If you have an in-progress discussion that's still finding its way, that's not going to happen. And the curators may or may not have their own agenda.
Perhaps technical discussions are just messy because the problems are messy and don't have obvious solutions.
In the context of Rust, the bigger problem, IMO, is that things that aren't Mozilla/Firefox driven don't get driven to conclusion/completion. The discussion simply peters out with no resolution and then remains open indefinitely (look at how many issues from Rust 1.0 are still open with no conclusion, discussion or resolution.)
I'm curious, what kinds of things do you see that are being prioritized for Firefox or Mozilla over things the community wants? I can't think of anything, but I am also extremely biased.
> (look at how many issues from Rust 1.0 are still open with no conclusion, discussion or resolution.)
Issues for the project are prioritized based on need and ability to complete them, not age.
> I'm curious, what kinds of things do you see that are being prioritized for Firefox or Mozilla over things the community wants?
It's not about prioritizing over--aka "Mozilla wants X and community wants Y and those are in conflict." I don't think I've seen that even though it probably has happened at some point.
It's more about "Mozilla wants X and allocates resources so that goes to completion while community wants Y and languishes because maybe one or two people are actually qualified to work on it and they have full-time jobs."
> Issues for the project are prioritized based on need and ability to complete them, not age.
Quite a few of the Rust 2019 announced objectives saw very little progress, and yet people wanted to move onto new objectives in 2020. Erm. Why did those 2019 objectives see so little progress when they were important? Why are they now unimportant in 2020? Apparently Mozilla didn't seem to think those objectives were too important as they didn't seem to allocate resources to attack most of them.
However, this isn't a new problem. For example, you transferred a bunch of issues for the 1.0 timeline. Some of them were transferred into things like "This should be a library". Okay.
Someone later wrote that library. It has become a de facto crate solution and people use it. Yet that RFC/Library issue still stands open. The library author may not even be aware that such an RFC/Library issue exists. Whose job is it to close that issue and anoint that library/crate as "The One(tm)"?
Some of these issues have been untouched for 4+ years. No one is fooled that these are "in progress". They're dead and closed even if Github still marks them green. They might as well be closed out anyway as the language is so different from that milestone that many of them aren't even relevant anymore.
None of this is new nor specific to Rust. However, Rust is getting hit with it a bit more strongly because it's in a growing-phase where there is far too much work to do and far too few hands to do it.
(Funny you should pipe up. I'm sorry, I am certain I filed a Github issue on a Rust library issue asking for one such library to get a pronouncement from you, but the issue seems to have simply disappeared down the memory hole. It is a little unnerving that Github can just wipe out something I filed without reference or history.)
Thanks, this is helpful. I still disagree with a lot of the characterization here, but it’s much more reasonable than what I thought you were saying, and regardless of my opinion on your opinion, I’ll try to think about why things may appear that way.
Thanks for giving me the slack. This probably sounds harsher than I intend. Putting negative information into text almost always comes off worse than the speaker intended.
"Timeliness in process" is extremely difficult. Python does better than most, but even they have issues with RFCs that never achieve consensus. Also, one of the Python koans is "The standard library is where modules go to die" so Rust pushing things out of stdlib has some advantages (as well as disadvantages).
Part of this is also probably personal frustration on my part. I'm apparently using Rust in a lot of ways that are "off the happy path" and I bump into "Here Be Dragons" sections of the Rust landscape quite a lot.
I haven't paid attention to the yearly goals, but the story you are telling is quite believable (and also sad). To me it sounds like a disconnect between people who do planning and people who implement things. Thank you for sharing!
If you have an in-progress discussion that's still finding its way, that's not going to happen. And the curators may or may not have their own agenda.
Perhaps technical discussions are just messy because the problems are messy and don't have obvious solutions.
In the context of Rust, the bigger problem, IMO, is that things that aren't Mozilla/Firefox driven don't get driven to conclusion/completion. The discussion simply peters out with no resolution and then remains open indefinitely (look at how many issues from Rust 1.0 are still open with no conclusion, discussion or resolution.)