Greetings list, Here is a nobody proposing something to be picked up by more skilled people. The urllib deprecation thread is scary for many reasons. I cannot just ignore it as it affects me as a noob user. The security issues makes one think if we continue shipping something that always has known open holes, can we sleep soundly? On the other hand, removing the library completely is weird. How can programming be fun if it does not provide 'internet'-related libraries? Heu, should users start diving into C stuffs? Mere patches will work but seeing advances in 3rd party libraries, one think that to have something comparable in the stdlib seems a big task. A minimal library will work but is it the solution? I had the opportunity to get some ideas about how the asyncio implementation went. Folks who had really amazing async experience helped, even wrote PEPs. I think setting up a workgroup with experienced urllib devs and 3rd party library authors will be very productive. It might be true that the days of batteries included are over, but, it might be more true that space is cheap, the world moves fast and more effort is required than it ever was. The way forward will be deprecating urllib in the real sense of the world until it's peacefully asleep, all while introducing a new library. If we go the minimal version route, it might be a good idea to introduce umbrella libraries. Libraries maintained by core devs, but not included in the stdlib. This allows the stdlib to be leaner, more straight to the point. On installation python can have a command to install those extra libs. Proposing this on -ideas as i no longer understand what the original thread is about. And as stated often, i avoid this module of the stdlib because of its non elegent API. If something's no longer used by normal users, it is maybe the time to do something complete and elegent about it. -- A noob Python user who wants his favourite language to be as good as possible Q: Why not post on the original thread? A: Two core ideas in here: viz proposal to set up a workgroup and the umbrella libraries. As such it feels more natural to post on -ideas than -dev