Convergence: The Hardware-Focused Discussion Thread



  • @wayneoutthere said in Convergence: The Hardware-Focused Discussion Thread:

    It's like this: if you could do this, why would you not:

    Because even with all the things I can theoretically do on my phone, it's still a phone, and will always be far behind a real PC in terms of raw power. There aren't 5GHz CPUs in phones, they don't have 32 cores, 128 GB of RAM, and multiple terabytes of storage capacity.

    It's something I want to replace a personal laptop with when traveling, but it's not something that can replace all my computing needs.

    Why do you even need an external display? I can pop my phone into a folding stand, pull out a folding bluetooth keyboard, and start coding, anywhere. And with the sizes of phones lately, you're practically carrying around a monitor anyway.



  • Hi @dobey -

    I suggested snaps because we'll need desktop apps, and snaps are an easy way to load them. As much as I'd like to see it, I can't imagine we'll get app builders writing apps for everything we need... Maybe some apps will just be for desktop mode.

    And regarding bluetooth: one of the things we struggle with is pairing with sound systems... I was just thinking that if an OEM built something, it could ensure that it works from a physical perspective.



  • To be fair though, @dobey, your computing needs are bigger than a lot of people's if you need 128GB of RAM.



  • @3arn0wl right. Dobey is not the 90% of the population, but absolutely valid. In the Audiocast I mention 'anyone who is not gaming or doing 3D rendering or compiling code'. Those people. Those people can live on a pocket computer in my opinion.

    I agree with @dobey about the 'big screens' though. I don't like them. I liked my HTC wildfire and convergence running on a super small screen would be awesome because it would allow me to have a 'phone sized phone' again instead of these 62" movie theatre phones! we used to mock those big screens and now that's all you can buy 😞



  • Really just depends on what you consider convergence. I'd love to see programs via appimage, flatpak, or even snap that could just be installed on one platform or another. The problem is all these platforms are different and the software developed on them are different as well with different goals. The closest I've found to a convergence system that works is Chrome OS and Android.

    Essentially you would want software that is scale-able regardless the platform. Have a program that you can install on the phone, then turn around and be able to install on the desktop and it scale up. The closest I've seen to something like this is the maui framework under Nitrux.

    This way the program being used is the same regardless the platform and makes everything feel seamless. I've also seen Chrome OS essentially do this with Android applications being run on their desktop OS. The only real difference in running an application on Android and Chrome OS (aside from the underbelly making things work on desktop) is you have a minimize, maximize and close slapped onto the application. It would be amazing for eample to see the open store ported to desktop and all applications just work.

    Hardware to me is kinda the beauty of Linux. 2 or more gigs of ram minimum and for most people you are set. Usage for a wide majority of people anymore consists of paying bills online, word processing, and surfing the web or multimedia.

    As for OS itself the closest I've seen to convergence is Android x86. Unity8 I could see doing the same as long as it's essentially scale-able. As of current it's not really like that. The phone OS and Desktop OS for nearly every major system out there is difference in appearance.

    For me either make it all function and/or look as close to the same on all platforms and slap on a minimize, maximize, and close for the desktop and I'm set. 😄



  • @doniks said in Convergence: The Hardware-Focused Discussion Thread:

    One thing that I find a bit surprising is that "convergence conversations" are usually of a either/or mindset. I want convergence, but I do not necessarily want it instead of another device/setup that I have now, I want it in addition.

    This!

    People thought a slab of glass would replace laptops. We still have laptops and we have a ton of tablets. People thought laptops would replace desktops. We have both. Folks thought cars would replace trucks. We have trucks. We have cars still. Both co-exist.

    I think it is important to observe history when it comes to how mainstream society adopts and adapts to technology and how we interface it...and get our design/product insights from that (not saying let it drive the entire ship, that would be silly) but at least let those things inform our opinions rather than hyper-personalizing how we use devices and trying to infuse that with an entire demographic of users who are not like us.

    When this happens, I think we have a better shot at moving the conversation forward.Thanks @wayneoutthere for posting the article. This is a great discussion.



  • @3arn0wl said in Convergence: The Hardware-Focused Discussion Thread:

    I suggested snaps because we'll need desktop apps, and snaps are an easy way to load them. As much as I'd like to see it, I can't imagine we'll get app builders writing apps for everything we need... Maybe some apps will just be for desktop mode.

    Agreed. Some snaps would not be built with convergence in mind...but some will. This is the hardest part - to get developers to think convergent. I think we will get there. IBM is still pushing out marketing about the benefits of "coding once, run everywhere".

    https://twitter.com/IBM/status/1101629012407922693

    This framework will greatly expedite convergent development because programmers are "lazy" and there is a cost benefit from doing so (more reach, more eyeballs, more eyeballs means more attention and potential for ad revenue). https://www.forbes.com/sites/forbestechcouncil/2018/01/08/developers-are-lazy-and-thats-usually-a-good-thing/#5ddc228a1391

    But I think Canonical had a good idea when they had the standard window size of their "staging" window when using split screen. It wasn't free form. Here is demo of what they came up with before they dropped it.

    https://youtu.be/fiEkCaA_P7c?t=155

    Lo and behold, those dimensions are oddly close to mobile phone dimensions. Was this on accident. No. You can create an app view as well as a multi-task pane for tablet use - perhaps without changing much code.

    P.s. Apple also has this framework for managing an "app" view while in multi-tasking/split window mode and standardizes that as well...this makes it easier for devs to come along and code for that standard. Here is a link to that. https://developer.apple.com/design/human-interface-guidelines/ios/visual-design/adaptivity-and-layout/

    PurismOS goes into this and does a nice job at explaining that this can, and technically has, already been pulled off.
    They describe it on the second half of the article.

    https://puri.sm/posts/converging-on-convergence-pureos-is-convergent-welcome-to-the-future/

    Being linux users in general, our crowd has been the one that championed innovation and forward thinking...after all, even people doubted GUI interfaces wholesale at one point believe it or not. Let's not get too close to that kind of thinking.

    they got gui wrong too

    alt text



  • Just to add, I think if "phone-to-desktop" becomes a thing, we won't have problems finding dock stations anywhere. I'm pretty sure facilities will cope up and we might see docking stations in public places like airport. A big concern with that though is security. We'll need USB condoms 😛



  • @kugiigi Good point about the USB condom thing. The public hardware security issue was lightly touched in a previous episode.
    It's very possible (https://www.geek.com/tech/this-malicious-usb-cable-can-remotely-accept-hacker-commands-1774220/), but the victim base isn't big enough yet. Having public docks could lead to that eventually.



  • @wayneoutthere There are lots of other things for which a phone cannot suffice. And most people do most of those things. This is why I will never see "convergence" as a means for replacing most peoples' hardware with a single device.

    Yes, for plenty of people that may be possible. However, the best part of "convergence" has nothing to do with replacing hardware. It's really about the apps. Being able to use the same apps on any device.

    This is my problem with the whole focus on "convergence" as "plug your device into an external screen." That is not what convergence is about, but it's what so many seem to try and focus on.



  • @kugiigi That won't work if you want more than charging though. You need the video and USB data connections for a dock. So you'll have to trust that your screen and keys aren't being logged.

    So basically, I hope we don't see that happening. 🙂



  • @dobey said in Convergence: The Hardware-Focused Discussion Thread:

    This is my problem with the whole focus on "convergence" as "plug your device into an external screen." That is not what convergence is about, but it's what so many seem to try and focus on.

    I see what you mean. But I think for you that isn't what its all about. For many it is. Canonical did showcase this aspect as a major feature and there is a company who is literally shipping a product that people are buying for this very purpose. But we can agree to disagree.

    Steve Jobs refused to release a bigger iPhone. He claimed it would cannibalize iPad sales. But now looking back we see bigger iphones and bigger iPad's. Both are selling. Nothing was replaced and I don't think anyone claimed converged devices would literally replace all our devices.

    I agree with your point though about keylogging on public docks. That would be a threat to seriously consider.



  • I think @dobey's right about convergence being all about the software, @Profetik777, and the idea of one code for everything is compelling. However it's a conversation about hardware, because it's hardware that sells.



  • @3arn0wl all about the software? Its Both thats all I was saying. You need both. Its up to the user to decide how they will want to use it. Some will be fans for different aspects. After all there are those who buy S9 without dex or using it like a convergent device. Others buy it w dex. Hence qualifying it for him. For him its all about software.



  • @Profetik777 said in Convergence: The Hardware-Focused Discussion Thread:

    @3arn0wl all about the software? Its Both thats all I was saying. You need both. Its up to the user to decide how they will want to use it. Some will be fans for different aspects. After all there are those who buy S9 without dex or using it like a convergent device. Others buy it w dex. Hence qualifying it for him. For him its all about software.

    It's not all about software. But hardware without the software to back it up is completely meaningless. Otherwise you'd all just be running Android, plugging it into a screen, and calling it a day. Because you can just plop on a Linux chroot in all kinds of ways, and run an Xorg session to get mate/xfce/whatever running on the external screen.

    The hardware is merely some way to sell the software. Because software needs some place to run, and hardware needs something to be running on it. The whole point of convergence is to blur the line, not to keep harping on about replacing other devices or whatever. And it's not about bringing traditional Linux distributions to phones. It's more about bringing many of the ideas of phone/tablet applications, security, power management, etc… to traditional PCs, and providing the same experience on all of your phones/tablets/PCs, so you can use the same apps on both, and for the people who want to use only a phone/tablet with an external screen/keyboard as their only "computer" if they think it's suitable, they can do that. But that itself is very secondary.

    If you're talking about Canonical wanting to sell Ubuntu 4 Android (and the Ubuntu Edge), that is much closer to Samsung DeX/MaruOS, than it is to what UT is today.



  • @dobey thanks for sharing. Not sure where I or anyone made the point you are attempting to counter with but good reminders overall and in general.



  • @wayneoutthere said in Convergence: The Hardware-Focused Discussion Thread:

    Is it better to have just one device?

    Probably not. It depends on one's needs. Most people will probably need a main machine for doing other things. There's also the issue of security and compliance to consider, in workplaces. To be in compliance with various regulations, many companies have hard restrictions on what hardware employees can use (or which they can connect to the company WiFi). There's a very wide range of uses, and while a few people may be able to suffice with only a phone in their lives, with an external screen/keyboard/mouse to use sometimes, in my experience, most can't.

    What kind of hardware is best?

    That's a very vague and subjective question. Best for what? And is it limited to only devices which are currently supported by UT? The most complete experience with a UT device on an external screen today, will probably be had with either one of the M10 tablets, or the Nexus 4. The device with the most powerful hardware, is the Pro 5.

    Does the market want this?

    Probably not? But it's not answerable when the market doesn't already have it. How can people know whether they want something or not, when they have no idea what it is? If you judge this based on say, sales of Samsung DeX devices, versus all the devices which can't work this way, the the answer is surely no. But, we get to define our own market, so it's a pretty irrelevant question. Also, we're not making any money off it, so whether others want it or not is irrelevant. We're building it for ourselves and letting others use it for free if they want.

    What kind of creative set up environments can we dream up?

    There's literally no context for this. What does it mean? How people would use their UT devices in the wild if full convergence was capable today?

    Personally, I would never plug my phone into an external monitor, as a means to replace a laptop/PC. Only to stream video or such. I would also, as a matter of security, only want to send a stream from a specific app to the external display. I would however, want to have my phone and main PC "connected" to each other so I could reply to SMS or be notified of incoming calls, from my PC if I choose. And I would carry the smallest folding bluetooth keyboard I could get, that was still usable, with a stand, so I could type complicated things while sitting at the pub or on a plane.



  • @dobey I want to argue with Dobey but I like what he's writing here. 🙂

    Especially the part about software. I actually... never thought much about this because up until recently I didn't know anything about coding. Now that he mentions it how awesome would it be for a horrible programmer like me to sweat for 1 year on some simple app and then magically have it usable on every platform. That is, truly awesome. And the amount of dollars saved for a software company to be able to do that would be impossible to record. This is truly a significant consideration and definitely a worth cause with or without the hardware to consider.

    Imagine even a 'tech support' environment where 99% of the learning was done and then you just have slight hardware related 'tweaks' to your base of understanding? This would be massive for any kind of IT support environment whether a third party company or an internal IT dept. Wow.

    And then comes in snaps and flatpaks. Amazing. Deploy a flatpak/snap and have it roll out to phones, tablets too? Dang. Amazing.

    So, I have to say 'thanks' to Mr. @dobey. This is huge eye opener thought for me and not sure how I didn't see this before ....



  • I agree with Rodney about convergence on the software-side. It will benefit both developers and users and I think it would be really awesome. But I guess this thread wants to focus on the "phone-to-desktop" side of convergence? 🙂



  • I was just listening to the Linux Action News podcast that was linked this morning, because UBports was featured - the two guys seemed fairly skeptical about converging apps...

    Are they more difficult to build?
    Are developers finding it tricky to get the design right for both touch and mouse?


Log in to reply