Switching the target to the NT 6.x kernel might save R&D time in the long run

Here you can discuss ReactOS related topics.

Moderator: Moderator Team

wdstudios
Posts: 35
Joined: Mon Aug 13, 2018 8:53 pm

Re: Switching the target to the NT 6.x kernel might save R&D time in the long run

Post by wdstudios »

"Expect" does NOT imply that you are giving orders! I can say "I expect Windows 12 to suck even worse than Windows 10" but that doesn't mean that I want Microsoft to make it happen.
florian
Posts: 510
Joined: Tue Nov 01, 2005 2:19 am
Location: Germany

Re: Switching the target to the NT 6.x kernel might save R&D time in the long run

Post by florian »

shunesburg wrote: Mon Jun 03, 2019 3:10 pm
dizt3mp3r wrote: Sat Jun 01, 2019 11:45 am It may be merely an English translation issue. Quim may not understand the nuance that his "we expect" comments give.
Maybe, for French for example, there is an "undefined" plural personal pronoun, "On".
It's a little difficult to explain, but it's like a "We" but in a general meaning. You can be include or not, it's mean like "The people", and when you use Google Translate, always it's translate by "We".

If it's that, the good translation should be: "The people expects" or "It's expected" and not "We expect"
And your French "on", by the way, is comparable to German - as well as Old and Middle English - "man". Or its Middle English allomorph versions "men" or "me". See Visser, Fredericus Theodorus 1970: An Historical Syntax of the English Language. Part One: Syntactical Units with One Verb. 2., revised edition. Brill Academic Publishers: Leiden. pp. 51–52, § 65 (man as Subject).
PurpleGurl
Posts: 1790
Joined: Fri Aug 07, 2009 5:11 am
Location: USA

Re: Switching the target to the NT 6.x kernel might save R&D time in the long run

Post by PurpleGurl »

wdstudios wrote: Tue Jun 04, 2019 7:21 pm "Expect" does NOT imply that you are giving orders! I can say "I expect Windows 12 to suck even worse than Windows 10" but that doesn't mean that I want Microsoft to make it happen.
Thank you. I think many of us would expect Windows 14 or whatever their next thing after that will be called to suck even more, until given evidence to the contrary. That is just based on precedent and the anti-Microsoft bias of many. We (I think I can safely use this) are the group trying to "make Windows great again." We are not like the Linux crowd who are like emigrants to another "country." We haven't given up on the Windows model or we wouldn't be here. And we play different roles. The devs are the true stars. The rest of us are like cheerleaders, janitors, groundskeepers, announcers, fans, crowd control, etc. Not everyone can be an athlete, as it takes "true" fans to keep the "athletes" wanting to stay in their "contracts." By "true" fans, I'm thinking of those who buy tickets to see a game, and TV networks that pay to get to broadcast the event. Our equivalent would be donors. Not quite sure of my role, I guess I'm more like a commentator. And in sports, there are often 2 varieties of commentators, and they are working together. One type is serious, knows all the facts about the game, and tends to stay focused. The other type is more about history, humor, inane details and makes the first type look good. So the less serious announcer draws more attention to the serious announcer and helps the ratings.

Now, it would be helpful, IMHO to get back to the original topic. It seems that at this point, we simply need to have any solid kernel and API set. Once it is complete, we can work on another set, since we'd have a solid foundation and developers who would know more about what they are doing. That is like in my QB programming days, where sometimes just getting a basic working program was the hardest obstacle. And once you get working code and know in and out how your existing code works and what the important aspects are, then you can optimize it, add to it, etc. An example of some code I was writing was a simpler version of the "Input" statement in Quick Basic. I wanted one that would not mess up the screen, accept no more text than what would fit after the prompt, would allow for backspace or left arrow to delete characters, and would not pull in a large library of functions. But it took a while for me to get it to the point where it worked as expected. And once I knew what I was doing, I was able to trim out more code. I later converted that to an assembly module that worked with QB's internal functions, which helped me to clean up the QB version even more. So you have to start somewhere, and starting ROS at the 5.x level is as good a place to start as any.

My point is that we need a solid, stable, working foundation of code to use. Once that exists, any of use can modify it, improve it, or whatever else. So whatever the original goal is, it seems it would be best to stay on that path. Also, understanding how Microsoft did things "back then," will help us to learn how Microsoft got to where they are with the code, and it would help us to maintain compatibility with older but useful software. So even if we were to go straight to NT 6.x, there would still be a learning curve in getting older but important/relevant stuff to work.

So everyone saying, "We should code it this way" or "We should reach these goals," are unintentionally impeding the project. If a developer stops and tries to answer such things and justify their current path, they're taking time that they could be using for coding. The same goes for those who don't understand the goals or want to suggest ideas outside of our current stated goals, such as saying, "We need to be able to use Linux drivers and run all Linux software." Then devs have to take time explain this is not Linux and not our mission. If you want something way outside the goals, then feel free to fork the project and make your own part NT, part Linux environment, make your own homegrown Windows 10, etc. If you don't like that devs are working on NT 5.x, then make your own 6.x fork, maybe interest some new devs to help you. If you do useful work in the fork, then you can talk about merging parts of yours back into the original ROS. And ROS just might be "mature" enough to take your patches.
shunesburg
Posts: 215
Joined: Wed Feb 21, 2018 3:46 pm
Location: Somewhere in France

Re: Switching the target to the NT 6.x kernel might save R&D time in the long run

Post by shunesburg »

wdstudios wrote: Tue Jun 04, 2019 7:21 pm "Expect" does NOT imply that you are giving orders! I can say "I expect Windows 12 to suck even worse than Windows 10" but that doesn't mean that I want Microsoft to make it happen.
If what I suspect is true. It's just misunderstood and not an order, just a bad translation.
florian wrote: Tue Jun 04, 2019 11:54 pm And your French "on", by the way, is comparable to German - as well as Old and Middle English - "man". Or its Middle English allomorph versions "men" or "me". See Visser, Fredericus Theodorus 1970: An Historical Syntax of the English Language. Part One: Syntactical Units with One Verb. 2., revised edition. Brill Academic Publishers: Leiden. pp. 51–52, § 65 (man as Subject).
"On" it's just an example for French, maybe the guy use an other language.
If it's that the good translation (by reverse engineering) could be:
Quim wrote: Thu May 30, 2019 7:59 am Maybe in 1 or 2 years, NT 5.2 kernel should be almost complete, and by extension the NT6.1 kernel as fast as they can....

There is no detailed technical information about how much of NT 5.2 kernel is complete? And how much of NT 5.2 is still incomplete?
Nobody gives an estimation... (example: "according to my experience or what there is to see, it's not bad to think that the NT 5.2 kernel is 75 % complete, more or less").
Post Reply

Who is online

Users browsing this forum: No registered users and 53 guests