On Wed, 3 Mar 2004, Ole Jacobsen wrote: > Paul, > > This is simply silly. > > What you are saying is that for religious reasons you are unwilling to use > FREE and widely used tools in order to help us develop our own. > > Next thing you'll be telling me PDF is a bad thing. > > If you want the IETF to be a place where more people can participate you > need to ditch some of this religion. ... > > the fact that realmedia and windowsmedia aren't interoperable means that > > we (this community) failed to recognize and address a common need, and > > that the world (including this community) is suffering for it. > > > > compounding this failure by adopting proprietary technology for the primary > > work of this community -- which is interior and published communications -- > > would be a bad, bad (bad) thing. This is not silly, it is just smart in a longer timeframe than you're thinking in. Proprietary tools that utilize a proprietary/non-open data interface are a serious problem for a variety of very sound, non-religious reasons (as well as for a variety of political and economic reasons, which is what I think you're calling "religious" reasons). "Free" is irrelevant to the issue, unless free means open. 1) Proprietary data formats and long term archiving of any sort of data are fundamentally incompatible. Ask anybody who has lived through the last twenty or thirty years of computer evolution how many documents they've lost or had to go in and rescue (sometimes at great expense) as the tools they were built with have disappeared. Sometimes along with their vendors. Other times the vendors simply decided to release a new version that was sufficiently incompatible that it could no longer manage the old documents. I think all of us can remember multiple instances where this has happened to us -- I personally have lived through wordstar, pcwrite, wordperfect, word, and several income tax programs (which are the worst, as one HAS to be able to access the records up to seven years later, which is a real problem with operating systems and Moore's Law). There is also the uncertainty even now surrounding the "encumbered" mp3 format versus e.g. the unencumbered ogg format. Formats used to encode long-term public records need to be open and tools to manage those records need to be available from many sources. So putting up realmedia shows is short-run glitzy and nifty and all that (even though lots of people won't have the players and cannot play the media) but it is long run foolish IF the production is intended as any sort of serious historical or archival record. 2) Using a proprietary data format that can only be accessed by using a proprietary tool (even a "free" one) leaves one vulnerable to all sorts of shenanigans and hidden costs. For example, nothing prevents the vendor from waiting until you have a large amount of valuable data built up with their format that would be very expensive to convert and then deciding to charge you. It's their tool, they can charge you if and when they please. Worse, since their tool is generally a closed source, proprietary object, there are the usual problems with libraries and compatibility when trying to get the tool to run on the wide range of platforms it is advertised for. "Free" may just refer to the cost of getting the program, but it may well cost quite a bit of time to install and maintain it, and time is money. 3) The Internet has been built on open standards from the very beginning. This is absolutely the key to its success and tremendous degree of universality and functionality to this very day. Any vendor can build a mail tool, an ftp tool, a tool using TCP/IP as a transport layer. Any vendor can build a browser, an http daemon. The specifications for those tools are laid out in RFCs, and modifications to the open standards proceed in a serious and systematic way. The Internet has RESISTED being co-opted by monopolistic vendors who have sought to introduce their own proprietary and "essential" layer of middleware on hidden protocols, although they continue trying. The DMCA makes it quite possible that if they ever succeed it will be tremendously expensive and damaging to the entire structure. You can call this a "religious" argument if you like, but I think it is really a statement of both politics and economics, in this case the politics of freedom and the economics associated with having lots of choices. So I'm afraid that I agree with Paul 100% on this one (although I respectfully disagree with him on others;-). The IETF absolutely should avoid using proprietary tools to create documents that they might wish to archive, and should strongly encourage the development of open standards for and open document formats (one data format, many tools both free and non-free) for data transmission on the Internet to ensure that the Internet NOT be co-opted by any single vendor and that records that might be archived today can still be accessed ten or twenty years from now without finding a DOS PC and a 5.25" floppy drive to reload an ancient piece of software. Or am I alone in think that this would be a Bad Thing? rgb -- Robert G. Brown http://www.phy.duke.edu/~rgb/ Duke University Dept. of Physics, Box 90305 Durham, N.C. 27708-0305 Phone: 1-919-660-2567 Fax: 919-660-2525 email:rgb@xxxxxxxxxxxx