Emulated desktop Operating system written in a high level language, compiled to an intermediate assembly language, run on a virtual machine written in JavaScript, executed in a javascript engine, run inside of a sandbox inside of a browser, inside of an operating system just like the proprietary one one being run at the bottom of the stack.
All so you can "create text files" on a proprietary platform that won't work without the internet. What a time to be alive!
Not 100% correct, but ok, you can say that as it stands.
If you inspect the IndexedDB when logged in, you can see that everything is stored locally already. Offline mode was planned from the very beginning. It will and can already work offline if I spare a couple of days on it. But I didn't see it as a priority right now.
Most consumer (hard/soft/firm/wet)ware is pretty much all around bad with very few redeeming qualities these days. You can almost feel the contempt for the user built right into the design from marketing all the way to implementation.
I struggle to understand the use-case. What does this actually offer me over just using a normal web browser with maybe a customised new tab page? I made an account with a throwaway email to play around with it, and I honestly didn't understand why I'd want to use this over whatever workflow I have now.
Came here to say this. I see these come up on HN every year or so. They're often beautiful and commendable for that reason, but I never understand what anyone would actually use them for. Typically I see them start out strong, marking that features are still in Beta, but then it just kind of ends there.
In the abstract, I can feel out there being some kind of use case I'm not thinking of. Maybe something along the lines of having a familiar mental model for organizing the data in the browser, (not the data in the web apps running in the browser). Things like bookmarks, tab groups maybe, browsing history.
But what is shown is more of an honest attempt to create a full OS in the browser which will always be inherently inferior to the operating system the browser is running in, even with advances to browser engines.
Coming back to this some time after my original comment, a few more thoughts in no particular order:
> a modern, desktop-like UI that meets today's needs
What are "today's needs"? Why does the standard desktop interaction paradigm not work with them? Why does "move it into the browser" solve the problem?
> it actually has some useful features to help with your browsing management
I can see some potentially-interesting features in the demo GIFs, but they don't make it clear why I'd want them through a desktop-style interface. Why not as just a browser extension or similar? What does the desktop paradigm provide that's a meaningful enhancement for the end user and not just a tech showcase?
> And then a "revolutionary" product comes out that puts our links & tabs collapsed on the side, with some extra features. Magical, right :)?
It's a bit overdone, sure. I won't disagree with that. But why does it keep happening?
imho it happens because the standard tab setup of "put all the tabs at the top of the window" had obvious shortcomings for a meaningful-enough group of users, and moving it to the side -- be that as a list, a tree, ... -- had benefits that those users appreciated, and that the more-average user may also find benefit in. Tabbed browsing had a significant benefit over the traditional windowed browsing, and imho it seems that moving them to the side has enough of a benefit over the "standard" one-row top-only tabs that people are moving to it.
But what does the desktop paradigm then offer over that? From my quick experimentation, I couldn't even figure out how to get value out of the base UI; this discourages me from then wanting to then try to ex. install your browser extension.
---
Who's the intended user? Is it specifically for desktop-heavy power users? Is it for less-technical users to gain a better interface for browser interaction?
Why floating window management? With a surprising amount of people's interactions with the web moving to mobile (smartphones, tablets), I could see the floating window interactions actually being a drawback. Tiling windows might be better, but again that's potentially very user-dependent.
Jumping off of the floating window question, where is the intended use-case? Is this for desktop users? Are tablets a supported use-case? Smartphones? Where's the line drawn? If you intend to support more than the desktop user, how will you handle the lack of precise touch? The current UI has annoyingly-small icons on desktop; if mobile is an intended use-case I could see this very quickly becoming infuriating.
How is accessibility handled? Is keyboard-only interaction possible? What about screen readers? As a disabled person, I'm pretty tired of seeing user interfaces that are incredibly unfriendly to me because they rely on precise click/tap targets that I struggle immensely to properly use at times.
The product might benefit from having an actual tour / demo / etc. It's very unclear what the true use-case for it is. Every major (desktop) OS already has drag-and-drop, multiple workspaces, floating windows, ... Having some kind of walk-through would probably go a long way towards helping people understand.
---
I don't say these things to discourage you, but this just doesn't feel ready for release yet. The sheer number of comments here on HN expressing negativity about this, at least to me, signal that the product's use-case isn't clear and that it's not obvious why this would be wanted. As a tech person, I very well understand the desire to make cool tech and show it off, but cool tech does not a product make. This reads as a tech demo being dressed up as a product, which explains a lot of the negativity here.
Your comment below[0] makes it a lot more clear that this is less "real serious product" and more "toy/tech demo/similar," and if it had been expressed as such from the start, I think the reception would have been better.
Hopefully these thoughts are of use! I do love seeing experimentation outside of the standard ways of interacting with things, and I think that with more refinement and experimentation you might find something interesting with this idea.
I appreciate your detailed response, so thank you for taking your time to explain.
First, I would like to point you to this page: https://savaos.com/about - I don't know if you've seen it already, but it explains in brief what we want to achieve/envision. Should explain some of your "what the hell is this" questions :) And as you can see, it's a very ambitious goal.
I agree with a lot of your arguments. To be honest, the most challenging part so far has been to try to explain what we're doing. "What would be the right MVP?", "How can we evaluate such a project?", "Is something like this even feasible?".. these kinds of questions constantly popped up in our heads. My point is, we had to start somewhere.
We don't know if this is the right direction yet, but so far we've seen some interest and already received valuable feedback. The best use-case right now is practically link & tab management. Definitely overkill for just that, but for "non-tech" people, it does feels familiar.
I'm not a UX designer, I tried my best with the design and I've personally worked on a lot of accessibility features before (in terms of code). It's not there yet I get that, but it will.
Regarding the comments, on the contrary, I don't see much of negativity here. Just people expressing their gut feeling, but I like it a lot. Much better then hearing the typical non-sense "congrats" & "all the best" kind of feedback. This is what opens my mind the most to make crucial decisions for the next steps.
By the way, I did mention that it's an MVP. In my opinion all MVPs are toys and demos of "what can become". Perhaps I'm too good of a designer for not being an actual designer, and it looks too good? I'll take that as a compliment.
First thing I though was that it looks like my Android phone when plugged into a monitor. So then the question is what do I gain/lose by running in a browser? I would expect the Android versions of the apps (Netflix, Spotify, etc) to be better and more efficient.
The only gain I could think of was the extra level of sandboxing the apps. That might be worth it for some apps, so I can always run those in a browser on my normal OS. For this to work out, it has to offer something else like Chromebooks which was packaged on low priced hardware and worked more like an appliance without PC maintenance. Running a browser OS on normal hardware+OS doesn't make any sense to me.
It was predominantly designed to be as dumb a server as possible and do everything client side. It implemented a basic file IO interface so you can write files and get directory listings.
Anything fancy client side could be leveraged by the ability to execute commands on the server and providing a local named socket for commands to connect to to establish a websocket connection to the client. The core server then plays no further part and it's up to the executed command to send a message to open a window and provide the html/js for the content which will handle all future websocket messages for that window. Essentially all windowed apps with a complex server side component are free to implement their own protocol.
I didn't think it could be a commercial product though. I appreciate that being a 'Product' might afford one the resources to develop such a thing more fully, but then it is just another thing to buy.
I fundamentally believe that things like this have to be free and open source to be viable long term. Otherwise you just have an impressive dead-end like GeoWorks was, back in the day.
For me it makes sense if you're using a bunch of shared devices but want to keep your tabs/projects/files all open in a safe place. It looks like google drive sync is in the works, and I think this would make it much more useful. Outside of a managed enterprise environment, though, I don't really see the advantage.
It looks like the idea is to make it so that every type of app can run everywhere within the browser. Once again, cool, but useless unless these apps optimize for smaller vs larger devices and different storage versions, at which point I don't see the difference from a pwa, unless I'm interpreting this incorrectly? Otherwise every other example on the webpage seems to be really basic like text files, which, as others have pointed out, have many existing apps with robust sync capabilities.
Technically, you can say that an OS is a term for managing complexity. We're managing the complexity that comes with the modern age of the internet & the web platform. At this stage, I agree it might sound like an overstatement, but it's aligned with our vision and where we eventually want to take it.
Right now I rely on TabsOutliner when using Chrome (which I only use for work). It lets me keep 400+ tabs open and stay sane. I like it so much that I've paid for it 3 times, and would have paid a 4th time but it seems you can't anymore, and I fear it's abandoned.
In any case, this is how I work. I use browser tabs as a kind of short- to medium-term memory. I open a bunch of things, and keep them open for as long as I might plausibly need them. To me this is just normal. I don't know how anyone lives with only 10 or 20 tabs open, or a 50 tabs in a single window. How could you remember anything? But without TabsOutliner or something like it, this becomes a sprawling mess, because the browser gives you no native means to search it, or "zoom out".
Unfortunately TabsOutliner isn't available for Firefox, which I use when I have a choice. So seeing SavaOS promote Chrome... I lose a little interest right away. If it doesn't work in Firefox it's not worth getting excited about, because Chrome as a piece of software treats me like an enemy and I don't like that. So: support Firefox!
That said, if SavaOS gives me the capability to organize my tabs, maybe treat them like files I can put in directories etc etc, that sounds awesome and I want to try it. At the very least maybe it's better than TabsOutliner.
I'm not sure why you got so many negative replies. As some other commenter pointed out, it can be a good use case for reusable, copyable desktops. Another use case I though about for some time - if we build an OS (a "traditional one") that consists of just a browser and only necessary stuff for running a browser, your product would be a great fit as well
So a less functional ChromeBook? I'm struggling to find the use case for your use case. Maybe for kids education if you want them to have limited capabilities? But otherwise it is just another OS with way less software support.
As it stands, that's true. However, with the continuous advancements in WebAssembly, and if we open a store where people can install open-source software, a lot of great things could come from that.
But... why not just let people install open source software as is? Stripping down the OS to "just" the browser leaves most of the OS. At that point we're arbitrarily restricting users to apps that run in the browser for no real reason or benefit.
I have not checked out any of this, but I guess the main advantage is that you can log in from anywhere, on any device that has a web browser, and have your full OS?
I've seen this "we need a stripped-down OS just for browsers" idea floating around a lot recently, and I just don't think it holds water. The browser is too big. Stripping down the OS just to accommodate the browser still leaves most of the OS.
Browsers do a lot. You're going to need the same kernel, almost all the same userspace libraries, and many of the same background processes as a fatter OS.
I have given your service a try, and while the UX was impressive, I was bothered by an apparent lack of encryption. The landing page, the privacy policy, and your mission statement all make many mentions of encryption and security, yet (besides TLS) there doesn't seem to be any encryption going on as far as my browser is concerned.
I created a new folder, created a text file in the folder, and added some text to the file. The name of the folder, the text file, and the content itself were sent to the server unencrypted (besides TLS).
I'm sure your server encrypts the data immediately, but this adds unnecessary trust when client-side encryption could be employed. It also enables an attacker or a potential future operator to sniff the data before it is encrypted. That's no good!
This is a really good point and I have thought about this multiple times along the way. Web Crypto API seemed ideal, but it brought its own complexities, especially if you want to have quick access on multiple browsers/devices.
It's true that as it is, it still requires trust. We do have our own custom servers, and we made sure that no logs related to personal data are ever stored, and encryption is done on the application level before it is sent on the DB server.
So what you've done is redefine "application" to mean half of your backend? Meaning your privacy page, which claims "we don't & cannot observe what you are doing", is outright lying? Meanwhile you're here commenting about how "the encryption & anonymity are rock solid" and nothing at all like the "encryption at rest" other services have. This is insanely sketchy.
Personally, I feel like the bold statements about encryption should be removed until this is implemented to avoid misleading users.
Out of curiosity, is the data encrypted with a client-provided secret (eg. a password hash, or something that would otherwise be impossible to extract from the server), or is the secret stored on the server?
I'm not sure I agree about it being a bold statement. Our description is very clear, and our approach is still much safer.
I see hundreds of products slapping "Encryption at rest" to make people believe their data is safe :) Yet, it's accessible by anyone that controls the server...
We also go further into details in the privacy page too.
The data cannot be decrypted without a client-provided secret. We'll make sure to be more transparent regarding all this.
In my opinion it is misleading. Your "privacy by default" section has three headings which claim encryption, and while none of them are false, you can still just log everything your server receives. This is less private than What's App, and it's marketed as an Operating System -- for everything that you do. I think it's worth considering moving the encryption to be done client-side as long as there are no performance concerns.
That's a lot of trust for a project that is proud for having it tested with one browser, and wants to hear from you on Facebook or Discord. My trust in such a project would be quite minimal.
It was tested and should work on all browsers (though the extension is only chromium based at the moment).
But for the sake of simplicity, we placed that badge since it's the most familiar icon to the wider audience. Initially most of our visitors thought it's an OS they have to download, this helped solve some of the doubts.
I had to create Facebook just for this, because whether we like it or not, it's a huge platform that shouldn't be missed.
This is pretty neat! I wish web browsers on Mac could span to 2-3 monitors because then you could really build a great spatial environment for web browsing.
This does remind me a bit of a personal knowledge management solution that has (fairly) recently gone out of beta, AnyType [0], as that too was sometimes referred to as an "operating system for the new internet" and received similarly skeptical reception. Seen them refine the concept into a more rounded project, since the closed alpha, that has been rather successful in the PKM space, though the end product never fully fit my personal use cases.
From previous projects [1] (during which I merely made some minor modifications to a pre-existing FOSS project and took a look at the code of different "web-desktops" codebases), I get how much goes on under the hood of any desktop-like interface running in JS, so seeing one that also integrates handling the tabs of the "host browser" is immensely impressive, though of course I understand those highlighting the inefficiencies such an approach can bring.
I very much appreciate the appearance, especially the subtle use of glass like transparency on window borders and the way third-parties like Soundcloud seem to be embedded.
I will have to agree with some commenters that perhaps the privacy promise should be made more clear as to what the current state is. I understand that implementation takes time, but to build trust, being as transparent as possible from the start is invaluable.
Furthermore, I see potential for this project, perhaps with a bit of tunnel vision due to me being immersed in the space, in the PKM field, as there, despite web-first applications being the norm, there is still a hole that I feel could be filled concerning making open tabs as part of an active project directly accessible and manageable, rather than needing to import web pages, then switch to the application, as is the case with most current solutions like MyMind [2]. Especially, as there must be people other than myself whose PKM does necessitate actively using roughly 100 tabs at a time...
Additionally, if I may, perhaps the potential would have become more clear to some on here if you had linked the learn section [3] in your original post. It doesn't appear accessible on the homepage either, which is a shame, because that gave me a far better idea of the project in its entirety. Especially the search implementation showcased there has potential depending on the execution and should, in my opinion, be something you highlight on the homepage too.
Lastly, I want to mention that you are doing a great job taking the occasionally a bit harsh, but generally fair reaction the HN crowd can have to showcases on the chin. That takes maturity and professionalism, especially on a passion project.
Tangential, but rclone[1] supports this kind of workflow with the `copy` subcommand, if you don't mind spending both download and upload bandwidth at once.
This will definitely be implemented, in fact we mention it on the website as the next major feature. We would also like to support NAS & FTP as well, but most likely start with the major cloud providers.
It reminds me of the initial Dropbox launch. Their pitch was a USB drive for the Internet, which was torn to pieces by this crowd. Then, Dropbox built a video showing how it worked, and their product went viral.
Folks here, myself included, struggle to understand why we need a desktop in our browser. Your animated gifs don't show enough for someone outside of your product to 'get it'. Record a video and walk us through your killer use case, I think it's there, but I don't quite see it yet.
Dropbox was torn to pieces because it was a feature (folder sync) raising money as a company.
It was obvious why it was a useful feature, but it wasn't obvious how they'd build a moat.
This product is unlike Dropbox in that it isn't solving a problem anyone has. I'd go further and say it immediately repelled me because it looks like it would make using my computer even worse.
Was it? I’m looking at the thread right now and the top level comments say things like: “This is genius”; “I must say Dropbox looks great!”; “This has great potential!”; “Cool stuff indeed”; “your app is something that I've been wishing someone would make for some time now”; “This is an interesting application”; “It looks great man”; “Nice Application”; “I like the app”; and on and on… It’s mostly praise and some well reasoned and polite criticisms.
And even if you’re referring to the infamous BrandonM comment, not only is that one comment but the reply (which is almost always glossed over) shows it was a valuable and positive interaction. dang had a nice writeup on it.
Dropbox solves a problem. I'm not sure what this is other than a tech demo. If it was a toy project, sure it's cool. Trying to sell it as a product? That's a reach...
You're right, we're not being clear enough regarding this. In the last section & in the letter we do mention & give hints of open-source, and that's because it will be.
We really hope to have the code & documentation ready soon, but I can't promise any specific timelines at this point.
We'll try to adjust the wording to be more concise
Not the original commenter, but "Own back your data & software" to me reads as if self-hosting is supported, at minimum. There doesn't seem to be an option to download the server software, however.
If you asked me to login with Google and got my email and other info, then don't ask me to also pick a password. Bait & switch and I won't stand for it.
I understand your pain, but this is the best we could do right now to be able to implement the privacy features we mention on the website. Think about it as MFA by default, and Google is a super simple way to pass the first method.
Links & tabs contain a lot of personal data, and I'm sick of hearing "Encryption at rest", while not knowing that all the admins can still see everything.
It sucks I know, but at least the encryption & anonymity are rock solid.
We'll definitely try to find other solutions in the future.. and by the way, passwords don't need to be complicated ;) There's no restrictions
But this is a web app, so the encryption happens in a blob of JavaScript that you can update at any time. Users still need to trust you as much as they would if the data was encrypted server-side. You could maybe claim "application-level" "rock solid" encryption if your app was entirely implemented within the extension, so users could at least pick one version of it to trust, but it doesn't look like that's what you did?
I'm going to give my honest opinion. (No, I haven't given you a login email to try it, it's not that compelling)
We've known how to put labels and little images together to make icons for decades, we're all familiar with the idea. And some windows, I doubt you've added anything new. And text boxes. Your splash image shows windows, with checkboxes in them.
Blatantly useless, so your presentation says, "This is just the familiar stuff, think of what you can do....." And what? What can I do? The browser is already the thing you're copying, in the browser. A "familiar environment"? Your "online home"?
It's a toy, a demo of your hard work, but it does nothing, and your writeup is nothing. I'm glad you had fun. That's it. Sorry.
I ask for nothing else other than honesty. And I thank you for speaking so bluntly about your thoughts.
At this point I think you're right, it is indeed a toy and a demo of my hard work. But it's also available for others to try it, and hopefully together we can turn it into something substantial.
I was thinking about something like this. It would be cool if each app has a discovery service of commands so AI can manage the ecosystem automatically.
That way AI can handle any arbitrary app so long as it has a public interface of discoverable commands. It will make sense of your apps and handle complex interactions between them.
Emulated desktop Operating system written in a high level language, compiled to an intermediate assembly language, run on a virtual machine written in JavaScript, executed in a javascript engine, run inside of a sandbox inside of a browser, inside of an operating system just like the proprietary one one being run at the bottom of the stack.
All so you can "create text files" on a proprietary platform that won't work without the internet. What a time to be alive!
VS Code's stack is approaching this level of abstraction/translation/insanity, but it's still performant and useful.
Not 100% correct, but ok, you can say that as it stands.
If you inspect the IndexedDB when logged in, you can see that everything is stored locally already. Offline mode was planned from the very beginning. It will and can already work offline if I spare a couple of days on it. But I didn't see it as a priority right now.
I'm just being snarky. Props for making something more polished than anything I've ever made.
Please do spare a couple of days on it. I think noone would really believe it until seeing it :)
I mean when you describe most modern consumer software products, they’re pretty much that bloated.
Most consumer (hard/soft/firm/wet)ware is pretty much all around bad with very few redeeming qualities these days. You can almost feel the contempt for the user built right into the design from marketing all the way to implementation.
And most software today requires an internet connection anyways, even if it is installed natively...
as a user, your first paragraph was just jibber jabber.
I struggle to understand the use-case. What does this actually offer me over just using a normal web browser with maybe a customised new tab page? I made an account with a throwaway email to play around with it, and I honestly didn't understand why I'd want to use this over whatever workflow I have now.
Came here to say this. I see these come up on HN every year or so. They're often beautiful and commendable for that reason, but I never understand what anyone would actually use them for. Typically I see them start out strong, marking that features are still in Beta, but then it just kind of ends there.
In the abstract, I can feel out there being some kind of use case I'm not thinking of. Maybe something along the lines of having a familiar mental model for organizing the data in the browser, (not the data in the web apps running in the browser). Things like bookmarks, tab groups maybe, browsing history.
But what is shown is more of an honest attempt to create a full OS in the browser which will always be inherently inferior to the operating system the browser is running in, even with advances to browser engines.
Coming back to this some time after my original comment, a few more thoughts in no particular order:
> a modern, desktop-like UI that meets today's needs
What are "today's needs"? Why does the standard desktop interaction paradigm not work with them? Why does "move it into the browser" solve the problem?
> it actually has some useful features to help with your browsing management
I can see some potentially-interesting features in the demo GIFs, but they don't make it clear why I'd want them through a desktop-style interface. Why not as just a browser extension or similar? What does the desktop paradigm provide that's a meaningful enhancement for the end user and not just a tech showcase?
> And then a "revolutionary" product comes out that puts our links & tabs collapsed on the side, with some extra features. Magical, right :)?
It's a bit overdone, sure. I won't disagree with that. But why does it keep happening?
imho it happens because the standard tab setup of "put all the tabs at the top of the window" had obvious shortcomings for a meaningful-enough group of users, and moving it to the side -- be that as a list, a tree, ... -- had benefits that those users appreciated, and that the more-average user may also find benefit in. Tabbed browsing had a significant benefit over the traditional windowed browsing, and imho it seems that moving them to the side has enough of a benefit over the "standard" one-row top-only tabs that people are moving to it.
But what does the desktop paradigm then offer over that? From my quick experimentation, I couldn't even figure out how to get value out of the base UI; this discourages me from then wanting to then try to ex. install your browser extension.
---
Who's the intended user? Is it specifically for desktop-heavy power users? Is it for less-technical users to gain a better interface for browser interaction?
Why floating window management? With a surprising amount of people's interactions with the web moving to mobile (smartphones, tablets), I could see the floating window interactions actually being a drawback. Tiling windows might be better, but again that's potentially very user-dependent.
Jumping off of the floating window question, where is the intended use-case? Is this for desktop users? Are tablets a supported use-case? Smartphones? Where's the line drawn? If you intend to support more than the desktop user, how will you handle the lack of precise touch? The current UI has annoyingly-small icons on desktop; if mobile is an intended use-case I could see this very quickly becoming infuriating.
How is accessibility handled? Is keyboard-only interaction possible? What about screen readers? As a disabled person, I'm pretty tired of seeing user interfaces that are incredibly unfriendly to me because they rely on precise click/tap targets that I struggle immensely to properly use at times.
The product might benefit from having an actual tour / demo / etc. It's very unclear what the true use-case for it is. Every major (desktop) OS already has drag-and-drop, multiple workspaces, floating windows, ... Having some kind of walk-through would probably go a long way towards helping people understand.
---
I don't say these things to discourage you, but this just doesn't feel ready for release yet. The sheer number of comments here on HN expressing negativity about this, at least to me, signal that the product's use-case isn't clear and that it's not obvious why this would be wanted. As a tech person, I very well understand the desire to make cool tech and show it off, but cool tech does not a product make. This reads as a tech demo being dressed up as a product, which explains a lot of the negativity here.
Your comment below[0] makes it a lot more clear that this is less "real serious product" and more "toy/tech demo/similar," and if it had been expressed as such from the start, I think the reception would have been better.
Hopefully these thoughts are of use! I do love seeing experimentation outside of the standard ways of interacting with things, and I think that with more refinement and experimentation you might find something interesting with this idea.
[0] https://news.ycombinator.com/item?id=41872816
I appreciate your detailed response, so thank you for taking your time to explain.
First, I would like to point you to this page: https://savaos.com/about - I don't know if you've seen it already, but it explains in brief what we want to achieve/envision. Should explain some of your "what the hell is this" questions :) And as you can see, it's a very ambitious goal.
I agree with a lot of your arguments. To be honest, the most challenging part so far has been to try to explain what we're doing. "What would be the right MVP?", "How can we evaluate such a project?", "Is something like this even feasible?".. these kinds of questions constantly popped up in our heads. My point is, we had to start somewhere.
We don't know if this is the right direction yet, but so far we've seen some interest and already received valuable feedback. The best use-case right now is practically link & tab management. Definitely overkill for just that, but for "non-tech" people, it does feels familiar.
I'm not a UX designer, I tried my best with the design and I've personally worked on a lot of accessibility features before (in terms of code). It's not there yet I get that, but it will.
Regarding the comments, on the contrary, I don't see much of negativity here. Just people expressing their gut feeling, but I like it a lot. Much better then hearing the typical non-sense "congrats" & "all the best" kind of feedback. This is what opens my mind the most to make crucial decisions for the next steps.
By the way, I did mention that it's an MVP. In my opinion all MVPs are toys and demos of "what can become". Perhaps I'm too good of a designer for not being an actual designer, and it looks too good? I'll take that as a compliment.
I wonder if I should have an LLM to sum it up...
First thing I though was that it looks like my Android phone when plugged into a monitor. So then the question is what do I gain/lose by running in a browser? I would expect the Android versions of the apps (Netflix, Spotify, etc) to be better and more efficient.
The only gain I could think of was the extra level of sandboxing the apps. That might be worth it for some apps, so I can always run those in a browser on my normal OS. For this to work out, it has to offer something else like Chromebooks which was packaged on low priced hardware and worked more like an appliance without PC maintenance. Running a browser OS on normal hardware+OS doesn't make any sense to me.
I have had a go at building one of these.
https://fingswotidun.com/images/notanos_4.png
It was predominantly designed to be as dumb a server as possible and do everything client side. It implemented a basic file IO interface so you can write files and get directory listings.
Anything fancy client side could be leveraged by the ability to execute commands on the server and providing a local named socket for commands to connect to to establish a websocket connection to the client. The core server then plays no further part and it's up to the executed command to send a message to open a window and provide the html/js for the content which will handle all future websocket messages for that window. Essentially all windowed apps with a complex server side component are free to implement their own protocol.
That simple architecture allowed me to construct some nice interactions https://fingswotidun.com/images/notanos_drag_drop.gif
I didn't think it could be a commercial product though. I appreciate that being a 'Product' might afford one the resources to develop such a thing more fully, but then it is just another thing to buy.
I fundamentally believe that things like this have to be free and open source to be viable long term. Otherwise you just have an impressive dead-end like GeoWorks was, back in the day.
We're definitely going to make this open-source. But the code and especially the documentation is not ready for that yet
For me it makes sense if you're using a bunch of shared devices but want to keep your tabs/projects/files all open in a safe place. It looks like google drive sync is in the works, and I think this would make it much more useful. Outside of a managed enterprise environment, though, I don't really see the advantage.
It looks like the idea is to make it so that every type of app can run everywhere within the browser. Once again, cool, but useless unless these apps optimize for smaller vs larger devices and different storage versions, at which point I don't see the difference from a pwa, unless I'm interpreting this incorrectly? Otherwise every other example on the webpage seems to be really basic like text files, which, as others have pointed out, have many existing apps with robust sync capabilities.
Why is it called Sava OS when it isn't an operating system. Shouldn't it be Sava DE or something?
Technically, you can say that an OS is a term for managing complexity. We're managing the complexity that comes with the modern age of the internet & the web platform. At this stage, I agree it might sound like an overstatement, but it's aligned with our vision and where we eventually want to take it.
I am plausibly a person this is for.
Right now I rely on TabsOutliner when using Chrome (which I only use for work). It lets me keep 400+ tabs open and stay sane. I like it so much that I've paid for it 3 times, and would have paid a 4th time but it seems you can't anymore, and I fear it's abandoned.
In any case, this is how I work. I use browser tabs as a kind of short- to medium-term memory. I open a bunch of things, and keep them open for as long as I might plausibly need them. To me this is just normal. I don't know how anyone lives with only 10 or 20 tabs open, or a 50 tabs in a single window. How could you remember anything? But without TabsOutliner or something like it, this becomes a sprawling mess, because the browser gives you no native means to search it, or "zoom out".
Unfortunately TabsOutliner isn't available for Firefox, which I use when I have a choice. So seeing SavaOS promote Chrome... I lose a little interest right away. If it doesn't work in Firefox it's not worth getting excited about, because Chrome as a piece of software treats me like an enemy and I don't like that. So: support Firefox!
That said, if SavaOS gives me the capability to organize my tabs, maybe treat them like files I can put in directories etc etc, that sounds awesome and I want to try it. At the very least maybe it's better than TabsOutliner.
But seriously: Firefox!
I'm not sure why you got so many negative replies. As some other commenter pointed out, it can be a good use case for reusable, copyable desktops. Another use case I though about for some time - if we build an OS (a "traditional one") that consists of just a browser and only necessary stuff for running a browser, your product would be a great fit as well
So a less functional ChromeBook? I'm struggling to find the use case for your use case. Maybe for kids education if you want them to have limited capabilities? But otherwise it is just another OS with way less software support.
As it stands, that's true. However, with the continuous advancements in WebAssembly, and if we open a store where people can install open-source software, a lot of great things could come from that.
But... why not just let people install open source software as is? Stripping down the OS to "just" the browser leaves most of the OS. At that point we're arbitrarily restricting users to apps that run in the browser for no real reason or benefit.
I have not checked out any of this, but I guess the main advantage is that you can log in from anywhere, on any device that has a web browser, and have your full OS?
I've seen this "we need a stripped-down OS just for browsers" idea floating around a lot recently, and I just don't think it holds water. The browser is too big. Stripping down the OS just to accommodate the browser still leaves most of the OS.
Browsers do a lot. You're going to need the same kernel, almost all the same userspace libraries, and many of the same background processes as a fatter OS.
I have given your service a try, and while the UX was impressive, I was bothered by an apparent lack of encryption. The landing page, the privacy policy, and your mission statement all make many mentions of encryption and security, yet (besides TLS) there doesn't seem to be any encryption going on as far as my browser is concerned.
I created a new folder, created a text file in the folder, and added some text to the file. The name of the folder, the text file, and the content itself were sent to the server unencrypted (besides TLS).
I'm sure your server encrypts the data immediately, but this adds unnecessary trust when client-side encryption could be employed. It also enables an attacker or a potential future operator to sniff the data before it is encrypted. That's no good!
This is a really good point and I have thought about this multiple times along the way. Web Crypto API seemed ideal, but it brought its own complexities, especially if you want to have quick access on multiple browsers/devices.
It's true that as it is, it still requires trust. We do have our own custom servers, and we made sure that no logs related to personal data are ever stored, and encryption is done on the application level before it is sent on the DB server.
This is something I want to see implemented 100%
So what you've done is redefine "application" to mean half of your backend? Meaning your privacy page, which claims "we don't & cannot observe what you are doing", is outright lying? Meanwhile you're here commenting about how "the encryption & anonymity are rock solid" and nothing at all like the "encryption at rest" other services have. This is insanely sketchy.
Personally, I feel like the bold statements about encryption should be removed until this is implemented to avoid misleading users.
Out of curiosity, is the data encrypted with a client-provided secret (eg. a password hash, or something that would otherwise be impossible to extract from the server), or is the secret stored on the server?
I'm not sure I agree about it being a bold statement. Our description is very clear, and our approach is still much safer.
I see hundreds of products slapping "Encryption at rest" to make people believe their data is safe :) Yet, it's accessible by anyone that controls the server...
We also go further into details in the privacy page too.
The data cannot be decrypted without a client-provided secret. We'll make sure to be more transparent regarding all this.
In my opinion it is misleading. Your "privacy by default" section has three headings which claim encryption, and while none of them are false, you can still just log everything your server receives. This is less private than What's App, and it's marketed as an Operating System -- for everything that you do. I think it's worth considering moving the encryption to be done client-side as long as there are no performance concerns.
Curious: is WebCrypto compatibility better now?
That's a lot of trust for a project that is proud for having it tested with one browser, and wants to hear from you on Facebook or Discord. My trust in such a project would be quite minimal.
It was tested and should work on all browsers (though the extension is only chromium based at the moment).
But for the sake of simplicity, we placed that badge since it's the most familiar icon to the wider audience. Initially most of our visitors thought it's an OS they have to download, this helped solve some of the doubts.
I had to create Facebook just for this, because whether we like it or not, it's a huge platform that shouldn't be missed.
This is pretty neat! I wish web browsers on Mac could span to 2-3 monitors because then you could really build a great spatial environment for web browsing.
https://www.youtube.com/watch?v=klTjiXjqHrQ&themeRefresh=1
If you disable the following setting, windows should again span across displays as they used to:
Settings > Desktop & Dock > Displays have separate Spaces
Oo! I need to try this thanks for the tip
Love it! It's like a Chromebook or Samsung DEX in my browser.
"there's ideas, then there's execution".
"there's beauty, then there's usefulness".
Sorry about the flak OP but it's better you hear these comments now than when money's on the line.
Now I can fight with the awful UX of managing windows within the window, too.
Windows 3.1 called and wants its MDI interface back.
Can't argue with that.. best I could do for now :)
I clicked the comments, searched for "MDI", and feel seen. Then I felt old. Bringing back old UX patterns like they are new.
This does remind me a bit of a personal knowledge management solution that has (fairly) recently gone out of beta, AnyType [0], as that too was sometimes referred to as an "operating system for the new internet" and received similarly skeptical reception. Seen them refine the concept into a more rounded project, since the closed alpha, that has been rather successful in the PKM space, though the end product never fully fit my personal use cases.
From previous projects [1] (during which I merely made some minor modifications to a pre-existing FOSS project and took a look at the code of different "web-desktops" codebases), I get how much goes on under the hood of any desktop-like interface running in JS, so seeing one that also integrates handling the tabs of the "host browser" is immensely impressive, though of course I understand those highlighting the inefficiencies such an approach can bring.
I very much appreciate the appearance, especially the subtle use of glass like transparency on window borders and the way third-parties like Soundcloud seem to be embedded.
I will have to agree with some commenters that perhaps the privacy promise should be made more clear as to what the current state is. I understand that implementation takes time, but to build trust, being as transparent as possible from the start is invaluable.
Furthermore, I see potential for this project, perhaps with a bit of tunnel vision due to me being immersed in the space, in the PKM field, as there, despite web-first applications being the norm, there is still a hole that I feel could be filled concerning making open tabs as part of an active project directly accessible and manageable, rather than needing to import web pages, then switch to the application, as is the case with most current solutions like MyMind [2]. Especially, as there must be people other than myself whose PKM does necessitate actively using roughly 100 tabs at a time...
Additionally, if I may, perhaps the potential would have become more clear to some on here if you had linked the learn section [3] in your original post. It doesn't appear accessible on the homepage either, which is a shame, because that gave me a far better idea of the project in its entirety. Especially the search implementation showcased there has potential depending on the execution and should, in my opinion, be something you highlight on the homepage too.
Lastly, I want to mention that you are doing a great job taking the occasionally a bit harsh, but generally fair reaction the HN crowd can have to showcases on the chin. That takes maturity and professionalism, especially on a passion project.
[0] https://anytype.io/
[1] https://ethical-ai.eu/
[2] https://mymind.com/
[3] https://savaos.com/learn
it would be cool. to sava be able open two cloud disk providers side by side and copy files from one to other.
Tangential, but rclone[1] supports this kind of workflow with the `copy` subcommand, if you don't mind spending both download and upload bandwidth at once.
[1]: https://rclone.org/
This will definitely be implemented, in fact we mention it on the website as the next major feature. We would also like to support NAS & FTP as well, but most likely start with the major cloud providers.
Tested on Chrome = Not built by a hacker. Immediate pass.
Everytime i see someone writing "journey" i instantly feel repulsed by a... tech bro VC vibe maybe?
https://ourincrediblejourney.tumblr.com
Wow, difficult crowd.
It reminds me of the initial Dropbox launch. Their pitch was a USB drive for the Internet, which was torn to pieces by this crowd. Then, Dropbox built a video showing how it worked, and their product went viral.
Folks here, myself included, struggle to understand why we need a desktop in our browser. Your animated gifs don't show enough for someone outside of your product to 'get it'. Record a video and walk us through your killer use case, I think it's there, but I don't quite see it yet.
Dropbox was torn to pieces because it was a feature (folder sync) raising money as a company.
It was obvious why it was a useful feature, but it wasn't obvious how they'd build a moat.
This product is unlike Dropbox in that it isn't solving a problem anyone has. I'd go further and say it immediately repelled me because it looks like it would make using my computer even worse.
> which was torn to pieces by this crowd
Was it? I’m looking at the thread right now and the top level comments say things like: “This is genius”; “I must say Dropbox looks great!”; “This has great potential!”; “Cool stuff indeed”; “your app is something that I've been wishing someone would make for some time now”; “This is an interesting application”; “It looks great man”; “Nice Application”; “I like the app”; and on and on… It’s mostly praise and some well reasoned and polite criticisms.
https://news.ycombinator.com/item?id=8863
And even if you’re referring to the infamous BrandonM comment, not only is that one comment but the reply (which is almost always glossed over) shows it was a valuable and positive interaction. dang had a nice writeup on it.
https://news.ycombinator.com/item?id=27067281
Dropbox solves a problem. I'm not sure what this is other than a tech demo. If it was a toy project, sure it's cool. Trying to sell it as a product? That's a reach...
Where's the code? Is Sava OS free software / open source? Your site implies that it is, but I did not see any link to a public repo..
You're right, we're not being clear enough regarding this. In the last section & in the letter we do mention & give hints of open-source, and that's because it will be.
We really hope to have the code & documentation ready soon, but I can't promise any specific timelines at this point.
We'll try to adjust the wording to be more concise
I didn't see anything on the landing page that implied free software / open source.
Not the original commenter, but "Own back your data & software" to me reads as if self-hosting is supported, at minimum. There doesn't seem to be an option to download the server software, however.
But I already have a desktop interface... my browser is running in it.
It just needs an "app" to run a JS-based X86 Emulator for the infinity mirror effect.
:D
If you asked me to login with Google and got my email and other info, then don't ask me to also pick a password. Bait & switch and I won't stand for it.
I understand your pain, but this is the best we could do right now to be able to implement the privacy features we mention on the website. Think about it as MFA by default, and Google is a super simple way to pass the first method.
Links & tabs contain a lot of personal data, and I'm sick of hearing "Encryption at rest", while not knowing that all the admins can still see everything.
It sucks I know, but at least the encryption & anonymity are rock solid.
We'll definitely try to find other solutions in the future.. and by the way, passwords don't need to be complicated ;) There's no restrictions
But this is a web app, so the encryption happens in a blob of JavaScript that you can update at any time. Users still need to trust you as much as they would if the data was encrypted server-side. You could maybe claim "application-level" "rock solid" encryption if your app was entirely implemented within the extension, so users could at least pick one version of it to trust, but it doesn't look like that's what you did?
I'm going to give my honest opinion. (No, I haven't given you a login email to try it, it's not that compelling)
We've known how to put labels and little images together to make icons for decades, we're all familiar with the idea. And some windows, I doubt you've added anything new. And text boxes. Your splash image shows windows, with checkboxes in them.
Blatantly useless, so your presentation says, "This is just the familiar stuff, think of what you can do....." And what? What can I do? The browser is already the thing you're copying, in the browser. A "familiar environment"? Your "online home"?
It's a toy, a demo of your hard work, but it does nothing, and your writeup is nothing. I'm glad you had fun. That's it. Sorry.
I ask for nothing else other than honesty. And I thank you for speaking so bluntly about your thoughts.
At this point I think you're right, it is indeed a toy and a demo of my hard work. But it's also available for others to try it, and hopefully together we can turn it into something substantial.
This is a really good and mature response. He may not see the killer use case yet, but he wants to share it more broadly in case someone does.
I was thinking about something like this. It would be cool if each app has a discovery service of commands so AI can manage the ecosystem automatically.
Not sure I understand what you mean about "each app has a discovery service of commands"
Something like this:
https://xmpp.org/extensions/xep-0030.html
That way AI can handle any arbitrary app so long as it has a public interface of discoverable commands. It will make sense of your apps and handle complex interactions between them.