Provide different "editions" of the Steam client, for different users' needs.

[Quint the Alligator Snapper]

The big conflagration over the new library UI, particularly how it takes out some features and forces others onto basically everyone (resulting in an endless stream of complaints on the forum and using a variety of workarounds to undo said new UI), made me come up with a new approach:

What if Steam provided different "editions" of its client program to users? All would come with the software authentication function, of course, but beyond that, different users want different features, so it might make sense to simply let users choose from a palette of different setups.

One idea:

* "Steam Universe", which is basically the new library UI, and it's fully-featured with everything, and it looks really pretty.
* "Steam Slim", which is basically the old client UI, with the specific stipulation that stuff doesn't use webpage rendering except the web browser.
* "Steam Bones", which is specifically JUST a command-line authenticator and multiplayer server connection, and lacks many fancier features, even the Overlay.

Or something like that. Or maybe Steam Bones could be a subfeature of Steam Slim, for example. But you get the idea.

Another idea:

Stuff could be built modularly. Steam Bones (or whatever the name would be) is the bare minimum, but then you can add the following to it, as you like:
* Overlay (with built-in web browser and screenshots shortcut)
* Broadcaster (to stream your games)
* List View
* Detail View
* Small Mode
* Grid View
* Collections Editor
* Advanced Library Filters (using tags from the store)
* Big Picture Mode
* Friends integration
* Steam Music Player
* Steam Skin Editor
and more.

TL;DR the point is to let users download different packages that contain different UI featuresets beyond Steam's core functionality.

[[Ӻ.ɧ] No.1 The Larch]

โพสต์ดั้งเดิมโดย Rebel Fleet Pursuit:

Modular, eh? Might be difficult and time consuming to code, but I would be surprised if a feature like that wasn't met with open praise.

Poppycock, Valve and Steam *should* have some of the greatest programmers in the industry and this is as simple as custom toolbars and layouts in office. If they couldn't manage this it would be pathetic. Internal office politics, IE the designers of the new layout REFUSING to admit anything wrong with it, is the only thing that would prevent this from happening.

If it really is difficult and super time consuming I would fear for the Steam platform as a whole.

[RiO]

โพสต์ดั้งเดิมโดย Gambit-3:

The thing I'm wondering is if steam just made an old UI skin in the new launcher, wouldn't that still contain all the performance issues that people say they are having.

Yes and no.
Yes, because it would all still be based on Chromium which is inherently kind of a resource guzzler.
But also - and profoundly - no. Because some of the most egregious resource-hungry stuff like the cover images used for shelves; the content in the activity feed; etc. would be ousted.

Also, people have been having issues other than just the performance issues with the new library UI. Don't forget that!

There's the fact that some are experiencing medical problems with the shelves triggering migraine or epilepsy (and yes; in particular when in motion, complex high-contrast color patterns can do that).

There's the fact that the friend activity feed can't be disabled like the community feed can, and shows spoiler-heavy achievements that ruin the experience for many story-heavy single-player titles.

Etc.

โพสต์ดั้งเดิมโดย Gambit-3:

Isn't Valves goal to completely remove all traces of the old UI (is it gdi?) over time, wouldn't it save client resources if they were able to remove the engine that runs the older parts of the client?

It's actually VGUI (Valve Graphical User Interface, iirc) - Valve's own UI framework that was originally designed and afaik is still in active use for Half-Life 2.

Removing the engine that renders the UI would be a cost reduction for the end-consumer in the order of a few megabytes. A literal drop in an ocean of storage space. The core parts of Steam that the UI integrates with should in fact already be modular enough to power both the old and new UI. They really have to be, because they also have to be used by the Big Picture UI. (And infact; there's a commandline headless interface specific for developers as well.)

The real burden here is not maintaining the old UI; it's bolting everything onto the new Library UI. Rather than eat the cost of maintaining a modular design, Valve is breaking the existing clean separation and cutting corners to only implement functioning changes in the web UI.

This is evidenced among others by uninstallation being broken in Small Mode when run with -no-browser, whereas all the uninstall logic should really already be modular enough to be kick-started from the Big Picture mode too.

Somewhere Valve incorporated an explicit hook into the new Chromium based UI for the library into the logic flow, which breaks things.

[[Ӻ.ɧ] No.1 The Larch]

โพสต์ดั้งเดิมโดย Doom:

โพสต์ดั้งเดิมโดย Rebel Fleet Pursuit:

Modular, eh? Might be difficult and time consuming to code,

They already have a framework for that in place. Big Picture uses that. The old UI can be added as "classic mode" in the same fashion.

Exactly, This.

[Baltist]

โพสต์ดั้งเดิมโดย Gambit-3:

The thing I'm wondering is if steam just made an old UI skin in the new launcher, wouldn't that still contain all the performance issues that people say they are having. Isn't Valves goal to completely remove all traces of the old UI (is it gdi?) over time, wouldn't it save client resources if they were able to remove the engine that runs the older parts of the client?

No actually. A lot of the problems are inherent to the new library's UI over reliance on the browser to show information (from a constant stream of news to game box art for the tiles) whereas the older UIs didn't.

[Gambit-3k]

โพสต์ดั้งเดิมโดย Quint the Alligator Snapper:

Multiple clients, multiple versions, multiple editions, whatever makes the most sense implementation-wise to get this result. Any disagreements over these labels at this point are essentially just semantic arguments over the choice of name.

I don't know, I feel like having multiple clients is way different than one client with multiple modes or UI styles (views). A minimal client could be stripped down to contain only what it needs for that specific version, while multiple views or UI's may have to load more dll's resources in order to easily switch between both versions. I don't know how they handle this with BPM, and BPM does seem to take a while to load up for me. But I do see that small mode still uses the same resources as the full large mode for instance. So if Valve made a minimal bones version as the OP asks, then that could and would use much less resources. I have also wished for a minimal client and have posts saying It would be something Valve would get a lot of credit for if they did it. While I have programmed a few apps in my life, I can admit I'm very much not an expert on this stuff though.

โพสต์ดั้งเดิมโดย Quint the Alligator Snapper:

I don't get why Kusa keeps saying that Steam hasn't ever had two separate clients. This isn't even the point of the thread.

Frankly, Kusa seems to be upset that other people have complaints about something Kusa is enjoying.

The OP asks very specifically for multiple client versions, and Kusa is saying they have never done that. I don't personally think that means you can't ask, but that is what he is saying from my observation. I just think in his opinion, you are wasting your time by doing so.

I know all these UI threads kind of bleed together. I have also been guilty of that, but it seems like asking for a separate UI inside of the main client would be more for the feedback or suggestion threads. It would make the people who's complaints don't have anything to do with performance happy though.

[Gambit-3k]

โพสต์ดั้งเดิมโดย RiO:

Also, people have been having issues other than just the performance issues with the new library UI. Don't forget that!

I believe the OP is asking for different versions like a bones version for performance reasons though. I know some people just don't like the UI at all (for many different reasons), but I guess my thought is that if they just made it look like the old UI, some people would not be satisfied with that, at least based on what they say.

The rest of your post is very interesting. I do notice that small mode still takes more ram than the old UI even though it's not rendering cover images, but I'm sure you would say that could be fixed with some code, I'm willing to take your word on most of that stuff since I don't have the technical knowledge to discuss it in detail.

[Gambit-3k]

โพสต์ดั้งเดิมโดย Baltist:

โพสต์ดั้งเดิมโดย Gambit-3:

The thing I'm wondering is if steam just made an old UI skin in the new launcher, wouldn't that still contain all the performance issues that people say they are having. Isn't Valves goal to completely remove all traces of the old UI (is it gdi?) over time, wouldn't it save client resources if they were able to remove the engine that runs the older parts of the client?

No actually. A lot of the problems are inherent to the new library's UI over reliance on the browser to show information (from a constant stream of news to game box art for the tiles) whereas the older UIs didn't.

Small mode takes just as many resources for me as large mode. Which in my opinion isn't very much, but other users are saying it's way to much for them and causing performance issues. If they could just put it in small mode and not have performance issues I think they would. So I'm thinking if small mode doesn't really reduce resource use to levels of the old client, how would a legacy mode be any different.

[Quint the Alligator Snapper]

I should point out that the idea of a modular or multi-edition client suite is meant to serve the needs of BOTH those people who like the new UI AND those people who don't.

So Kusa's statement of "I'm not having problems with the new UI" is just fine; people who like the new UI ought to be able to keep using it.

Meanwhile, people who don't ought to be able to change it to something else that still has those same basic functions, like the ability to authenticate the user, the ability to install and uninstall and launch games, and so on -- just without the many bells and whistles, for example.

[Nx Machina]

โพสต์ดั้งเดิมโดย Quint the Alligator Snapper:

I don't get why Kusa keeps saying that Steam hasn't ever had two separate clients. This isn't even the point of the thread.

Because it never has.

You did suggest multiple versions (clients) in this thread.

โพสต์ดั้งเดิมโดย Quint the Alligator Snapper:

1. Having multiple versions =/= reverting the UI, except for individual users as they choose. Having multiple versions = users who like this new UI get to keep it while users who don't like it can choose something else.

2. If the new UI doesn't function as well for someone as the old UI, then reverting it will be an upgrade for them.

But more importantly your thread title which I am answering.

Provide different "editions" of the Steam client, for different users' needs.

[Quint the Alligator Snapper]

โพสต์ดั้งเดิมโดย RiO:

โพสต์ดั้งเดิมโดย Gambit-3:

The thing I'm wondering is if steam just made an old UI skin in the new launcher, wouldn't that still contain all the performance issues that people say they are having.

Yes and no.
Yes, because it would all still be based on Chromium which is inherently kind of a resource guzzler.
But also - and profoundly - no. Because some of the most egregious resource-hungry stuff like the cover images used for shelves; the content in the activity feed; etc. would be ousted.

Yeah, basically this. I'd expect it to still feel bloated and slow but probably with significant performance boosts due to simple things like not loading hundreds of images.

โพสต์ดั้งเดิมโดย RiO:

The real burden here is not maintaining the old UI; it's bolting everything onto the new Library UI. Rather than eat the cost of maintaining a modular design, Valve is breaking the existing clean separation and cutting corners to only implement functioning changes in the web UI.

This is evidenced among others by uninstallation being broken in Small Mode when run with -no-browser, whereas all the uninstall logic should really already be modular enough to be kick-started from the Big Picture mode too.

Somewhere Valve incorporated an explicit hook into the new Chromium based UI for the library into the logic flow, which breaks things.

Seriously, why the flying frack did they have to do that.

[Quint the Alligator Snapper]

โพสต์ดั้งเดิมโดย Kusa:

โพสต์ดั้งเดิมโดย Quint the Alligator Snapper:

I don't get why Kusa keeps saying that Steam hasn't ever had two separate clients. This isn't even the point of the thread.

Because it never has.

You did suggest multiple versions (clients) in this thread.

โพสต์ดั้งเดิมโดย Quint the Alligator Snapper:

1. Having multiple versions =/= reverting the UI, except for individual users as they choose. Having multiple versions = users who like this new UI get to keep it while users who don't like it can choose something else.

2. If the new UI doesn't function as well for someone as the old UI, then reverting it will be an upgrade for them.

But more importantly your thread title which I am answering.

Provide different "editions" of the Steam client, for different users' needs.

You're talking about what Steam has been like in the past; I'm suggesting what Steam could be like in the future.

[obliviondoll]

Making the Steam client modular enough that independent components can be downloaded and installed separately is a great idea.

Would also encourage more advanced modding options to allow even greater variety in user experience without Valve having to be the ones doing all the work.

โพสต์ดั้งเดิมโดย Kusa:

โพสต์ดั้งเดิมโดย Quint the Alligator Snapper:

I don't get why Kusa keeps saying that Steam hasn't ever had two separate clients. This isn't even the point of the thread.

Because it never has.

Except... like... literally every time there's been a beta test of a major client rework. But other than the multiple times there have been two separate clients, sure.

(also you're still wrong about that being directly relevant to the topic, as has already been covered)

[Trev]

The best idea ive ever seen on here. +1 from me for sure.

[sku]

โพสต์ดั้งเดิมโดย Kusa:

There is not an endless stream of complaints. There are some who do not like the new ui who feel the need to express an opinion and want what they want to the exclusion of all others including Steam. No other game client has multiple versions, they have one unified version.

There are options out there if you are so intent on disliking the new Steam ui. There is Gog Galaxy 2.0 currently in closed beta which I am testing and so far is a very good alternative to other clients although those other clients still need to be installed.

it's pretty endless bub

[Barking_Death_Squirrel]

What I want is the nice, stable, minimalist UI that existed before this should-not-have-been-released-from-beta pile of resource eating fecal matter was forced upon us.
That is all I want. The "special kids" that love it can keep it.