Microsoft Rep To Keynote Unix Conference
An anonymous reader writes "According to ZDNET Microsoft is going to be keynoting the Australian Unix and Open Systems Users Group conference. From the article: '"Don't be put off by Chris' Microsoft badge -- he is actually a long time Unix hacker," the user group said today in a statement updating users on presentations at the conference ... Green, Microsoft's local Unix Interoperability and High Performance Computing specialist, will update the conference on his company's "Unix and open source-related activities, including their efforts to provide a POSIX environment in Windows, and to integrate Windows and Unix systems."'"
Support reiserFS and ext2 / ext3 file systems in windows!
Thank you.
Yeah, Like Microsoft would ever offer anything POSIX compliant (coughSFUcough), like Linux does across the whole system (coughpthreadscough).
It's funny they say that because I went to Microsoft career talks at my university (York University in Toronto) and every time a (different guy) came in and the first thing they said was:
..."
"... I was unix hacker for a long time before I decided Microsoft's the way to go
I recall a story I heard from someone years back, when MS spoke at a conference of UNIX geeks. A man stood up and criticized MS's POSIX subsystem for Windows NT 4.0, stating that a feature in Korn shell wasn't compatible with true Korn shell. The conversation went something like this:
UNIX geek: Feature X in your korn shell implementation isn't true to the korn shell spec. Wnen do you plan to fix that?
MS guy: We're certain it's copmatible with the standard. Are you sure you don't have it wrong?
UNIX geek: Yes, I'm sure. It's broken.
MS guy: And who are you?
UNIX geek: I'm David Korn.
May or may not be true, but it was an amusing story, nonetheless.
Diplomacy is the art of saying "nice doggy" until you can find a rock.
No, it was a necessary feature thrown in to allow the government to avoid having to throw away all their software once the operating systems they originally developed on were no longer optimal. Games of "catch the moving API" can be fun and profitable for operating system vendors, but they're not so great for third party developers and users
.net, qt, gtk, xlib, perl, python, php, java, etc etc etc. So where exactly am I locked in, again?
1.) That's funny -- when Microsoft does this, it's called "vendor lock-in".
2.) Microsoft is notorious for backwards compatibility in their APIs. Probably a bit too much, actually.
3.) Microsoft's XENIX was still going strong back then.
4.) Microsoft wasn't a 300lbs gorilla back then, they were the IBM underdogs just over their honeymoon period.
The idea behind having a portable interface was to allow customers to choose different operating systems based on price, features, and performance.
s/different operating systems/UNIX/
The POSIX spec is based off of, and therefore highly prejudice towards UNIX. And since there weren't but a few major versions of UNIX, there wasn't really much choice involved. You picked your OS, and then got locked in via server hardware and maintenance contracts.
Obviously that's not the kind of market that a vendor can siphon tens of billions of dollars of profit from, however - I'm sure Microsoft much prefers the current situation where customers can choose different operating systems based on price, features, performance, and having to rewrite or replace all their unique applications.
Microsoft's monopoly appears to be dwindling, either due to the rise of opponents like Linux, or per the natural cycle of life and death.
What API CAN'T you write for on Windows? We have the shitty POSIX subsystem, SFU, cygwin, win32,