PDA

View Full Version : MSDOS 1.1, 2.0 and Word for Windows sources released



Al Kossow
March 25th, 2014, 11:13 AM
http://blogs.technet.com/b/microsoft_blog/archive/2014/03/25/microsoft-makes-source-code-for-ms-dos-and-word-for-windows-available-to-public.aspx

cr1901
March 25th, 2014, 11:20 AM
Damnit, you JUST beat me to it XD. Now we have a legitimate download of MS-DOS 1.1 source code and a not-so-legal copy of 6.22 :D

EDIT: Version 2.0's source code (the first useful DOS) is also available legitimately! Perhaps eventually all of the versions will be available legitimately, including the DOS-based Windows :D1

Unknown_K
March 25th, 2014, 11:55 AM
Didn't Microsoft steal or license third party utilities and or code on late versions of DOS that would keep them from giving out the full source code? DOS 6.2 with stacker comes to mind.

krebizfan
March 25th, 2014, 12:11 PM
MS licensed from Central Point (anti-virus and undelete), Norton (disk checking and defragmenting) and I believe Vertisoft (for the original disk compression software that led to trouble with Stac). Those specific components might not be releasable (depending on contracts unless Symantec is willing to modify) but the rest most likely could. The current release should keep scholars busy for a few years anyway. Though with how overloaded CHM's servers are right now, it might take a few years just to download DOS 1 and 2.

pearce_jj
March 25th, 2014, 02:33 PM
Super!

Hatta
March 25th, 2014, 02:48 PM
Do note that this code is not redistributable. For reference only.

cruising
March 27th, 2014, 08:46 AM
This is nice, then you can make your own "upgrade/better version" or what ever with it?

krebizfan
March 27th, 2014, 08:58 AM
This is nice, then you can make your own "upgrade/better version" or what ever with it?

Sure, you can. Just can not send the better version to anyone. But the Research provisions should allow someone to write their own code but refer to the MS code to double check for any bizarre cases where execution differs from specification. The license seems more intended for historical articles which could be accompanied by up to 50 lines of source.

sergey
March 27th, 2014, 09:34 AM
Although MS indeed opened the source, their license is very far from being an open source license. For example their license prohibits redistribution of the source or derivative works based on it.

WARNING: Do not download or read the Microsoft DOS source if you are working or planning to work on other projects (be that open source projects, or something you do for profit, or at work) that might potentially use any knowledge about DOS internals (for example developing a device driver, or a FAT filesytem driver). Potentially having access to the DOS code can cause IP contamination issues with all kind of legal consequences.

Chuck(G)
March 27th, 2014, 10:05 AM
Although MS indeed opened the source, their license is very far from being an open source license. For example their license prohibits redistribution of the source or derivative works based on it.

WARNING: Do not download or read the Microsoft DOS source if you are working or planning to work on other projects (be that open source projects, or something you do for profit, or at work) that might potentially use any knowledge about DOS internals (for example developing a device driver, or a FAT filesytem driver). Potentially having access to the DOS code can cause IP contamination issues with all kind of legal consequences.

Hasn't enough been published already about DOS internals--even by Microsoft itself? It's not rocket science. And the source for 6.0 has been kicking around on the web for several years, as well as bits of 2K.

Besides, there's always FreeDOS if you want some source to look at...

sergey
March 27th, 2014, 11:25 AM
Hasn't enough been published already about DOS internals--even by Microsoft itself? It's not rocket science. And the source for 6.0 has been kicking around on the web for several years, as well as bits of 2K.


Right, not a rocket science. Published documentation/books is one thing, published source is another. Usually there is no problem with re-using examples from books/documentation in your code, but reusing somebody else's code in yours is a potential legal trouble.
Also Microsoft somehow managed to get various companies to pay royalties for using FAT. And we also had SCO vs. IBM case regarding Linux contamination...


Besides, there's always FreeDOS if you want some source to look at...

That would be a better choice...

geoffm3
March 27th, 2014, 06:36 PM
With a license like that, you almost have to wonder what their play is.

SomeGuy
March 27th, 2014, 07:13 PM
It would have been more interesting to publish the code for Word for DOS 1.00 from 1983, but then they would have to explain this:

http://toastytech.com/guis/word115trashing.png

Or other dirty things they may have done at the time.

Shadow Lord
March 27th, 2014, 08:49 PM
It would have been more interesting to publish the code for Word for DOS 1.00 from 1983, but then they would have to explain this:

http://toastytech.com/guis/word115trashing.png

Or other dirty things they may have done at the time.

Interesting - is that a self destruct? Never knew about that one. Although presumably this only occurred on a copy or hacked disk so its not like it is a big deal as the original disk would be unaffected.

Chuck(G)
March 27th, 2014, 09:38 PM
Right, not a rocket science. Published documentation/books is one thing, published source is another. Usually there is no problem with re-using examples from books/documentation in your code, but reusing somebody else's code in yours is a potential legal trouble.
Also Microsoft somehow managed to get various companies to pay royalties for using FAT. And we also had SCO vs. IBM case regarding Linux contamination...

That was also based on copied source code that could be demonstrated by simple object code comparison. If you're stupid enough to copy actual code rather than paraphrase what's being done, you deserve what you get. Note that there's some guy who claims that MS copied DRI's source code. A close examination of his claims shows that he's basically full of beans.

I seem to recall that there was a case involving Mentor Graphics (IIRC) that darned near sunk the company doing the copying.

SomeGuy
March 27th, 2014, 10:09 PM
Interesting - is that a self destruct? Never knew about that one. Although presumably this only occurred on a copy or hacked disk so its not like it is a big deal as the original disk would be unaffected.
That message appears if it thinks it detects a debugger. Which meant some systems with special or development software would always get that message, even from a genuine original disk. However, it doesn't actually erase anything. This anti-debugging feature was also in some of their other products at the time.

But if they had the audacity to put something like that in there, then who knows what else is underneath the hood?

Shadow Lord
March 29th, 2014, 08:33 AM
That message appears if it thinks it detects a debugger. Which meant some systems with special or development software would always get that message, even from a genuine original disk. However, it doesn't actually erase anything. This anti-debugging feature was also in some of their other products at the time.

But if they had the audacity to put something like that in there, then who knows what else is underneath the hood?

Well, a lot of companies had that I believe. If I recall correctly ICE had a number of mechanisms to try and defeat the detectors.

Chuck(G)
March 29th, 2014, 09:59 AM
Anti-pirating code can be interesting. Some of the dongle-dependent code would plug the interrupt 3 vector (used by many debuggers as a breakpoint trap) before decrypting the application code.

Trixter
March 29th, 2014, 04:34 PM
Indeed. Others would check time via the RTC before + after an operation to see if the operation was taking too long (as the RTC advances at a constant rate when in a debugger) and, if so, silently decrypt with a garbage key.