Hello, In the Message; Subject : Re: packaged version of postfix Message-ID : <98a8401f-8964-36c2-adee-b8a8008bf132@j4computers.com> Date & Time: Thu, 6 Jul 2023 10:38:22 -0400 [J] == joe a has written: J> >> Have I a clue? MN> > ChatGPT replies; MN> > Based on the information you provided, it seems that using LMTP for MN> > communication between Dovecot and Postfix on the same box is not MN> > recommended. According to Dovecot documentation, you can use LMTP MN> > and the Postfix setting “reject_unverified_recipient” for dynamic MN> > address verification. It's really nice because Postfix doesn't need MN> > to query an external datasource (MySQL, LDAP…)*1. MN> > *1 https://doc.dovecot.org/configuration_manual/howto/postfix_dovecot_lmtp/ J> > J> > Does this help you？ [...] J> ChatGPT needs a tuneup. Dovecot in fact, to my readings, shows that LMTP is J> preferred in this case. The link is one I followed, far as I can tell J> correctly, which appears to describe setting up LMTP. Thank you for the verification! As you probably already know, ChatGPT searches by keywords, combines the hits, and displays the results in human spoken language, but, 1. arbitrary use of keywords 2. Sentence structure that does not care about logical contradictions As a natural result, it may show false conclusions without hesitation. Therefore, we should be careful when using ChatGPT, but since it is a human language, it may mislead people's minds..... BTW. I'm installing postfix, but not dovecot. But, I get the same result as yours; # postconf -a cyrus dovecot # postconf -A cyrus I wonder why ..... ? Regards. --- ┏━━┓彡 野宮 賢 mail-to: nomiya @ lake.dti.ne.jp ┃＼／彡 ┗━━┛ " Hassabis says that no one really knows for sure that AI will become a major danger. But he is certain that if progress continues at its current pace, there isn’t much time to develop safeguards. "I can see the kinds of things we're building into the Gemini series right, and we have no reason to believe that they won't work," he says." -- "Google DeepMind's CEO Says Its Next Algorithm Will Eclipse ChatGPT" --

Hello, In the Message; Subject : ChatGPT was: packaged version of postfix Message-ID : <20230707124701.5e7f6286@acer-suse.lan> Date & Time: Fri, 7 Jul 2023 12:47:01 +0100 [DH] == Dave Howorth has written: DH> On Fri, 07 Jul 2023 11:16:58 +0900 DH> Masaru Nomiya wrote: [...] MN> > Thank you for the verification! MN> > As you probably already know, ChatGPT searches by keywords, combines MN> > the hits, and displays the results in human spoken language, but, DH> I don't think posting ChatGPT results is very helpful. AI seems to be DH> better received where its results have been looked at and interpreted DH> by an expert who can check the results for accuracy. So I'd suggest we DH> keep this list to the output produced by human users as before. Sorry for bothering you all. At the request of a young colleague at my former workplace, I investigated the effectiveness and limitations of ChatGPT and reported on it by email the day before yesterday, and was asked for a report to show the members, which I have just sent. Now, I have paid attention to the following two points in addition to the accuracy of the description of ChatGPT. 1. how to avoid the legal risk of copyright infringement 2. how to avoid the legal risk of use of personal information As you know, even the Web, which we can freely view, there are many websites that we can freely view, but that claim to have copyrights and prohibit secondary use. In this context, I have found that there are two ways to avoid legal risks related to copyright: 1. make it difficult to identify the source of the quotation, and. 2. display the results in an ambiguous manner. Specifically, in the case of Linux-related articles, the phrase "written in a forum or bulletin board" often appears in the results, without a link, but no such thing can be found by any separate search engine. The same is true for social science questions, which I assume is the result of a uniform measures against risk aversion. Surprisingly, two American SF authors recently sued openAI copyright infringement. In other words, science fiction writers usually create new words when they depict their imaginary future, just as Chapek created the new word "robot" in the Czech Republic, but ChatGPT used the new word in the results display without knowing it and was sued for copyright infringement. You may have noticed that openai has also been inactive for some time, which I believe is a clear indication of poor risk management. I am sure that they modified their program during the hiatus, but I am afraid that this will make the results displayed more and more ambiguous. Secondly, the ambiguous display of results is evident in the response to Peter. My simple question is, How do I set up processor.max_cstate=5 in openSUSE? The answer given by ChatGPT is not an answer for openSUSE users, as Andrei also points out. This can be seen in ChatGPT's answers to joe and Marc, as well. Furthermore, CahtGPT's way of avoiding the legal risk of using personal information is to enter your name into ChatGPT and ask who you are, and you will be figured out. It told me, I am an associate professor at Shanghai Institute of Technology. Why doesn't it just make me a Hollywood handsome star? Anyway, the use of social sciences is even more disastrous. As you know, in social science, it is important to explain the occurrence of a problem phenomenon based on past facts, but ChatGPT fails to do this and tries to explain everything in relation to current events. This makes it completely useless. In this case, it starts its answer with the usual phrase, "There are many opinions". For these reasons, I will not use ChatGPT. Fortunately or unfortunately, since yesterday, it is no longer possible to use ChatGPT over vpn. Sorry for the long sentence.

Op vrijdag 7 juli 2023 14:45:46 CEST schreef joe a:

On 7/6/2023 10:16 PM, Masaru Nomiya wrote:

...

# postconf -a cyrus dovecot # postconf -A cyrus

I wonder why ..... ?

Regards.

--- ┏━━┓彡 野宮 賢 mail-to: nomiya @ lake.dti.ne.jp

I was mildly (?) chastised on the postfix list on that issue. Essentially, working as designed, "postfix does not do client side".

Lack of fully understanding postfix, or even having a good working knowledge, appears to be part of my issue. Or I am getting too old for this stuff.

joea Below is the description of what postconf -a and postconf -A is about (from https://www.postfix.org/postconf.1.html):

Options: -a List the available SASL plug-in types for the Postfix SMTP server. The plug-in type is selected with the smtpd_sasl_type configuration parameter by specifying one of the names listed below. cyrus This server plug-in is available when Postfix is built with Cyrus SASL support. dovecot This server plug-in uses the Dovecot authentication server, and is available when Postfix is built with any form of SASL support. This feature is available with Postfix 2.3 and later. -A List the available SASL plug-in types for the Postfix SMTP client. The plug-in type is selected with the smtp_sasl_type or lmtp_sasl_type configuration parameters by specifying one of the names listed below. cyrus This client plug-in is available when Postfix is built with Cyrus SASL support. This feature is available with Postfix 2.3 and later. It only lists what is available, so what values smtpd_sasl_type and smtp_sasl_type or lmtp_sasl_type can have and depends on what has been build- in. -- fr.gr. member openSUSE Freek de Kruijf