating site for KIDSPHERE. Distribution via World Wide Web is another attractive option. Both these mechanisms are efficient and flexible.

Digest service. This is an option available through many mailing list management packages. It combines all messages for a given day into one long message, with a message index at the front of the single message. This does not reduce the volume of mail that any subscriber receives, but it does place it in a single message so that KIDSPHERE mail does not obscure a subscriber's personal mail.

Custom mail filters. These filters would enable subscribers to request that only those articles which contain certain keywords will be sent to them. While this sounds attractive, there are some disadvantages to the idea. For one thing it requires people submitting messages to the list to make consistent use of these keywords if their articles are to be selected by the filters. And segregation of message traffic tends to destroy the idea of community-building with which KIDSPHERE has been so successful.

Sublists. The idea here is to split the large KIDSPHERE list into a number of smaller mailing lists -- one for various topic areas or various grade levels, for example. This would, however, split the online community, and many readers has expressed a reluctance to see this happen. Of course any reader could subscribe to several or all of the sublists. The problem that arises in this situation is that people would begin to post their messages to several of the sublists, and the total message volume would actually increase, with no increase in content.

Searchable archives. This is an extension of the ideas of a headline service or of custom mail filters. People could search through the archive for articles on particular topics.

Multiple moderators. This is another way of splitting up some of the activities associated with KIDSPHERE, but it would occur at the provider's end rather than at the readers' end. Submitted articles would be sent out to one of several moderators, the choice either being a random one to reduce the load on any given moderator or a selection by topic area to provide input from subject experts for the moderation procedure.

Edited summaries. A "Best of KIDSPHERE" summary could be prepared on a weekly basis. This summary could be distributed online -- or it could be prepared for print distribution to help teachers and administrators at schools which are not yet online see what sorts of resources are becoming available to educators on the Internet.

The majority of these new mechanisms would be desirable to implement. Since each entails new administrative overhead or new demands on system hardware, it would be impractical to implement them without corresponding new funding. Once such funding becomes available, specific mechanisms could be provided on a user-selectable basis. That is, those subscribers who wanted to receive a headline service or a news distribution could do so, while others might elect to retain the conventional distribution through electronic mail.

C. Funding mechanisms

The total cost for new KIDSPHERE options would be quite modest to implement. There is a need for hardware, perhaps at the level of $10,000-$20,000. Half-time technical support would suffice to set up the various services described above, and support for one or more moderators working for a total of 40 hours a week could deliver these services to the readership. The total cost is thus in the range of $100,000 -- pretty much the same as that of a medium-sized face-to-face conference.

One can imagine several avenues for developing this level of support. There has been some discussion of these options on KIDSPHERE, and a few clear preferences have emerged.

Subscription fees. An annual fee of $50 per addressee could cover the costs of proposed new services, assuming that the number of subscribers would not fall if such a fee were instituted. This assumption is likely to be false, and the arithmetic ignores the cost of administering a fee service. Furthermore, fees of any size will drive away many new users of the network, and this is precisely the group which has benefited the most from KIDSPHERE's services in the past. Hence this option would not appear to be a viable one for this service.

Federal, state or private grants. This is an obvious mechanism to employ. It is one that is commonly-used to support traditional community-building activities such as conferences and workshops. But few funding agencies have recognized the value of online community building, perhaps because few of these agencies use the Internet themselves. It is worth pursuing this option, if only to sensitize the funding community to the value of online interactions. They could be using the network themselves to great advantage, and they might be supporting many more network projects if they understand the full potential of this medium.

Integration into larger projects. This is an option which has been available to KIDSPHERE for some time. It has been resisted out of the feeling that KIDSPHERE has a unique character which can best be maintained if KIDSPHERE's independence is maintained. It would also be desirable to see an explicit recognition of the value of this type of service by the agencies which fund other school networking projects.

Commercial support. If school networking expands so as to reach classrooms across the United States, it will be a multi-billion dollar business in that country alone. Worldwide the potential is perhaps ten times as large. Companies seeking to do business in this market have a clear interest in reaching the people who are designing, implementing and using the technology, and KIDSPHERE is an excellent way to do this. Commercial sponsorship of KIDSPHERE and associated printed summaries could provide a stable basis for these activities and a way of bringing the sponsors' names before the KIDSPHERE readership without interfering with the mailing list's content in any way. The greatest obstacle to implementing this option is probably the administrative time required to set up the necessary commercial relationships.

Among these options some combination which incorporated KIDSPHERE into a larger framework and sought commercial sponsorship for that larger activity seems to be the most practical approach at the present time. Under this arrangement subscribers would experience no great change in the service aside from the prospect of a greater variety of distribution and retrieval options. The cost of soliciting funds for KIDSPHERE would be subsumed in the costs of funding the larger activity.

D. Conclusions

We have argued that KIDSPHERE has grown to provide an essential service in support of school networking around the world and, more broadly, in support of educational reform efforts across the curriculum. It is a resource that teachers new to the Internet can use to orient themselves in the online community and a tool that all teachers can use to help develop and make use of new network resources. With modest funding the service will be able to expand and keep pace with the rapid increase in the size of the online community, while retaining the personal flavor that has enabled it to be successful in the past. There are a number of mechanisms available for securing this funding, and it seems likely that the service will continue to thrive as wide-area networking reaches an increasing fraction of the world's teachers and students.