Path: ...!3.eu.feeder.erje.net!feeder.erje.net!news2.arglkargh.de!news.karotte.org!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail From: VanguardLH Newsgroups: alt.comp.software.firefox,news.software.readers Subject: Re: Keywords header Date: Thu, 6 May 2021 05:52:32 -0500 Organization: Usenet Elder Lines: 102 Message-ID: <5xpuhyxthcy3.dlg@v.nguard.lh> References: <1k7vwsvkotabm$.dlg@v.nguard.lh> Reply-To: invalid@invalid.invalid Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Trace: individual.net HY0IDUvqn8NJPPYXWpaV8g4iv2zr0isTaeE1u3VxgzyOB8CudY Keywords: VanguardLH VLH811 Cancel-Lock: sha1:m6IcBy78hkbIjxop1p2fLwSJgOM= User-Agent: 40tude_Dialog/2.0.15.41 Bytes: 6054 Eli the Bearded <*@eli.users.panix.com> wrote: > Crossposted and follow-ups set. FollowUp-To ignored. It is rude to yank away a conversation from other readers in the original newsgroup to which you posted by redirecting replies to elsewhere than the original location. > VanguardLH wrote: > >> I see you decided to steal my Keywords string. > > It is a rn / trn feature to preserve the Keywords: in follow-ups. I > seldom examine them, since they are usually blank. Checking my post > archive I see this is not the first time you've gotten me like that. > ... > > Can I ask why? Or what you hope to get out of those keywords? In addition to the From header, I use both the right token of the Message-ID and the Keywords headers to make sure anyone that wants to identify me, even to plonk me, has multiple and stable headers on which to filter. If they wanted to ensure their filter only targeted me, or they wanted to ensure a search only showed my posts, and not accidentally on someone else, they can test on: - Path injection node - AND From - AND Organization - AND Message-ID - AND Keywords. Or, they could use just the From header for easy if filter definition, too, but it's easy to catch forgers, especially since their Path injection node won't be the same as mine (well, not for long since I use responsive Usenet providers that can kill the forger's account very quickly, and another reason I quit using freebie Usenet providers since the forger would have to pay to get an account). I give lots of compounded targets to identify me. https://tools.ietf.org/html/rfc1036 Section 2.2.9 A few well-selected keywords identifying the message should be on this line. This is used as an aid in determining if this message is interesting to the reader. https://tools.ietf.org/html/rfc5322#section-3.6.5 The "Keywords:" field contains a comma-separated list of one or more words or quoted-strings. ... These three fields are intended to have only human-readable content with information about the message. ... The "Keywords:" field contains a comma- separated list of important words and phrases that might be useful for the recipient. I would also use the "Comments:" header to further strengthen my Usenet identity, but my client doesn't let me add that one. No, I'm not going to PGP-sign my posts, because it is stupid since no one in Usenet is going to bother doing the lookup. While I can select a view that always shows all headers, that is usually a bunch of noise (as is often the attribution lines where posters think they have to add lots of duplicated info that is already available in the headers). I only occasionally look at all headers, so it is possible that I previously missed someone just copying my Keywords header into their reply. From what I seen in many NNTP clients, they generate their own Keywords header, if specified (non-blank), not forward a value from what some other client specified in a parent post. The RFC definition of the Keywords header is rather vague and very terse. Just didn't figure any client would not use its own value. I don't delete any unwanted posts. Instead my filters colorize them and add an Ignore flag. I use a default view of Hide Ignored Posts; however, if I need to check my filters for false positives or someone mentions something in an otherwise hidden post, I can just switch to the Show All Messages view. When I showed all messages, including the ignore-flagged ones, I saw your post. It was colorized, because it originated from Panix. So, I looked at the raw source of your message to see your Keywords header duplicated mine instead of your client adding its own value. Panix has been ignored-flagged by me ever since they decided to spamify all posts that originate at them by appending a deliberately invalid signature (so clients that hide sigs won't work). I wasn't interested in seeing posts by users of a spamifying Usenet provider. I've see Avast users, and other anti-virus program users, that spamify their Usenet posts (and e-mails). Avast does the same as did/does Panix: use an invalid sigdash line followed by 1, or more, lines of spam announcing the users employs Avast. The default config in Avast is to add the invalid sigblock. I'll alert such users that they are spamming their choice of anti-virus software, and to turn it off. If they continue to refuse, and because they choose to be spamming affiliates, they'll get kill filed. I did the same to Panix when they were appending their invalid sigblock to all submissions. Has Panix ceased spamifying the articles submitted to them? I don't see it in your posts. Did they ever offer free trials, and those are the submissions they spamified (as a lure to get those users to move to their pay service)? If Panix is no longer spammifying their articles, I will modify my filter on them to stop flagging them as ignored. I'll still colorize them for awhile to watch if any spammified posts show up.