Given Microsoft's recent moves towards a freemium desktop operating system, with its numerous privacy concerns and baked in advertising, it's great to see an actively developed alternative for those who make use of software originally developed for Windows. Lock-in is a terrible thing.
Oops, yea, I really meant the whole Windows-as-a-service direction. I could see Microsoft extending the free offer, though, since they appear to be aggressively pushing people to enroll in the service.
For most end consumers who didn't realize that they paid for a windows license when they bought their computer, it's just a free upgrade, no different than their iPhone updating.
Indeed, and I'm definitely not complaining that Win8 SP2 is a free upgrade from Win7+ ;) Still, only holders of existing Windows licenses qualify for it - "Windows for free" is something different, IMNSHO. But you are quite correct that an average consumer would see this distinction as pointless nitpicking.
Considering that people have been complaining about Windows lock-in for 2 decades, I can't really see the point. Chasing Windows compatibility is not the answer. Getting software to run on other platforms and reducing Windows dominance is more important. Make it easy for people to switch platforms so companies need to listen to consumers.
> Chasing Windows compatibility is not the answer. [...] reducing Windows dominance is more important. Make it easy for people to switch platforms so companies need to listen to consumers.
Agreed, but not everybody's seen the light, aka POSIX.
I also view ReactOS as a way to subvert Microsoft's control over users and developers who enjoy aspects of Windows-like operating systems, such as the UI or the APIs.
I think it could be a really interesting model for folks who need to run legacy Windows software in the future, as well. Instead of keeping a few XP machines around (lack of security patches and all) why not just spin up a lightweight ReactOS VM for just that one application?
Think of all the weird desktop apps, custom databases, etc. that small businesses, schools, and other non-technical orgs have baked into their workflow. Yes, eventually we'll have wonderful legacy-free versions of all of these available on the web but in the meantime wouldn't it be nice to give them some outlet to use them w/o paying an MS tax or dealing with the infinite upgrade treadmill?
I also like to think that some of the ideas implemented in this project might make it's way into things like Wine as a major side benefit. I agree this is a very important project.
ReactOS is a cool idea. Use WINE to create a Windows compatible FOSS Operating System. They aim for XP/2003 compatibility so it doesn't run a lot of newer stuff. But it is good for legacy Windows apps. With the 0.4.0 release they added more driver support, Wifi support, Audio support, etc. They need more people to work on it, and I hope that Google funds them some money as well to pay people.
So far it is in the alpha stage and needs to go to beta stage soon. It only has a 48M overhead and loads faster than the real Windows. It has EXT file system support as well as FAT support but NTFS is read only.
Someone can make a ReactOS bundle with PHP, MySQL, Apache2 so it can be used as a web server. If someone ports CLRNET or the Dotnet open source programs over to it, it can run ASP.Net stuff as well.
You're severely underestimating the massive amount of resources needed to accomplish such target, especially when compared to what the target it.
There isn't much point in throwing a massive amount of resources for building an AMP hosting environment. If somebody needs to start an AMP server from scratch and he really needs to save money, there is LAMP.
The only realistic use case of ReactOS is large-scale deployments of machines which need Windows (driver) compatibility, in which case ReactOS would save large amounts of money.
Security is another major problem with this OS. They are proud to have produce 9M LoC, but they don't advertise how they assure this code is not full of vulnerabilities.
Do they audit the code ? Do they use static analysis tools ? Do they have specifics security practices and processes ?
> The problem with ReactOS is its lack of automatic or semi-automatic OS update.
Among many other reasons why you shouldn't use it as your main OS. But not even the React Devs would suggest that you do, so I'm not sure what your point is. It's clearly still in the development stage and is almost certainly just going to be run in a VM for the time being anyway.
In a recursive fashion, I had to scroll past all of the comments against hijacked scrolling to get to the first comment about the subject matter. This is not the first time on HN I have had to do this. Maybe a single comment, without all of the piled on comments would lessen this distraction? My intent is serious, and I hope not taken the wrong way, but the irony is there.
Yes, I also have to do a 'double-take' on these names. I know it's not the point of your comment, but ReactOS started first for those who don't know already.
Why doesn't Google make a free, compatible Windows clone, for the same reasons Microsoft as well as Google made free, compatible browser clones? I honestly don't understand it.
Why would they? It would cost billions, take years and mostly consist of dead-end features which Windows keeps around for compatibility. The end result would be an almost worthless product which always trails behind Windows and which MS can dominate with their existing software. It would also massively increase the market for "real" MS Windows.
Don't extend your competitors platform. Chrome took IE users and made them Chrome users because Google's platform is the web, not the desktop. So Chrome has been pushing forward traditional desktop features such as offline, OpenGL, and sockets, to displace desktop software.
Look at what Apple has done with the iPad, they've been the market leader and MS has had to follow them, only recently shipping Office apps for iOS. They're blazing ahead with their own platform which serves their goals, not tired old legacy software from 30 years ago.
Because that's an enourmous amount of work and likely to attract hostile legal action? They didn't even create their own operating system for their phones.
Why then does MS work on Bing? Bing is more likely to infringe on Google's patents than a Windows clone is likely to infringe on MS's patents, not? And Bing is certainly lots of work.
Hardly a replacement for Office. They're good to collaborate and work with people over long distances, but when documents grow to not so big sizes, problems begin. I even have problems to load 30 page documents. Word 2016 flies on Windows 8 and 10, and it takes 60-100 MiB of RAM.
Also, there are countless document templates (which plague different work fields) that are utterly incompatible with Google Docs and break all over.
I suspect that Google's view is that the local Operating System is decreasing in importance over time.
They might argue that we're spending more time in web applications, rather than purely in native desktop apps - This is reflected in their Chromebook offering, which largely strips away the non-web experience.
ReactOS has been going at this problem for a loooong time. It's a pretty amazing initiative, and the fact that it's taken so long goes to show how insane a kludge Windows is under the surface.
I wonder if the reason we haven't seen more work on ReactOS compared to Linux development is the commercial backing that Linux development receives from the big guys (Red Hat, IBM, et. al.) I wonder if people are wearisome of funding ReactOS beacuse of Microsoft.
While I'd love to throw poop at Microsoft, I think this is correct. If someone had to create an operating system that accepted GNU/Linux binaries without the benefit of looking at internal communication or the source code, I bet it would take many years as well.
Accepting binaries is relatively easy, you can often stub features and have the thing mostly work. See Wine project.
Accepting drivers is super hard and this is the main part of ReactOS - the kernel.
Based on what I see from the development blogs, I see that most of the effort is just fighting against the huge complexity of building an O/S, more than the quirks/bugs of Windows itself.
You can see very often the pattern of "X was implemented using the library from Y, which works but not well" or "Y was implemented and working for Z case, but it's effectively broken" and so on.
I've seen comparatively little complaints of the kind "bl00dy Windows performs this functionality in an awkard and messy way".
I don't really have much faith in ReactOS, although I would really love to see it mature.
Following its evolution with the time, I can see clearly that it's a type of product which is doomed to be "90% functional" for its entire life (unless somebody will throw a massive amount of money at it).
While you could consider Wine to have the same fate, there is a distinction in the fact that Wine is more directed to a per-case basis, which makes the "90% functionality" state acceptable.
If you want to talk about bad scrolling experiences, then I give you Hacker News. Half the time when there's a submission about some deeply technical topic... before seeing any discussion of the actual topic, you have to scroll through 3 or 4 pages of debate about the ergonomics of the font kerning on the blog post or something.
This is a group of C/C++ hackers who have spent the past 18 years re-implementing Windows 95. The fact that there's any CSS at all on their website is a minor miracle.
I don't know how strongly I feel about non-responsive websites displaying a fixed header bar. However, I'm pretty sure that most people who do feel strongly about it either way aren't going to be using or hacking on ReactOS.
I have this reliable bookmarklet which makes comments collapsible, however it stopped working once I got a downvote arrow: http://pastebin.com/raw/iMkHgmNZ
I hate to bikeshed, but the amount of wasted vertical screen space [1] makes me almost a little angry. Web designers: if you think it's a good idea to reimplement basic browser functionality (scrolling), or to emulate the UX of HTML frames, please think again.
So many mobile sites implement this same unpleasant UI control where the header follows your scrolling and stubbornly takes a portion of the screen without adding any useful functionality, that I wonder if people know it's possible to scroll to the top of any page and see the header there...
<rant>I also object to menus that hide when scrolling up and show when scrolling down. I usually place the row I am reading at the top of the window to make it easier to follow visually. When I need to scroll back a little, the pesky menu comes out instead of the text. I think scroll up should be the exact inverse of scroll down, not a menu display trigger. Or, at least, offer us a few lines back scroll before showing the damn menu.
I tried to explain the retardedness of this to management, before it was implemented. They were like "but people share more when the widgets are always visible, we A/B tested this!" It's like talking to a rock sometimes.
That's like saying "but people click the ads more when they're obscuring most of the page or right next to the download button." They certainly do. Whether they intended to, however, is a different matter.
I would much rather see an implementation of the Web where by default websites need to ask for your permission to load and execute scripts. This would prompt the creation of a widely accessible "reach web" which would never break screen readers or URLs, wouldn't invade your privacy, and could be accessed with a text-only browser. Then there would also be the "rich web" - parts of sites which behave like an app and consequently need user approval to run, rather like you need to "approve" an actual app by downloading it. Many challenges to consider but I think this would solve a lot of problems.
Actually, I was oversimplifying when I said I have JavaScript disabled. I’m actually using JS Blocker¹, which gives me your first sentence². Of course, the end user using a browser extension to gain that functionality instead of it simply being a part of the Web does nothing to actually encourage the creation of a “reach web”.
Every time I see comments on how a site tries to override the browser functionality, I'm once again reminded of why I don't have JavaScript on by default. I don't notice these things at all as a result, and this site is as pleasant to read as any other.
Agreed. I tend to scroll in small increments (maybe every other line of text) in an attempt to keep the content I am looking in the same position relative to my eye. This site's scrolling kept behaving in a way that was inconsistent with my expectations, and forced me to actively pay attention to what should have been an intuitive interaction. Irritating and distracting.
Google's Blogger does the same sort of thing, and it really annoys me. Swerve left or right as you're scrolling through the page and it "helpfully" takes you to a different article.
That is so infuriating, especially for code blocks with no word wrap! Oh you want to scroll to read the rest of the line? How about this random article instead?! Pfft.
My mouse is currently with the scrollwheel having some issues.
On this site the result was a total disaster: any attempt of scrolling made the site jump around randomly, and immediately stop right back at the start again, making scrolling with the mouse impossible.
I seem to remember blinking text, animated GIFs, slow-loading Flash and Java applets back in 1995, although I'm not sure if things were ever as bad as it is now. Then I guess around 2005 web design was generally quite good. And lately it has become pretty awful because many people obviously aren't able to restrain themselves with Javascript.
From their webpage: "Nearly ten years ago the ReactOS Project released version 0.3.0. Today we are proud to announce the formal release of version 0.4.0. ".
This is the problem. Going from 0.3 to 0.4 in TEN years means that you cannot be commercially successful.
As much as I admire ReactOS' contributors and their efforts, I have little faith in their ability to make a dent in the market.
Linux went from 2.6.0 to 2.6.39 in 8 years. That's not even 0.1 change! It can't be commercially successful...
</sarcasm> Version numbers don't mean anything out of context. Is there any specific issue you wanted to point out that took them too long to implement?
This doesn't apply to Reactos though. The api is "whatever windows is doing". Unless they remove something supported by windows N-1 and instead implement replacement interface from windows N, there's not going to be a "breaking" release. But as far as I know the API target is not going to change for a while - it's 2003 kernel / 8.1 user.
