The Internet has been with us for only about a decade. Users of the WWW should be aware that there is still more information, literature, data, etc., existing only in printed form, than is available on the Internet. While the possibilities of information and data retrieval have advanced at a tremendous pace in recent years, there is an infinite number of possible improvements. I shall mention only a few very subjective ones as an example here.
The increasing presence of commercial companies on the Internet is both an enrichment and a plague, the latter because more and more frequently unsolicited emails are being sent to global distribution lists with commercial offers. While this is annoying, and measures should be taken against it (see cdsweb.u-strasbg.fr/~heck/spams.htm), I do not think that it is a reason for astronomers to refrain from being listed in email guides or from making these guides available among colleagues. The damage to easy communication among scientists would be too severe.
The transition from printing large tables on paper to publishing them in electronic form (Section 3) either in the ADC and CDS archives (or on the AAS CD-ROMs), raises the question about the future of marking-up tables for printing. For many years authors have been obliged to convert their data tables to LATEX format. Ironically, AAS requests a charge of US$ 50 for the service to convert the data tables back to plain ASCII format for publication on their CD-ROM, except for tables marked up with the AASTeX macros (see www.journals.uchicago.edu/AAS/cdrom/). It may be anticipated that the ``publication'' of tables in electronic form will eventually release authors from this task. However, special non-ASCII symbols, like e.g. Greek letters, will require to be transliterated to ASCII characters in the electronic version.
Unfortunately the journals in astronomy do not yet oblige authors to provide their tabular data to a data centre, as a requisite for publication. An agreement between all major journals and the data centres ADC and CDS is highly desirable, not only for the sake of the completeness of their electronic archives of tabular and catalogue data, but also to remedy the following problem. The clearing house of the IAU Task Group on Astronomical Designations of IAU Commission 5 (cdsweb.u-strasbg.fr/iau-spec.html) has frequently come across unconventional namings of astronomical objects causing confusion and redundancy of names in object databases like NED and SIMBAD. My experience in the Task Group was that the standard refereeing system of journals does not help to avoid this problem. Ideally, these tables should be run through an automatic cross-checking routine prior to publication or acceptance. For this purpose they should have at least two identifiers (a name and a coordinate) and could then be compared with databases like SIMBAD, NED or LEDA, in order to check the consistency of names and coordinates, and perhaps even part of the data. Of course this is useful only if the objects were known previously.