report from the 4th World Wide Web conference (4WWW95) in boston

IMPORTANT DISCLAIMER:

this is neither an official document in any respect nor should any statement be taken as a fact, this is just my personal opinion. reto ambühler, informatikdienste of the ETH zürich

first impression:

the fourth World Wide Web conference (4WWW95) took place from december 11th to december 14th 1995 in boston, USA.
the copley place marriott hotel and the adjacent hynes convention center easily hosted the over 2000 attendees and provided every service one could ask for. the conference was very well organized. a total of approximately one hundred workstations, Macs and PCs provided access to the Internet. the equipment functioned well and the transfer speed was excellent.

when we arrived in boston on friday afternoon, the city greeted us with crystal blue sky. but on sunday, it started to snow and it was very cold. by tuesday, the charles river was frozen. but this could not stop us from doing some sightseeing: we visited the world famous Massachusetts Institute of Technology (M.I.T) and the Harvard University.

we, that is:

• louis perrochon from the computer science department of the ETH
• roberto mazzoni from the computer center of the university of zürich, see his trip report
• myself

according to the printed list of attendees, there were 31 people from switzerland.

pictures and impressions from boston at winter - a wonderful sight ...

key messages:

to me, the following are the key messages from this conference:

• there will be a tough battle over the HTTP and HTML standards in the future !
  amongst other vendors, NetScape and Microsoft push ahead very hard with their implementations of the extensions of the current HTML V2.0 standard. there are already two different proposals for TABLES around and the current browsers support HTML in various ways, from strict V2.0 to any extensions suggested.
  i do fear that we may end up with multiple, vendor specific "HTML V3.0" standards and that the World Wide Web Consortium (W3C) may lose control over the Web standards.
  the same is basically true for HTTP, especially as far as extensions for support for enhanced security and privacy are concerned.
  
  
• the Web goes commercial !
  while at the first three Web conferences many talks were about how difficult it is to find the information you need on the Web and various attempts to solve that problem were presented, this time, the majority focused on security, authentication and privacy. these seem to be the major concerns nowadays. these issues are not necessarily only applicable to commercial applications - authentication for example might also be very interesting for scientific publications - but they are strongly driven by commercial information providers.
  
  
• "where is NetScape ? is there anybody from NetScape here ?"
  almost every speaker asked these questions. there were no talks and no representatives from NetScape at all. most speakers made jokes about the BLINK-tag etc., but at the same time, many speakers used the famous browser with the shooting stars in the upper right corner ...
  
  
• Java
  of course, Java was a hot topic. Sun tried hard to make everybody belief, that not the invention of the Web, but Java is the big deal. certainly Java may have a big impact on Web applications, but after all - as somebody stated at the conference - Java is basically nothing more than just another programming language. but the fact that both companies, NetScape and Microsoft have bought a Java license, may be a good indication, that Java WILL have a bright future. and in contrast to the HTML and HTTP standards, Java may stay a unique standard as long as Sun sells licenses and stays in control of the Java standard.

tutorials and sessions i have attended:

• monday, 11-dec-95: tutorials
  • an in-depth view of forms, tables and cgi
  • security, authentication and privacy on the Web (best tutorial of the conference !)
  
  
  
• tuesday, 12-dec-95: 1st day of conference
  • opening plenary session
  • objects on the Web
  • caching:
    • making WWW caching servers coorporate
    • proxies: limitations and potentials
  
  
  
• wednesday, 13-dec-95: 2nd day of conference
  • low level security in Java
  • security:
    • CCI-based Web security
    • smart tokens and their implementation
  • authoring tools
  • visualizing the Web
  • closing plenary session
  
  
  
• thursday, 14-dec-95: developer's day
  • HTTP V1.x
  • HTML V3.0

my conclusions:

the conference was titled "the Web revolution". some people belief the invention of the World Wide Web will have a similar influence on mankind as johann gutenberg's invention of the movable type printing process more than 500 years ago. while some people belief, the Web will replace printed books and newspapers, i agree with those who think, printed books will not be replaced by electronic media. however, there are printed documents which can and shall be replaced by electronic documents, like the phone book, maybe time tables of trains and airplanes and other documents, which change often. newspaper and magazines are also documents with a very short lifetime, but since less the 0.5 percent of all human beeings have access to the Internet, there is indeed a need for other means of distribution of the latest news.

in my opinion, this was a very interesting conference, even though the developer's day was somewhat disappointing. i got some ideas where the Web might go in the near future. i guess we will see a lot of debates about the new standard of HTML, at the same time, i'm afraid the we will end up with various standards. this may lead to a similar situation as with the various UNIX operating system flavors we have to deal with today.

i also belief we will see more and more commercial applications on the Web. the security first network bank (SFNB) has already been established and many vendors offer "safe" Web servers and clients. authentication and privacy are the big issues now. some of them may be addressed by the new version of TCP/IP (V6), others by secure protocols, "smart cards" and so forth.

as far, as our local ETHZ Web is concerned, i think the following questions should be discussed:

• how can we reduce the network traffic ?
  most clients have built in caching to reduce network traffic. but in an organization of the size of the ETH zürich, one or more proxy servers with caching capabilities may reduce the network traffic much further. there are already some proxy servers set up, but they have relatively small caches and they are not (yet) well known. it might be desirable that the "informatikdienste" set up a well known proxy server with a large cache. also the caching strategy needs to be discussed:
  1. what should be cached ?
  2. how long should a document be kept in cache ?
  3. what documents should be deleted from the cache if it runs out of space ?
  
  
  
• do we care about security on the Web ?
  as discussed in the security tutorial, there are various aspects of security on the Web. in an educational environment, i would expect that privacy is not a big issue (this is only true for the Web, but NOT for other Internet based services like ftp or telnet !). but authentication might be vital for scientific publications. it would allow to reliably identify the author of an electronic document, e.g. by means of digital signature.
  
  
• do we want the information providers to follow the standards ?
  despite of the fact that the current standard of HTML is V2.0 !, most information providers use any HTML extensions that are available. TABLES are already common and supported by most of the current browsers. but there are at least two different proposals which are similar for the basic tags, but different if it comes to more complex structures.

  there are tools available to check HTML documents. should we establish a service to check Web documents and what should be "our" standard ? HTML V2.0 ? HTML V2.0 plus tables ? the W3C's draft for HTML V3.0 ?

  but the more important question might be: how do we handle "non standard" documents ? shall we ban them from our servers ? do we refuse to set up links to "non standard" documents ?

in my opinion, the simple days on the Web are over. it was easy to bring a few Web pages on-line in the beginning, because it was more or less a game. but the Web has become a serious business today. during the Web conference, the Web was on the front page of the "USA today" almost every day. one article was about how Web surfers cost the industry thousands of dollars every day because they browse the Web instead of doing their job. in addition, they use up the bandwidth of the network, which disturbs the colleagues who try to do their "real" job. this means, not everybody loves the Web and there are serious concerns about the Web. some organizations have already restricted the access to the Internet or are at least considering to set up such restrictions.

another fact that makes life more complicated on the Web are the various HTML extensions: not only that you need a particular browser to enjoy many of the pages out there, but it seems the appearance of the information has become more important than its contents, which i consider a tragedy ...

hardcopy documents:

the following books are available at my office for anyone who is interested:

• attendee list
• poster papers
• tutorial notes
• conference proceedings

next WWW conference:

the 5th WWW conference will take place from may 6 to may 10 1996 in paris, france !

other trip reports:

• trip report from the 3rd WWW conference in darmstadt, germany, 10-14 april 1995
• sorry, but i missed the 2nd WWW conference in chicago, USA !
• trip report from the 1st WWW conference in geneva, switzerland, 25-27 may 1994