ICOM started using the Internet as a means for self-service access to its core documents at a time when it was by no means clear that the Web would attain its current prominence. It was clearly stated at the outset that the "ICOM Document Repository" was intended, "to provide material to as large a segment of the global community as possible." The initial structure of the repository was therefore such that any document in it could be located and retrieved via e-mail, FTP, Gopher or the Web. Surprisingly, the e-mail retrieval service was almost never used. Although it remains in operation, it is not maintained actively and only the older documents can be accessed by it. Similarly, with the Web's ascendancy the Gopher service fell into disuse and was subsequently terminated. The FTP engine, however, is still in frequent use and may be used for the retrieval of any document in the repository which, for whatever reason, cannot conveniently be obtained directly via the Web. The entire repository is available from mirror sites on several continents: two in Europe, two in North America, one in Australia. Additional mirrors in Africa and Asia will hopefully be in operation by the time of the October conference. The parent site has a massively wide and reliable connection to the Net, as do most of the mirrors. Any suggestions about further measures for overcoming bandwidth constraints will be gratefully received. Until a few days ago, the repositories consisted entirely of documents of relatively small size. However long tranfer times might have been, nobody found it worth the effort to complain about them. This situation has now changed. The Handbook is a prodigious mass of interlinked files and its successful viewing (regardless of the mode of transfer) is utterly dependent on software that supports HTML frames. As far as I am personally concerned, this latter device causes much more trouble than it is worth. Any successful document using it must provide transparent means for viewing in an environment that does not support frames. In this sense, the present version Handbook is clearly inadequate. In fact, it is beset by a number of additional technical shortcomings. The alternatives at this point would have been to keep the thing off the Net, entirely, while the producers modified it to provide support for a broader range of viewing environments, or as was done, to make the document available to anyone who is capable of utilizing it in its present state. The paradox here -- the real bone of contention -- is that the document's explicitly labeled target group is among the more likely to have technical difficulty in utilizing its contents. Despite this, a hardcopy version of the Handbook is at public disposal and anticipated access constraints to the current electronic version therefore provided no justification for abstaining from its release. Frames are a legal feature of HTML and the Web contains countless documents that use them with varying degrees of success. Indeed, it is hard to imagine a Web version of as intricate a document as the Handbook without them. Nonetheless, its designers would be well advised to incorporate a frames free text and graphics alternative in a future version. Rest assured that this advice has already been conveyed and that a more flexible Handbook Mark II is very likely to make its appearance. If this should prove too difficult, at the very least, a PDF version of the hardcopy version will be made available. In the interim, and with the admission that my not having done so at the outset was an outright blunder, I will be adding the usual "Frames capability needed" label to the links to the Handbook. Cary Karp mailto:[log in to unmask]