The great pain and setbacks is the trifecta of the documentation stating that `request` has many options that it does not, the parser not rejecting unsupported request options, and the configuration payload receiver not warning about receiving unsupported payloads.
Then there is a string of bugs that made figuring out the actual problems much harder. > On Jan 6, 2025, at 7:56 PM, William Rusnack <williamrusn...@gmail.com> wrote: > > To Stuart, > > Thank you for your civility. > > For the copyright, the code changes I have submitted are minimal and follow > the patterns that already exist in the source; the documentation submissions > are minimal but I could rewrite them by hand if necessary. > > I have not submitted questions with the bug reports but I have submitted some > refactorings which is probably inappropriate. Where is the documentation on > the appropriate procedures for submissions based on the category? > > "20-odd reports in a short period whatever the reason is hard to deal with” > most other projects have a triage process and issue tracking. Is this not in > place? Also, Reyk Floeter <r...@openbsd.org <mailto:r...@openbsd.org>> does > not seem to be present anymore and it seems like there is no specific > maintainer for iked that is able to quickly access these. In light of this, > is there a way for specifying the “severity" of the bug report? And, how > should I limit the number of bug submissions for the bugs that I do find? > > Lastly, if there is not opposition to receiving more legitimate bug reports, > I am planning on auditing unbound and nsd soon. > >> On Jan 6, 2025, at 4:06 AM, Stuart Henderson <s...@spacehopper.org >> <mailto:s...@spacehopper.org>> wrote: >> >> On 2025/01/05 19:46, William Rusnack wrote: >>> To Peter Hessle, >>> >>> I said that I use (utilize) multiple AIs but I do not solely >>> rely on them for generating the content. I have: >>> - actually been using iked >>> - looked into the sources extensively >>> - found multiple actual issues >>> >>> Believe me, I do not want to be looking into the iked source >>> code, but multiple sloppy implementation and documentation >>> issues in it have caused me great pain and setbacks. >>> >>> I have not been able to find any stated policy that AI cannot >>> be utilized for assistance in contributions. >>> >>> If you have problems with the actaul submission from my >>> indepth audit on configuration payloads then please be >>> specific. >> >> Apart from anything else, the copyright status of llm-generated code and >> doc changes is unclear. As far as I'm aware provenance isn't usually >> taken into account by those systems and there isn't usually an option to >> use only sources with an acceptable license. >> >> Also a dump of 20-odd reports in a short period whatever the reason is >> hard to deal with. Worse when it's a mixture of some things which are >> bugs, some things which are questionable, and some are misunderstandings >> (but we don't know whether that's a misunderstanding by you or whatever >> system you used to "help"). >> >> If you could explain which of the issues that you reported have caused >> "great pain and setbacks" that would help prioritise things because I'm >> not seeing which of the reports where that might apply. >
