Alex also does the NT5.0 kernel not have patch guard support? It should be a good "secure" feature to add? So there is not that many change on karnel side between 5.0 and win10 one a part from the secure kernel stuff?
2016-05-17 20:58 GMT+02:00 Riccardo Paolo Bestetti <riccardo.kyo...@live.it> : > Il 17/05/2016 20:23, Magnus Johnsson ha scritto: > > I am an utter scrubnub, but has followed this mailing list for.. more than > a few years. Its not unheard of to get an angry response to a commit saying > that its an API that is newer than the targeted version. > > Being a scrubnub lurker I can't help but think that being somewhat > agressive on compatibilty modes would be nice, but.. I'll shut up now :'). > Just wanted the "people are angry when non-targeted versions of API's are > implemented" thrown in there. > > +1 to this, forgot to mention it. > > > Il 17/05/2016 20:04, Alex Ionescu ha scritto: > > The project doesn't have to be hard-coded to NT5. For example, I am > building a UEFI loader/bootmgr based on Windows 10, because 2003 > doesn't boot on UEFI systems. > > That being said, I don't see any good reason for us not to still > mainly focus on 2003 for the kernel. The kernel is NOT what's > preventing apps from working, or hardware from working. What's > preventing that from working is: > > 1) Lacking user-mode APIs, and in some cases Rtl APIs (sure, implement > Win 10 ones!) > 2) Lacking hardware support for things like UEFI (I'm working on it), > AHCI (we have a student working on it), USB 3 (someone can implement > this...but USB 2 barely works), etc..etc..etc.. > > Find me a single device driver that *only* works on NT 6... Server > 2003 is still a support MS OS, so by definition there's still drivers > for it. > Best regards, > Alex Ionescu > > Hello Alex, > > You can definitely focus on 2003 for the kernel, however keep in mind that > you would be re-implementing 14 years that is an advanced state of the > process of migration towards new software. > Yes, user-mode APIs and lacking hardware support is what prevents apps > from working now. Lacking kernel support is what will prevent apps from > working in a few years. > > A clear example: most of new Windows application are being written against > the .NET libraries, and the last version of .NET has already dropped > Windows XP support. I doubt you can run .NET 4.5 on a 2003 type of kernel > even if you implement the last user-mode interface (and afaik you can't > even do that unless you implement some parts of NT6), so this is already > significant portion of apps that ReactOS will never be able to run if you > stick to NT5. Unless you want to reverse .NET 4.5 too and rewrite the it to > run on NT5... The same goes for all other examples of this kind, both from > Microsoft and from 3rd party vendors. > > I've also found an example of a device driver that only works on NT6, and > in fact I didn't even have to look for it: it was the first device I saw on > my desk. Elgato Game Capture HD, and that's pretty much the entire YouTube > and Twitch gaming business (a multi-million dollar industry, for those who > aren't familiar with it). > Also, some of the newer printers in my father's business, as well as the > GPS navigation system map update driver for some of his trucks, to give > some "job industry" examples. > > Windows Server 2003 is definitely supported by Microsoft (which doesn't > mean it is supported by others, and it mostly isn't), but it is almost > dead, and it would only do harm to deny that. > Implementing NT5 with the "NT6" user-mode API would lead to an Open Source > Windows Vista, without the "wow Aero looks so cool" factor. Please don't > waste all the magnificent work you've done so far like this. :) > > Best regards, > --- *Riccardo Paolo Bestetti* > > _______________________________________________ > Ros-dev mailing list > Ros-dev@reactos.org > http://www.reactos.org/mailman/listinfo/ros-dev >
_______________________________________________ Ros-dev mailing list Ros-dev@reactos.org http://www.reactos.org/mailman/listinfo/ros-dev
