Any plans for a better Editor?


#1

The build/create menu is still miserably bad to work with, i’m not even going to butter that up. I’d rather lick paint than build with it right now.

We need a more streamlined way to assign textures we’ve already uploaded ATP to assets in our Domains

I’m nto saying Second Life’s Editor is perfect, David’ made a decent editor this time last year but it breaks now as far as I can tell. We need more ease of use features. Things that take me minutes to do or test in other platforms wind up taking hours.

Is anyone interested in working on a new Editor with me in the future even?.


#2

It’s simple since around april building got complete non interesting anymore. Since the moved all desktop stuff in tablet type dock things work not good. But the build tools where never really good.

The tool is just good enough to load oject look at it. mabye try to move it. Not hope it get better soon. And now i would sy not care about it anymore. it did take to long.


#3

Its been said quite a bit of times since the alphas and beyond that to be honest. High fidelity them selves don’t have any plan on that, but it always helps if we push a bit and do tickets and so forth.

Last week finished up some QoL improvements / Rewrite of the Particle editor that should make adjusting particles much more easier with an included responsive texture preview, but it still adhering to the Hifi Tablet UI they seem to be pushing for. It is still waiting for Code Review and eventual merge of master.

The new Particle UI is a bit more responsive and was a step ladder I am trying to eventually use to improve the main editor UI later on down the line. but right now busy with other tickets.

To be brunt It is a massive undertaking. The entire Edit / Create tool is a hodge-podge of multiple methods, working in 3 different layers ( in-world, qml, and webview). Already on the UI side there probably is 6-8 different types of methods and logic all over the place. The UI library that was developed for the particle editor would make it a bit more maintainable, but it still wouldnt address the selection, drag, and widget logic which happens inworld, which honestly need most of the improvements on, which david’s script was doing to a degree, rather than the tablet ui.


#4

Would really enjoy just a small tweak to make the ‘up-and-down’ arrow control easier to grab, visually separated from the sizing controls, without tearing the whole thing apart…


#5

Yep, even that “small tweak” is a doozy because everything is so decentralized atm without extensively falling back to jquery, making the entire thing a even more of a slog to load. Have you seen the condition the code has been in?

Otherwise all of these would have been done already.


#6

Yeah now that i’ve opened things up i can tell it’s gonna take a good deal of work. I honestly feel that it’s probably working well worth in the long run. Maybe i’ll start with David’s Edit script as a base it’s more in line with what i’d like to see, my biggest concern is building on top of the mess that it currently is.

Unfortunately i think that for a great deal of the Alpha and even currently we’ve been blinded by the wonders of VR, but now that my headset has broken and i’m trying to build something instead of play with my VR toys i’m seeing a lot of things that need a lot of work. Grabbable objects are super cool and all, but they’re pointless if building an environment to interact with each other in is currently a huge chore.

The JS API documentation is pretty out of date at the moment too.


#7

Yes, unfortunately my editor has languished and there are half a dozen reasons why. One of them being an announcement at a weekly meetup about a year ago that they were going to redo their editor, though obviously this hasn’t happened and I don’t think they have any plans to.


#8

Hmm, yuck, this is the downside of open source (of course lots of upsides). Sounds like many have had their hands on it without a consistent vision…

Can we create a clear set of requirements of what needs to be done? This might be a first step to building something more useful.

That being said… i have finally gotten used to the edit controls, took a while. As clunky as they are, they do work sorta-ok-for-now-ish.


#9

what we have is based on https://www.tinkercad.com/
to give u a clue what they are aiming at


#10

List of Requirements on my end at least would be pretty simple

  1. A Maya/Blender Style Interface for positioning/scaling/rotating objects
  2. A “drag and Drop Material Editor”
  3. Easier editing of Compound Objects (That’s the hard one i’m betting)

What we currently have is barely passable. I constantly end up accidentally copying things, or dragging something hundreds of meters away.


#11

|Sorry to sound negative.
But We talked for 1 year against a wall i think.
The result is thing got worse. Not better.

This year it got clear that opens source always seems to have the same problems.
And i have experience with opensim.


#12

No i get that. but we can’t realistically expect a lot of creators to come over from platforms where the ease of use is much higher.

This is the major thing i hear from other SL Creators and Builders when i mention HiFi to them, and let’s be honest Second Life players are HiFi’s market whether HiFi wants to admit it or not.

Something needs to be done HiFi is in “beta” right now but i really can’t see much different that alpha other than being able to sit on things.This platform has so much potential but once again I feel like we’re being blinded by VR. Some changes really need to be made to the Desktop portion of HiFi to make building VR Experiences rewarding and streamlined.


#13

Who say that SL users go HiFi and not that other linden lab platform ?
The make a cgood change compared to the statee of high fidleity is right now.
I tried to avoid that i think the last 15 months. But i get a feeling the lose. Or are far behind.


#14

That’s other thing is all prebaked. It’s graphically nice but if you can’t rez… It’s going to be another close market… So I think HF has still good chances to be the logical continuity of SL.


#15

This is a good list, it’s a final state that should be table stakes for a release product. It seems to cross a few different internal plumbing systems, are there intermediate steps we need to get there that are more manageable for our (often volunteer) developers?

I think its kind of confusing when they are trying to use the same tools for editing a ‘primitive’ in world as opposed to a ‘model/entity placer’.

In the short term if the up and down arrows could be made taller so it goes out the top and bottom of the object, and a checkbox to hide the sizing controls might make it more like a model placer.

For materials wishlist would be integration with substance designer .sbsar files we would pack and drop in world.


#16

If Sansar is going the direction we’re seeing it going it won’t be very Second Life-like, High Fidelity as of when I had access to Sansar was more like Second Life than Sansar.

Most coders and modelers I know are a lot more interested in High Fidelity. I’ve managed to network with a lot of talented people in Second Life and almost all of them have really just been waiting for this feature or that feature to be added to High Fidelity, but the big one that’s keeping people away right now actually isn’t monetizing their assets, it’s the Edit interface and the focus on VR. Building in VR is cool but it’s very hard to build a large scale experience in HiFi right now because of the clunkiness.

The good news is that we CAN actually make the features we want if it’s out of the worklist for the company, we just need to know if the current edit.js is going to be expanded upon or not like was brought up a year ago. I also saw something like what David made which he said above on the Projects Tab with people bidding on it all the way back in 2014. I don’t know what came of that because some things changed with the site structure, ect.

As for Materials the big thing is that it just needs to be more streamlined for sharing mats across assets. A simpel drag and drop or even something that isn’t so error prone than the current method like copy paste fields for each materials would be ideal. Right now we have the filepath code and so many times i’ll glance at it and make a mistake. Granted i am legally blind but still i know other people with the same issue.

Thanks for hearing me out guys i’m always happy to post something here because i don’t get flamed for an opinion, everyone is still super productive.


#17

Someone should probably do a quick test if the drag to get file path works from Desktop /Browser -> High Fidelity -> qtWeb (which is used for the edit).

In HTML this unfortunately it isnt possible to get the path via drag, instead it is just the file (encoded), but High Fidelity needs valid paths for the linking, not the files the selves, as there is no upload method other than ATP which would make the assets domain specific only…

Thats the only suggestion on your feedback that makes me question if there is need for client code changes.


#18

I’ve tried out my entities editor and it working reasonably well, especially if you do not have High Fidelity’s edit.js running at the same time (interferes with some things like the Del key and undo/redo). Ctrl-E or Edit > Edit… to open/close dialog.

To help people try it out I’ve provided a modified defaultScripts.js at the following URL: http://ctrlaltstudio.com/downloads/hifi/scripts/defaultScripts.js

  • Removes High Fidelity’s edit.js
  • Adds my edit.js
  • Adds my modified version of inspect.js (use for camera control)
  • Adds my nametags.js script

More info on above scripts: http://ctrlaltstudio.com/hifi


#19

Who say you cannot Temp rezz ? Or rezz basic things. Ill wait until i can see it myself.
Most domains in high fidelity are no rezz to. not see a differene. Also, no rezz have it’s good things to. .sbsar is not option, there are more programs that are used to create materials. Not sure if the all support .sbsar

Building in VR is a nightmare and almost impossible. because you cannot access all the other desktop tool. The did waste to much time on get the edit tool right. Something like that where 10 months earlyer needed. Not in 2018

Drag and drop ? that’s the most clumpsy not practical way todo something.
I vere rare use drag and drop in windows. because i rarly use 'This pc" i use total commander , that is much betetr for file handling. Just a file open dialog is much better solution everybody understand and it works. But not that stupid QT file dialog the still used for ATP i think. ATP is monster that did need update 10 months ago too.


#20

You may not use it, but it is still handy way of setting a texture or a model file for some software without having to figure out the local file path or copy pasting url from a browser. But with any ui shorthand, It however should never be the sole method of doing things. However the same issue also applies for file manager, browsers never transmit the file path, instead its the file it self that is set to the field, which is damn annoying, but understandable protection from Script running / downloading / compromising your pc from the perspective of the web.

On further testing it is only doable between an image displayed in the webbrowser and high fidelity. Files however are not doable due to the security stoppers in browsers (which the edit window is).
ATP needs further investigation however.