Errors in recently published IEEE Std 802.1H-1995 (collected Email) Date: Fri, 1 Dec 95 11:18:33 CST From: keen@ncrssd.StPaul.NCR.COM (Hal Keen ) To: K.Doty@IEEE.org, p8021@nic.hep.net Subject: serious error in 802.1H-1995 Cc: henri.moelard@utrecht.NCR.com I just heard from a colleague who tried to use the newly published Std 802.1H-1995. He couldn't understand Figure 6 (the guts of the Standard), and when I checked it I discovered why. It was wrecked in post-ballot editing. The most important problem is the instruction to translate (tunnel) an RFC 1042-encoded frame going from CSMA/CD to an 802 LAN. This doesn't work, and isn't what we balloted or approved. Lesser problems are: The footnotes are mixed up, and they have been reworded to talk about table entries "needing" to match things, an anthropomorphism not in the original text. 802.1 and the IEEE need to figure out (a) how to prevent further promulgation of the error, (b) how to get the document fixed, and (c) how to prevent the IEEE editing phase from introducing technical errors in our standards. Regards to all, Hal Keen Date: Fri, 1 Dec 95 14:08:36 EST To: keen@ncrssd.StPaul.NCR.COM, lidinsky@hep.net From: k.dittmann@ieee.org (Kristin Dittmann) Subject: 802.1H errors Cc: k.doty@ieee.org, p.kelty@ieee.org, v.zelenty@ieee.org Hal and Bill, I am very concerned about the errors introduced into the standard post-balloting. Bill should provide a detailed listing of errors and I will investigate each of them with the project editor and technical editor of the document. There are several things we can do to ensure that the errors are not promulgated. First, we will issue an errata sheet and mail it out to all past purchasers. (Copies of the errata sheet will also be made available at the next 802 meeting for anyone who got a free copy of the standard.) I will process an errata sheet as soon as I receive the listing of errata from Bill. Second, we will announce the availability of the errata sheet in the IEEE Standards Bearer and on our WorldWide Web page. (If 802.1 has a WWW page, this might be a useful news item.) The errata page itself can go up on the Web for people to view and print. We will of course fix our electronic files of the standard so the working group doesn't accidentally promulgate the errors in a revision. Finally, I will try to ascertain what caused errors to be introduced in the IEEE editing phase. Our editors are meticulous and are instructed not to change the technical content of the standard in any way unless instructed by the technical editor or working group chair to correct a previously undetected mistake. I know that the technical editor from the working group made some "corrections" to figures, and it is possible that in doing so he inadvertently introduced new errors--and, of course, it is possible that the mistakes are entirely ours.We do, however, rely on the working group review to ensure the technical accuracy of the standard after editing. Perhaps my investigation of the problems in this published standard will reveal some weaknesses in the editorial, production, and review processes that need some consideration. In any case, I will report back to you after I have had a chance to investigate. Kristin Dittmann, IEEE Standards k.dittmann@ieee.org tel. (908) 562-3830 ******************* IEEE Standards Web site: http://stdsbbs.ieee.org/ Date: Fri, 1 Dec 95 13:56:31 CST From: keen@ncrssd.StPaul.NCR.COM (Hal Keen ) To: k.dittmann@ieee.org, p8021@nic.hep.net Subject: Re: 802.1H errors Cc: k.doty@ieee.org, p.kelty@ieee.org, v.zelenty@ieee.org Kristin, Thanks for the quick reaction from the IEEE to my report of the error. As I sent the original alert to the entire 802.1 participants' list, I will copy them on this exchange as well. I'm glad to hear we can distribute an errata sheet with a high probability of good coverage. It would seem most important that the errata sheet is complete, so we should make certain the entire published document gets reviewed against the final (post-sponsor ballot) Working Group draft (draft 5). I will try to do a parallel reading over the weekend, and others are invited to do likewise. We probably need to allow at least a week for that; the weekend's already begun in some very relevant places. As to how the error got in, I hope you understand my reference to the "IEEE editing phase" was intended to include all participants in the progression to publication. I'm not throwing blame at anyone, especially Standards Department staff, and I think directing our attention to process questions is best. I'd have been more careful in phrasing that, except I wanted to get a "heads up" out to 802.1 ASAP. Best regards, Hal Keen Date: Fri, 1 Dec 95 17:03:30 EST To: keen@ncrssd.StPaulMN.ATTGIS.COM (Hal Keen ) From: k.dittmann@ieee.org (Kristin Dittmann) Subject: Re: 802.1H errors Hal, Please stay in touch with me about this. Just so you know, I'll be out of the office next week until Friday. When I get back I'll be sure to treat this as a high priority. Thanks for your help. Regards, Kristin From: Andy Davis Date: Fri, 1 Dec 1995 21:37:55 GMT To: p8021@nic.hep.net Subject: 802.1H problems I'm flabergasted if we got it wrong. I'm also otherwise unable to comment, since my genuine printed copy is in a different country to the one I'm in right now. I certainly did not read the printed version in detail, since after all, I've got the original master to look at. Ummm, at least I thought I did! A quick flurry of finger pointing reveals that Alan was the last person to touch the document in the long path to printing (sorry Alan), but I still feel some responsibility too, and so I will gladly help out - but it will be another 7 days before I am re-united with my real book in order to compare and contrast the two. Meanwhile, if there's anything constructive I can do, please let me know. Andy Davis PS: Y'know we really should have just published it as an RFC instead. Not only would it have been right that way, it would have been published in the year 1993 :-)