New builds for 3.6.0. LazyDodo and Thomas helped me out with the buildbots, so my builds are now in the patch build section on the blender website. Tons of time saved, and no limitations as far as file size (250MB limit for free stuff on gumroad) or with HIP and Windows.
Builds for 3.6.0 - PR109150
Source
More details in the blender artists thread here:
is this going to be a part of official Blender?
Thatās the plan, yes. Lcas proceeded based on a bit of feedback from Julien Kaspar, moved settings to the toolsettings and made a design doc based on his advice.
The devs have been kind and generous, but I havenāt heard anything outside of the help theyāve offered along the way. I want to make it clear that nobody asked for the things that Iāve done. Iām not entitled to anything, not even a āthanks, but no thanksā but Iād like to hear anything they have to say.
Thereās been very little interest from:
My expectations were low, but the average user could not care any less. Iāve learned a lot, and got some features I wouldnāt have otherwise. Itās been fun, but Iām moving on to better things.
Oh that would be a bummer. The whole time I was reading the thread I was pretty much expecting this to go into main at some point, somehow. Who do we have to poke or which petition do we have to sign to get the attention to get this considered? 
Honestly it was kind of expected. People just want to be able to select polygons in Blender without face dots being mandatory. If they are instead presented with an essay about dozens of different options, almost no one will spend mental energy to understand whatās going on. If this was a simple selection without facedots patch, it would have likely gained much more interest.
I was told that the amount of different options and approaches was just for research purposes, and that there will be eventually just simple, condensed patch incorporating the results of that research and exploration, but a year later, it has obviously not moved past the researching of too many options and approaches phase.
The features are split up, and there is only one approach to them. I made a custom build that incorporates all of the different things Iāve done, beyond the ones below that deal with selection.
The devs do not appear interested in any of them. Like you say, it was expected. The fact that so many of them helped me, time and again, shows you that they are great people.
What I am saying about the average user, is that they donāt care how selection works. You present them with blender, and they will use it however it happens to be.
Iām watching your progress since the great debate that happened on phabricator in mesh-select-through task.
My guess is that all of those who cared are burned out of this topic and accepted the state of selection as it is. Itās kind of sad, but that does not mean the problems with selection in Blender does not exist or that your work is pointless.
Personally Iām mostly interested in select through, and would really like to have it in Blender. Rest options also looks good. Iāve briefly tested one of the previous builds, but havenāt got enough time to delve into all the options and make a proper response. There is a lot to take in. Iām also waiting to test your Linux build with AMD GPU, but the model Iām using right now is disabled because of HIP compiler bug.
I think that the main problem is lack of feedback from the devs in the first place. I assume that the main point of making a design task is to gather feedback from developers and decide if the direction is OK and that is not something users can decide on.
Thanks, that means a lot. Youāre right about the design task, it was to get some developer feedback. Based on past experience I assumed that would be like it is, crickets. So I also wanted to go to youtube, blenderartists, etc, and guage what people thought outside of here. If there was anything, go from there, make more videos about the process. Maybe that leads to something. But the initial response was non-existent on all fronts, so thereās no point. Disappointing, but I got other things Iām ready to put more focus into now.
FWIW - I just didnāt respond much here because I was mostly silently agreeing that new selection methods would be apreciated if done right. I just assumed you had the feedback from devs since you kept going for so long. 
Thanks, I didnāt do anything with any serious expectations outside of personally using some of the features and to learning how to code. Itās a different language but the concepts are the same. You get an idea, and look for different ways to solve a problem. A hobby with some potential later, maybe.
I donāt think the devs were ever interested, they were just being helpful, and I needed help for almost every feature on some level. Never worked on it very consistently, but still, it took enough time where the amount of things remaining that I could figure out or clean up for general use were not worth it without some outside incentive. Either from the general blender community, or the devs in a more official way. It went nowhere, but it was worth the time I spent trying to be more presentable.
I think these options are essential for Blender.
There are 30 artists using Blender in my studio and 90% use the Xray selection addon. Fortunately, this addon does a very good job and I thank its author, but it would be nice to have all this by default.
Icas has done a great job and the current version brings powerful options without being too intrusive. You can still use blender as you did before if you donāt like it.
This is very true. A great deal of dev time is devoted to new features but the basic interactions like selection and navigation were never brought into an acceptable state, IMO. I desperately want these features in blender but Iāve mainly been quietly watching from the sideline as you all worked out the nitty gritty of best presentation and places in the UI for features to go, and such.
I wonder if @Draise and co. would be interested in incorporating the diffs into the bforartists fork or if they think the maintenance overhead would be too much work.
A great deal of dev time is devoted to new features but the basic interactions like selection and navigation were never brought into an acceptable state, IMO.
I feel the same. I donāt want to complain, because Iām grateful for the time theyāve spent helping me. But I also donāt think any of them care what I have to say, or would bother reading it in the first place, because they have better things to do. I mean that for real, not in some complainy type of way. Same goes for the rest of what Iām about to say.
This is my outsider take on the situation: If Iām advised to ping the guy and post a design doc in the module chat, but when I do that itās as if I donāt even existā¦ that guy is way too busy, and unfortunately everything has to go through them and a couple other top level people. I donāt know how these top devs keep their collective heads above water, but somehow they keep things going. There isnāt enough people to sift through the bs. That is the answer to why anything in blender is subjectively (sometimes objectively) inadequate, weird, or broken. Add to that the human nature aspect of, āLook, flashy new thing that people will get excited about, letās do it.ā
But I am pretty sure that isnāt the whole story. Because if you could just fix something as vital as selection, why wouldnāt you? Select through, fine, that has been rejected for what I would call philosophical reasons. But near select grabbing unseen edges and faces is pretty terrible. Unusable in my opinion. That one should be contenting for priority one, and who knows maybe it is, and it just takes a really long time to get anywhere.
Out of necessity, I think the only people who get any heat are the ones who earned it, who proved themselves, and who really know what they are doing. That obviously aint me, for a number of reasons. If I could get someone lower down but on an āAble to get the time of day from higher upsā basis to fix up some of the things Iāve done and put their name on it, maybe it would get some traction 
Wow such a shame!
Ok, letās assume that the main problem is that Managers/Code Reviewers ecc.
Have no time to spend on project like yours.
Actually we donāt really 100% know if thatās the main issue, but letās assume that it is the case.
I wonder if having a couple of people more in the dev team dedicated just to manage, guide and review this type of projects and in general spending time on fixing long lasting usability problems, while all of the rest of the team focuses on new features would be a good strategic solution.
Iām not talking about directly coding , but ājustā manage and review a limited but promising contributors like you seem to be, so that the hard design and development work doesnāt go wasted.
A little bit like GSOC , but with a couple full time dedicated resources in the official team that just manage these project. The number of projects choosen of course depend on the capacity of those resources to cover the work to be done.
Then again - isnāt exactly this where an Open Source project should be different? That you can just do something and propose it as an official patch for consideration no matter how insignificant it seems at first?
I get that any patch has to conform to the projectās design philosophy and that there will be times where something will be absolutely rejected for this reason (at least in the official main branch). But I feel like this is where Open Source is always promoted as being this counter ballance where everybody from the āoutsideā who can add something that fits the paradigms at least will get considered. Unlike comecial projects which simply are a blackbox to anyone not working at the company or maybe being a big player who can weigh in on future features.
I mean - Iām saying this in a very general way, here. I havenāt looked into what is the official proposed approach to any feature being considered into Blender. If there is an upfront approvel process before you even begin your modifications or whatever - or if this thing followed it or not. Or how much noise it has to generate in order to be noticed or reviewed.
Cause from the outside (without knowing if some established protocol wasnāt followed or whatever) this seems kind of discouraging to new developers. When the general tone is one one hand: āactually present something before getting active to see if itās a good idea to followā and then just not getting any feedback on actually presenting something. I feel like no feeback at all is actually the worst thing because itās just undecided and up in the air.
Iād like to think itās slightly more nuanced. The devs did help, because they see there are advantages to be had in the area. They are probably short of hands, seeing that the Snapping work (also very much a core feature) has only been recently picked up again by Germano.
People who exclusively use Blender donāt really know about these features and what advantages they have. Yet, many who migrated over from Max or Maya did request these kind of things in the 2.8 development cycle. Though perhaps they are already satisfied with the X-ray Selection addon. I definitely think you work remained under the radar for most people, otherwise we would surely see more people responding.
And maybe people are too lazy to read the lengthy descriptions. Iāll see what I can do by plugging your youtube video in some discord groups.
Also, whatās the process of assigning reviewers to a patch? @ThomasDinges or @LazyDodo is this something you either one of you could chime in on?
@Hologram The process is described on our wiki: Process/Contributing Code - Blender Developer Wiki
Let me know if you have any further questions. 
PR author must tag relevant developers from relevant modules as reviewers. Author of this PR did not He just added bunch of stuff on development and expected developers to discover them and get mindblown by amazing features. There is a system in place without which nothing will work, and everybody has to follow those. You canāt complain about why isnāt everybody happy you did this and takes over and pats you on the head. Iām sure if PR author followed, and actually learned the procedure for merging PRs it would go much better.
But regardless I still hope most of the features proposed donāt get added in main, there is a reason most of the Blender users donāt need these selection tools, we donāt come from archaic paid softwares.