When will 0.3.15 release?
Moderator: Moderator Team
-
- Posts: 128
- Joined: Sat Dec 22, 2012 7:36 am
Re: When will 0.3.15 release?
At what point will ReactOS be sufficiently complete to warrant a 1.0 designation?
Re: When will 0.3.15 release?
Hi Aenas!Aeneas wrote:A question on the matter:
Have we got any expectations? Like, "release till June" or "release within 2013" or "the day after tomorrow"? I mean, anything relating to time and not just to "we have to do X, we have to do Y" (as that is not the question)? - If not, is there an idea when this will be known?
(Yes, I want another release; I know there are nightly builds, but for all I hear, they are of worse quality than a release.)
Well..My personal bet, but keep this in sssh-ecret...is that during the next ReactOS Meeting(next week) 0.3.15 will be one of the main points of the meeting. So I'd wait till next week to know the opinion of all the ReactOS Devs.
My particular opinion: 0.3.15 is closer than we think. Arguments:
1)The level of stability at MM is more or less the same as 0.3.14(I'd even say a little bit better, less crashes due to MM bugs).
2)The level of software compatibility has increased a lot. Foxit, Openoffice, SoftmakerOffice and other big apps are beginning to be testables.
3)Some of the devs wants to release.
4)ReactOS is going to be soon in another Event, and I'm sure we want to have an Official Release before that date.
5)But...
.....We need Testers to find regressions before releasing. One whole year without a release has this counterback: We need to detect all the regressions or the Release quality will be severed compromised. There are always regressions and we have to fix them before a release is done.
And sadly this 5) point is going to be a serious bottleneck. The later we begin trying to find regressions, the later the release will take place.
That's the reason we began asking Officially to test GoldenApps in order to detect asap regressions that could stop the release.
Since almost a week there is a post called "[Official]Testing call" in order to test just 30 apps (10 guys->30 apps: 3 apps each one->30 minutes of testing).
After a whole week, just 4 guys are cooperating. Thanks Fulup,milon, Black_fox, oswetto, for your effort,time and help
And this is really a pity because my idea behind that Official call was to have arguments to push a Release during the March Meeting which will take place next week.
Releasing is not just a "Developing"(coding) process, but it includes the cooperation of the ReactOS Community.
In other words: without Community testing, we can release tomorrow but the release is going to be a piece of s....t. No doubts.
Re: When will 0.3.15 release?
Well, that's a good question. That's up to the dev team to decide, but I would hope for most of the following:BlackRabbit wrote:At what point will ReactOS be sufficiently complete to warrant a 1.0 designation?
Runs the vast majority of software that will run on NT 5.
Supports SATA, USB, Wifi, and printing
Has an explorer that doesn't suck
Is highly unlikely to corrupt your data
Is not frustratingly crash-prone
Today entirely the maniac there is no excuse with the article. Get free BeOS, DOS, OS/2, and Windows games at RGB Classic Games.
Re: When will 0.3.15 release?
I'll do some testing this weekend. I spent all weekend and Monday and Tuesday doing major upgrades on my website. Tonight I have an Annual General Meeting to attend that could go to midnight due to elections. Tomorrow could be a busy day, depending on what, if anything, my family does for my birthday. I'm sure that others will answer the call, but they may be busy doing their taxes or digging out from the blizzard that's into its second day, despite the stupid groundhog saying that spring was just around the corner over 6 weeks ago.vicmarcal wrote:And sadly this 5) point is going to be a serious bottleneck. The later we begin trying to find regressions, the later the release will take place.
That's the reason we began asking Officially to test GoldenApps in order to detect asap regressions that could stop the release.
Since almost a week there is a post called "[Official]Testing call" in order to test just 30 apps (10 guys->30 apps: 3 apps each one->30 minutes of testing).
After a whole week, just 4 guys are cooperating.
Today entirely the maniac there is no excuse with the article. Get free BeOS, DOS, OS/2, and Windows games at RGB Classic Games.
Re: When will 0.3.15 release?
I don't know in case of apps, but in case of OSes...the completeness helps a lot with stability.BlackRabbit wrote: A product can be stable without being complete.
A complete MM helps to have a stable OS, as well as a complete CC or a complete journaling file system help.
There is also the problem about what does "stability" mean for an User?
Or what an User thinks "Beta" mean?.
Is "Beta" a concept about "stability" or a concept linked to "usability"?
My bet is that the user doesn't mind about "stability" but about "usability".Using your nice "Sausage" comparison: Stability is something internal that an user doesn't care at all, they just care at stability when it breaks usability. Usability is the macroview,the big picture, here.
In the other hand a lot of Users are confusing ReactOS Stability with ReactOS Usability.
When an User launches an app and the app shows a messagebox with a red cross saying "Dll entry point not found", he thinks ReactOS is not "Stable".
When an User launches an app and the app dies silently(but in a controlled way), he thinks ReactOS is not "Stable".
When an User launches an app that uses an API in a fancy-undocumented way and leads to a crash, he thinks ReactOS is not "Stable".
Of course there is a lot of work towards Stability, but an User mainly cares about Usability.
Why an user installs ReactOS and leaves?because poor Stability or because ReactOS is not Usable from his pov and needs?
Impossible. We can't declare before hand the exactly month and year when a beta is going to take place. Believe my unwise words Again, and as I said, we should have to define first what "Beta" means. Stability?Usability?Completeness?A mix of all?It can be declared, for example, that 2013.09 is the next beta release.
Btw, and excuse my short-knowledge...but what's a gamma release?a daily build?If we tell them that there are no gamma releases available, then there are no gamma releases available.
fred02 wrote:On the other hand I don't see how the YYYY.MM.DD is more informative a new developer. It convey nothing. Is it stable? No idea. Is it feature-complete? Dunno. Is it maintained? Maybe, if the date is not to old. As you said yourself with 0.y.z at least it can be inferred that the project is not production ready and is probably half-working. As for the date of the latest release, it is probably next to the download link. After that one have to roll up the sleeves and see for himself
I agree with both. They are both meaningless. I wouldn't change to another system which has the same problems and maybe others hidden(as PR, Marketing and attractiveness).BlackRabbit wrote:The version 0.y.z is meaningless to me.
This conversation about usability is giving me a nice idea...
-
- Posts: 128
- Joined: Sat Dec 22, 2012 7:36 am
Re: When will 0.3.15 release?
I think that waiting for ReactOS to become a full Windows replacement is a mistake.
There are different, legitimate interpretations of an OS, each of which can be useful. Some people think that an operating system is critical kernel-mode components and critical user-mode components. Others think that an OS is the preceding, plus all the applications that use those critical components. I believe that an OS does not necessarily include non-critical user-mode components. A browser is a non-critical user-mode component.
More importantly, there is value in a stable OS core. For example, I had two clients in recent years who would have paid for ReactOS. One client might have paid ~$17,000US. The other client might have paid $220,000US. Neither client was aware that ReactOS existed (and they are still unaware), even though they are both Windows-based companies. In the end, Linux was selected for the second client, because it was cheaper than paying Microsoft a $3US per-unit fee for Windows CE for a $260,000,000US/year project. Let us just say that paying $220,000 to ReactOS would have been a lot cheaper than what they would have had to pay Microsoft. Yes, it is more expensive than $0US per unit for Linux, but embedded development on Linux has pro's an cons. For example: The synchronization subsystem on Linux is fundamentally incompatible with that of Windows.
There are many corporations like this in the USA: they are predisposed to using Windows, but embedded development using Windows Embedded presents a cost-disadvantage when compared to Linux. And yet still, there are companies that are paying Microsoft ~$90US per-unit for what is essentially ReactOS minus numerous drivers, etc. Ideally, they want to have a Windows like-OS for free, but nothing like that exists, so they choose Linux. And of course, if they have a choice between paying $300,000 to Microsoft, or $3,000 to the ReactOS Foundation, without losing any functionality, the choice is very easy, especially since what they really want is Windows-like development.
By declaring that ReactOS will not be complete until Firefox (for example) runs well...a lot of opportunity is being lost. An alternative strategy is:
The ReactOS volunteer contributors would continue to do what they do now, and of course, their development will be easier because their platform will be stable enough to support more effective development.
There are different, legitimate interpretations of an OS, each of which can be useful. Some people think that an operating system is critical kernel-mode components and critical user-mode components. Others think that an OS is the preceding, plus all the applications that use those critical components. I believe that an OS does not necessarily include non-critical user-mode components. A browser is a non-critical user-mode component.
More importantly, there is value in a stable OS core. For example, I had two clients in recent years who would have paid for ReactOS. One client might have paid ~$17,000US. The other client might have paid $220,000US. Neither client was aware that ReactOS existed (and they are still unaware), even though they are both Windows-based companies. In the end, Linux was selected for the second client, because it was cheaper than paying Microsoft a $3US per-unit fee for Windows CE for a $260,000,000US/year project. Let us just say that paying $220,000 to ReactOS would have been a lot cheaper than what they would have had to pay Microsoft. Yes, it is more expensive than $0US per unit for Linux, but embedded development on Linux has pro's an cons. For example: The synchronization subsystem on Linux is fundamentally incompatible with that of Windows.
There are many corporations like this in the USA: they are predisposed to using Windows, but embedded development using Windows Embedded presents a cost-disadvantage when compared to Linux. And yet still, there are companies that are paying Microsoft ~$90US per-unit for what is essentially ReactOS minus numerous drivers, etc. Ideally, they want to have a Windows like-OS for free, but nothing like that exists, so they choose Linux. And of course, if they have a choice between paying $300,000 to Microsoft, or $3,000 to the ReactOS Foundation, without losing any functionality, the choice is very easy, especially since what they really want is Windows-like development.
By declaring that ReactOS will not be complete until Firefox (for example) runs well...a lot of opportunity is being lost. An alternative strategy is:
- Get the core stable.
- Get a minimal set of applications to be stable.
- Follow a Red-Hat/etc. model of providing value to companies like the second company that I mentioned.
- Take the money and re-invest it into the ReactOS Foundation.
- Repeat, aiming for stability, always, before completeness.
The ReactOS volunteer contributors would continue to do what they do now, and of course, their development will be easier because their platform will be stable enough to support more effective development.
Re: When will 0.3.15 release?
So what exactly would indicate that the core is stable, and which applications would be in that minimal set? I doubt any developer is currently wasting their time on unimportant stuff. Some random application failing or a regression could very well be tied to core instability. What to call 1.0 shouldn't really matter that much.BlackRabbit wrote:
- Get the core stable.
- Get a minimal set of applications to be stable.
-
- Posts: 128
- Joined: Sat Dec 22, 2012 7:36 am
Re: When will 0.3.15 release?
It is the release that you give to the customer, meaning the person who is uninterested in the mechanics of your sausage factory.vicmarcal wrote:Btw, and excuse my short-knowledge...but what's a gamma release?a daily build
See the following graphic from this Wikpedia page:
[ external image ]
One can infer that, if gamma comes after alpha and beta, it must be more stable than either of them. One can infer from that image another fact: gamma is not so well-known as alpha and beta releases because it has been suppressed by software developers for whom the idea that a product should stand on its own, to be used by the customer without constant maintenance by the team who developed it, is toxic. Notice how, in the Wikipedia graphic, alpha is clear. Beta is clear. But after that, there are all kinds of things: [Release Candidate, RTM, GA, Production, Gold, etc.] I have seen worse. To me, this simply means that the development team is seeking to excuse itself for releasing incomplete software while pretending that it is complete. [Side Note: Google has a habit of doing the opposite - releasing relatively complete software as beta]. Adding new names to releases is not a substitute for self-discipline. Some colleagues of mine and I, faced with this situation, decided that we would no longer use terms like "release candidate". When the product is ready for the customer, we call it gamma. Before then, it is either alpha or beta.
First, as vicmarcal pointed out, an end-user will generally be unable to distinguish between the OS being incomplete and unstable. If an application fails to load because an exported function in a DLL is absent, an end-user will regard that as instability. We might regard it as incompleteness. That said, it would take some thinking, but I would ask:Pesho wrote:So what exactly would indicate that the core is stable, and which applications would be in that minimal set?
What is the minimal ReactOS that would be useful to anyone, especially someone who might be willing to pay?
Inherent in this question is an implication and another question.
Implication: There are multiple customers of ReactOS.
- embedded-software house
- accountant using Microsoft Excel.
- United States Department of Defense
- right-brained individual who is too cheap to by Windows and is trying to run ReactOS on a laptop from year 2005.
Question: Is it prudent too try to serve the last customer first?
I think the answer is NO. Those types of users already have Windows XP. They are not going to remove XP from ther machines and re-install ReactOS to recapture revenue that, in their minds, they have not lost. Hackers might do this to spite Microsoft. Microsoft would not care.
So stability is a relative term. If I were the customer, I would make a list of kernel-mode and user-mode components that need to be present. NTOSKRNL.EXE, NDIS.SYS, TDI.SYS, WIN32K.SYS, TCPIP.SYS, KERNEL32L.DLL, USER32.DLL, GDI32.DLL, ADVAPI32.DLL; essentially everything that allows the computer to boot and be marginally useful in an embedded device. I would definitely exclude the vast majority of user-mode applications. Notepad and DOS-like tools might be the only familiar user-mode application I would need. The rest, I would not care about. This, alone, would have great value to a very large number of people, because third-party companies could build devices based upon this core in the same way that Google is pushing Android. But to do that, the developers need a stable platform that become available for production sooner than later.
This is the opportunity for ReactOS: getting inside the 1+ billion devices that will be sold each year, by giving the people who make such devices a reason to use ReactOS.
Re: When will 0.3.15 release?
None of the core devs work on applications in any meaningful way, their focus is effectively entirely on the OS. Our applications are either from Wine or were contributed by people who were interested in writing something to learn the win32 API.
Re: When will 0.3.15 release?
ARM port. Are there rapsberry-pi x86 alternatives? Or Interl x86 micros similar to ARM ones for phablets/tablets devices?BlackRabbit wrote: This is the opportunity for ReactOS: getting inside the 1+ billion devices that will be sold each year, by giving the people who make such devices a reason to use ReactOS.
-
- Posts: 128
- Joined: Sat Dec 22, 2012 7:36 am
Re: When will 0.3.15 release?
Yes there are.ARM port. Are there rapsberry-pi x86 alternatives? Or Interl x86 micros similar to ARM ones for phablets/tablets devices?
At AliExpress, it is evident that mobile devices are practically begging for full commoditization. Do a search on that site for smartphone, and you will see that there are plenty of smartphones, phablets, tablets...all for under $100US. These smartphones, mostly made in China, could run ReactOS. The only reason they are running Android is because Android is available and free. Think about all the developers, who are thoroughly familiar with the Windows ecosystem, who would like to write killer applications for Windows-based mobile devices, but cannot because Windows Phone 7 and Windows Phone 8 are both highly restrictive. ReactOS would quickly rise to the leader position for mobile OS's beecause, contrary to popular belief, neither Android, nor iOS, nor WP7, nor WP8, and now, not even Tizen, is open. All of these OS's are actually sand-box the developer. People may say what they want about the benefits of a locked-down OS, but no one can argue with unbridled power of unrestricted development. A torrent of Win32 applications would overwhelm these other OS's. Developers would resurrect their desktop Windows applications and tweak them to use touch input and conform to the screen's form-factor. Even CMD.EXE would be highly useful to those customers who want to use the smartphone as tool. The possibilities are unlimited.
And we must remember: Right now, it is possible to take a Windows application written in C/C++ and recompile it, using Visual Studio, for an ARM device. The problem is that the resulting application will be rejected by Windows Phone 7 and Windows Phone 8. It will not be rejected by Windows CE (now called Windows Embedded), but Window CE is not what you find on consumer-oriented devices. [This explains why Microsoft did what they did - they wanted to make it so that native applications could not run on consumer-oriented Windows mobile devices.]
-
- Posts: 123
- Joined: Wed Aug 04, 2010 7:34 pm
Re: When will 0.3.15 release?
You do know that there is nothing stopping you from writing native applications on Android and Tizen that leverage the full functionality of the underlying Linux kernel? E.g. for Android there exists an app that contains a QEMU binary or one that provies an IDE for a native Android port of the Free Pascal compiler. On Android Google does not stop you from doing this. And Tizen is even more a "normal" Linux than Android is.BlackRabbit wrote:ReactOS would quickly rise to the leader position for mobile OS's beecause, contrary to popular belief, neither Android, nor iOS, nor WP7, nor WP8, and now, not even Tizen, is open. All of these OS's are actually sand-box the developer.
Regards,
Sven
Free Pascal compiler developer
Re: When will 0.3.15 release?
PascalDragon, and what about the Android's Java VM?
-
- Posts: 128
- Joined: Sat Dec 22, 2012 7:36 am
Re: When will 0.3.15 release?
Forgive me for being skeptical, but every time that I hear a claim of "native" Android support, I go to take a look, and what I see is, to put it lightly, fictitious. I have taken quick looks at Android's NDK for example. This is not a native application platform development tool, despite what Google claims. Not really. They allow some native code, then they force you to invoke the native code using JNI. I have no interest in JNI whatsoever.PascalDragon wrote:You do know that there is nothing stopping you from writing native applications on Android and Tizen that leverage the full functionality of the underlying Linux kernel? E.g. for Android there exists an app that contains a QEMU binary or one that provies an IDE for a native Android port of the Free Pascal compiler. On Android Google does not stop you from doing this. And Tizen is even more a "normal" Linux than Android is.
The same thing goes for WP8. Microsoft claimed that WP8 would support "truly native application development". But when i saw Joe Belfiore of Microsoft make this claim, his body language indicated to me that he was not being entirely honest. I saw someone else at Microsoft make the same claim on Channel 9, and again, the body language indicated to me that he was lying. So I asked one of my engineers to investigate, and he informed me that there really is no true native development on WP8. Microsoft allows the compilation of native code, but that native code is invoked inside a managed shell.
QEMU on Android, if I understand correctly, does not count. It is an emulator.
Tizen, from what I understand, is trying to go the HTML-5 route with web applications. For native applications, a framework is provided, to which the developer must conform. If the developer tries to write a "normal" Linux application, it will not work. I could be wrong, and would be happy if I am.
What I want is true native development, the same kind that I will get with GCC on a Red-Hat distribution of Linux, or Visual Studio on Windows 7, with no intervening JIT compilers, sand-boxes, emulators, translation thunks, frameworks, shells, interpreters, etc.
ReactOS could be the first OS to provide that, on heavily-deployed mobile devices, and the market response would be outstanding.
Re: When will 0.3.15 release?
All those are ARM devices. They aren't currently ReactOS compatible. So again, are there x86 Tablets/Phone devices?BlackRabbit wrote: Yes there are.
At AliExpress, it is evident that mobile devices are practically begging for full commoditization. Do a search on that site for smartphone, and you will see that there are plenty of smartphones, phablets, tablets...all for under $100US.
Who is online
Users browsing this forum: No registered users and 27 guests