windows research kernel

Discussions on more advanced topics such as monolithic vs micro-kernels, transactional memory models, and paging vs segmentation should go here. Use this forum to expand and improve the wiki!
User avatar
qw
Member
Member
Posts: 792
Joined: Mon Jan 26, 2009 2:48 am

Re: windows research kernel

Post by qw »

Hey, my XML reader did not interpret that correctly... It said 'ignore="on"', didn't it?
OSwhatever
Member
Member
Posts: 595
Joined: Mon Jul 05, 2010 4:15 pm

Re: windows research kernel

Post by OSwhatever »

Solar wrote:<rant offtopic="true" ignore="on">

...

</rant>
Yeah, this should really be taken to another thread.
rdos
Member
Member
Posts: 3286
Joined: Wed Oct 01, 2008 1:55 pm

Re: windows research kernel

Post by rdos »

OSwhatever wrote:I never understood the hatred towards Microsoft among many software developers.
Just as Solar points out, it is often based on their activities during the 80s and 90s. This was when I had to cope with their products that had many serious problems from a technology point of view. MS launched fix after fix for their old bad products (VCPI, DPMI, XMS, EMS and DOS-extenders). The worst one was probably DPMI, since this interface promised interoperability, but it was never delivered, and instead MS broke the whole thing. Their proprietary network protocols was another "feature" that probably made many people unhappy. We also have the aspect of how MS competed. The issue of the web-browser being so tightly intergrated into Windows that it was "impossible" to remove is another factor that contributes. Especially since this tight intergration has many stabillity-related problems.
eddyb
Member
Member
Posts: 248
Joined: Fri Aug 01, 2008 7:52 am

Re: windows research kernel

Post by eddyb »

OSwhatever wrote:
Solar wrote:<rant offtopic="true" ignore="on">

...

</rant>
Yeah, this should really be taken to another thread.
Maybe it should be put somewhere on the front page or given a wiki page. That's a really good explanation Solar, thanks :).
User avatar
bluemoon
Member
Member
Posts: 1761
Joined: Wed Dec 01, 2010 3:41 am
Location: Hong Kong

Re: windows research kernel

Post by bluemoon »

OSwhatever wrote:I never understood the hatred towards Microsoft among many software developers. Microsoft have problems and some products have been bad. However, I wouldn't call Microsoft kernel engineers incompetent.
The whole idea of *A and *W function make me go nuts. They could have think twice and designed it better and use utf8 instead of doubling the APIs (and require you to recompile and maintain multiple binaries for deployment).
User avatar
Owen
Member
Member
Posts: 1700
Joined: Fri Jun 13, 2008 3:21 pm
Location: Cambridge, United Kingdom
Contact:

Re: windows research kernel

Post by Owen »

bluemoon wrote:
OSwhatever wrote:I never understood the hatred towards Microsoft among many software developers. Microsoft have problems and some products have been bad. However, I wouldn't call Microsoft kernel engineers incompetent.
The whole idea of *A and *W function make me go nuts. They could have think twice and designed it better and use utf8 instead of doubling the APIs (and require you to recompile and maintain multiple binaries for deployment).
UTF-8 was developed as a standard (by the Plan 9 guys) in January 1993. Before then, you had UCS-2 or UTF-1. UTF-1, while a variable length 8-bit coding like UTF-8, was an absolutely stupid standard which was, IIRC, incompatible with ASCII. It was certainly a horrendous mess to encode and decode (requiring multiplication and division by non-power-of-two values). Windows NT 3.1 (the first release) was released in July 1993. Using UTF-8 throughout NT would have required a time machine, or a major redesign which would have pushed back the release date significantly.

Besides, at the time the official line was that Unicode would never surpass 16-bits to encode a character. UCS-2 was a fixed length encoding, and this had processing advantages.

You could argue that they should have developed their own version of UTF-8 - but then people would be complaining instead that Microsoft had invented its own proprietary 8-bit Unicode encoding (Their July 1993 release would have meant that UTF-8 would have already existed).

But anyway, now you're ending up with Windows having three 8-bit encodings in play. UTF-8, your declared system wide Unicode encoding. The legacy 8-bit ("ANSI") encoding, for 16-bit apps. And the other legacy 8-bit ("OEM") encoding, related to the DOS emulation layer and some aspects of the console. Its bad enough that they even have one in play in the first place; things are now just getting silly.

It is best to think of things this way:
  • The W APIs are the native APIs. You should use these on all versions of NT. You should use them on all versions of Win9x when the MS Layer for Unicode is available
  • The A APIs are the compatibility APIs for Win9x apps, and for ported Win16 apps
You never need to compile multiple versions of your app (unless you want it to run on Win9x and object to MSLU. In which case, you're a fool).

Also: Standardizing on UTF-8 presupposes that it is a better character set than UTF-16. I would disagree. (UTF-16 is far simpler to decode)
rdos
Member
Member
Posts: 3286
Joined: Wed Oct 01, 2008 1:55 pm

Re: windows research kernel

Post by rdos »

Owen wrote:Also: Standardizing on UTF-8 presupposes that it is a better character set than UTF-16. I would disagree. (UTF-16 is far simpler to decode)
UTF-8 is superior for most languages, simply because it preserves ASCII-compability, which unicode and UTF-16 does not do. If it is harder to decode for odd characters is not an issue.
But then MS are known to break standards and compability.
User avatar
Solar
Member
Member
Posts: 7615
Joined: Thu Nov 16, 2006 12:01 pm
Location: Germany
Contact:

Re: windows research kernel

Post by Solar »

eddyb wrote:
OSwhatever wrote:
Solar wrote:<rant offtopic="true" ignore="on">

...

</rant>
Yeah, this should really be taken to another thread.
Maybe it should be put somewhere on the front page or given a wiki page. That's a really good explanation Solar, thanks :).
Doesn't really fit into the OSDev Wiki, so I put it on dev.rootdirectory.org for reference. If I find the time, I'll juice it up with a couple of contemporary screenshots and other artifacts.
Every good solution is obvious once you've found it.
UX
Posts: 15
Joined: Sun May 08, 2011 2:21 pm

Re: windows research kernel

Post by UX »

Solar wrote: Once upon a time, there was a real choice of operating systems. MacOS, AmigaOS, RiscOS, AtariTOS - the competition fueled quantum leaps in OS design, look & feel, functionality, and interoperability.
question is, was that choice a good thing in terms of standards?
For OS hobbyst, I believe it was great.
OpenSource showed a way how this monopoly could be broken - until that path was monopolized by the GPL/Linux community in their own, special way. Well, at least their pressure made Microsoft move again, but as far as choice is concerned...

It wasn't monopolized by GPL/Linux...there's just cannot be 10 more OS on the "market". People chose Linux as a flag carrier of OpenSource. It isn't Linux's fault...
If it wasn't Linux, it would be something else.
Well, we're stuck with "pay-per-view" Windows (with the "pay-per-view PREMIUM" bastard MacOS), and "wontfix" / "works-as-designed" Linux.
umm...MacOS...

And that won't change anytime soon, if at all: You play proprietary, Microsoft will crush you with patent sues as soon as you show up on the radar. You play OpenSource, Linux will assimilate you.
Not sure what assimilate means in this contex, but it largley depends on who or which company is using Linux and which distribution. Linux is in mess, but it's getting cleaner and cleaner.
The goal is: If I buy brand new PC, I really don't want to give extra 100$ just to make it run. OS should be standard, like any other open standard in the industry. I don't care weather it's Linux/MacOS/MSWindows.
It should come with free compiler, documentation, browser, CD/DVD/BR burner....everything to make the damn machine usable...and every other specialized app. like Autocad, Maya, Photoshop...should be payed and OpenSource alternatives will just force them to be better and cheaper products. Thats good, thats healthy economy environment.
UX
Posts: 15
Joined: Sun May 08, 2011 2:21 pm

Re: windows research kernel

Post by UX »

OSwhatever wrote: I never understood the hatred towards Microsoft among many software developers.
Well, Solar said it right....you are so used to it that you cannot see beyond it. When you pay for your Windows box, you never own that software, but only premission to use it. They would sell you the brick if they weren't forced to advance.
Microsoft was never the most inovative company in the industry and with it's monopoly, has killed many others smaller companies which have had a potencial to deliver great products and there for create more competative enviroment which is always in the interest of the end consumer.
Luckily, things are changing...slow but are changing. Businiess model of the 80s is not so dominate like it was 10-15 years ago.
More companies are finding their way in OpenSoruce...Google as the biggest example, and Windows is loosing the battle in the mobile device arena which should affect desktop market in the years to come. Server market varies, but Linux is practically killing every Unix based OS and growing stronger in datacenters(ask Solaris guys about it). Some time ago, no one would even dare to install it and it was a joke...especially to MS. Now, it's not so funny anymore.
User avatar
Solar
Member
Member
Posts: 7615
Joined: Thu Nov 16, 2006 12:01 pm
Location: Germany
Contact:

Re: windows research kernel

Post by Solar »

Solar wrote:..."pay-per-view" Windows (with the "pay-per-view PREMIUM" bastard MacOS)...
Every good solution is obvious once you've found it.
UX
Posts: 15
Joined: Sun May 08, 2011 2:21 pm

Re: windows research kernel

Post by UX »

Just an observation....maybe not very accurate...but why MacOS is way more popular in US then in Europe? I mean...here it's more exotic then...uugghh..OpenVMS.
Yes, they make expensive hardware, but man...Switzerland, Sweeden, Finland, Holland, Germany...all have much higher standard then US.
OSwhatever
Member
Member
Posts: 595
Joined: Mon Jul 05, 2010 4:15 pm

Re: windows research kernel

Post by OSwhatever »

UX wrote:Just an observation....maybe not very accurate...but why MacOS is way more popular in US then in Europe? I mean...here it's more exotic then...uugghh..OpenVMS.
Yes, they make expensive hardware, but man...Switzerland, Sweeden, Finland, Holland, Germany...all have much higher standard then US.
I'm not sure if I agree with that. What I've seen of the US, the standard of living is quite good and just as good as parts of Europe. I guess it really depends where you live in Europe and USA. I haven't observed the differences of Mac usage between Europe and the US but I would definitely say that it depends on the profession so maybe the concentration of different professions in different areas would have an impact.
User avatar
NickJohnson
Member
Member
Posts: 1249
Joined: Tue Mar 24, 2009 8:11 pm
Location: Sunnyvale, California

Re: windows research kernel

Post by NickJohnson »

berkus wrote:I guess it's because there's no California in Europe?
Yeah, but there's no China (where the computers are actually assembled) in the US either. It's at least not a matter of shipping distance. I would probably point to advertising instead.
User avatar
Solar
Member
Member
Posts: 7615
Joined: Thu Nov 16, 2006 12:01 pm
Location: Germany
Contact:

Re: windows research kernel

Post by Solar »

I think berkus tried to point out the effect of a computer as a status symbol / "chic" gadget.

I don't see many Macs around here in Germany. Actually, outside of computer fairs and demo parties, I have yet to see one "in the wild".
Every good solution is obvious once you've found it.
Post Reply