Slashdot Mirror
Peter Tattam Of The PetrOS Project Talks To OSNews
Eugenia writes: "Trumpet Software is mostly known for their Internet communications software package, Trumpet Winsock, which has been adopted by the Internet world back in 1995, at the times where Windows 3.1 and Win95 did not come as standard with full internet connetion capabilities. But the main product these days for Trumpet Software is PetrOS, a 32-bit Operating System, which has the goal to be compatible by all means (binary and API compatible) with Microsoft Windows. OSNews is interviewing the main architect behind the project, Peter Tattam, who talks in depth about PetrOS, and also there is shown an early screenshot of the PetrOS GUI, which is still under heavy development." And it's been (not surprizingly) under heavy development for a while. Building a Windows-compatible OS from scratch surely isn't easy, but from this interview (including screenshots) they're having quite a go of it.
It's an interesting goal and everything, but it's a moving target, as far as I can tell.
Every few months I get a new copy of the MSDN library (basically documentation for all Microsoft's APIs), and every time it's a bit different from the one before.
Of course I don't want to tell anyone how to spend their time, but if it were me, I'd spend my energy on building something new, rather than just trying to be compatible with something that'll just be obsolete by the time I'm done anyway.
- In Capitalist America, law violates YOU!
I can't imagine why anyone would try to base a company on cloning Windows. So I read the interview.
A) Cheap alternative for desktop users -- users say "they wished they had something better without having to pay big big bucks." Win2k is, what, perhaps $200. PetrOS will have to sell for $50 or less, then.
And it'll be obsolete the moment MS changes an API. Or the moment MS makes MSIE crash when it detects PetrOS.
B) Embedded market -- er, no. The embedded market wants Linux, QNX, EPOC and other OSes. They're either free, hard realtime, or extremely small.
C) Servers -- er, no. If you want cheap, then you choose a BSD or Linux. If you want to be able to blame someone, you choose MS. You don't go out and buy some $50 clone of MS.
D) Clustering -- er, no. Not unless you're just goofing around. Kind of money invested in building a cluster, you don't go pick up a $50 clone of Windows to run it!
While this is a pretty cool project, I simply can't see that it's a profitable one...
--
Don't like it? Respond with words, not karma.
You know back in the day....well not that far back Trumpet was the best connect software rolling around. What did you have maybe 2 or 3 that did anything worth talking about, that you configure and get working with any consistancy? Cammilion and Trumpet where the best, please correct me if I am wrong, and Cam was sad.
Interesting enough with the os that would be compat with winOS that is free. It would seem that the linux community would have done this years ago if for nothing else to put a funnel on the venture capital. If Linux had become 100% comp with windows software, via a deamon or what have you....things could and would be very different in many ways. Then again part of problem with windows is the software, so Linux would end up bringing bugs in...it is kind of catch 22 sale to the devil....
Neck_of_the_Woods
#/usr/local/surf/glassy/overhead
I can't decide if this is the work of a sincere person who is sadly deluded, or a marketing ploy to flog a few more dollars out of investors. It's hard enough getting your foot in the door with a pure desktop operating system that is better than Windows (BeOS, OS/2), much less one whose entire goal is to play catch-up to Windows itself. There is nothing here to appeal to people who already know and like Windows, and it's certainly not going to appeal to people who don't like windows either.
Beyond that, the technical feasibility of it is questionable. Microsoft is well known for making its Win32 API, filesystems, etc, moving targets. It's taken the Wine and NTFS teams a long time to get where they have, and even then they're pretty far from complete compatibility. What makes these guys think they're going to get any closer?
Oh well. I guess it goes to show you, there's always someone trying to ice skate uphill.
Sure sounds like a good idea reading about it but it's totally unpractical. He wants to aim for the desktop, which means he's gonna have to go for the average user. Since windows comes "free" on all mainstream PCs today, the average user will never ever have the need to run a windows clone.
Sure, us geeky people will have fun messing around with it, but we already have our joys in running unixes.
486 DX processor.
2MB System RAM (4MB recommended).
At least 2MB Hard drive space. This struck me as funny. I understand the point they're trying to make, but does this mean I'd better go out and get that 2MB upgrade I've been looking at for the past ten years for the 486DX I intend to run this on?
"We're well aware of the dominance by the key player in this market - we just want to coexist, not supplant." - Peter Tattam
"I once preached peaceful coexistence with Windows. You may laugh at my expense -- I deserve it." - Jean-Louis Gassée
"Those who cannot remember the past are condemned to repeat it." - George Santayana
If you read the article, you would notice that they're using Borland's Object Pascal. Go ahead, everyone, read the Brian Kernighan article. Then go and look at the Object Pascal language reference. Go ahead...I'll wait.
Notice anything? Like how almost EVERY SINGLE objection he raised is NOT APPLICABLE to Object Pascal?
Object Pascal should be called "Sensible Pascal". It allows you to break all of the rules, just as C does. It just makes you jump through a few hoops to do it. That way you generally know what you're doing isn't such a great idea.
Pointer arithmetic? OP allows it. It's dangerous though...inexperienced programmers can blow themselves up pretty easily doing it. So...they make you do some castings to get around it. Good idea? Bad idea? Depends on what you're trying to accomplish.
I'm sure all the hard-core hackers here will laugh and snort and generally dismiss anything I say about OP, but those people aren't who OP is aimed it.
OP is aimed at programmers who WANT strong type checking and WANT the safety net the Pascal language provides and WANT a language that allows rapid object-oriented development in a language that is easy to read and understand.
C lends itself to fast, loose code, but is hard to learn and tends to be rather terse to read. It's definately not for learnin'!
OP allows the newbie to step in and test the waters with simple language specific things. However, as you become more comfortable with the language, you can EASILY do just about everything C allows you to do.
The learning curve is a gentle slope, and doesn't have any unexpected drop-offs where the language fails you.
Anyone that dismisses OP out of hand hasn't used it. Anyone that has used a recent OP doesn't use C anymore if they can help it.
Whoa! This guy has guts. Not only is he trying to battle William Gates III he's also named Bill's OS after himself! Even Bill Gates himself wasn't arrogant enough to instill a "BillOS" upon us! Sheesh.
Your pizza just the way you ought to have it.
The two primary objections I see here could both have been raised when Linus decided to write a UNIX clone in 1991:
1. "Nobody could possibly write his own UNIX kernel! It's too huge of a system!"
2. "UNIX has been dying since the mid 1980s. Who would want to start writing a system that will be dead on arrival?"
And what's more, the person who submitted the Slashdot article had the gall to steal the name "Eugenia" from the person who wrote the linked article! It's not just copyright infringement, it's identity theft as well!
The "secret APIs" are not a rumor. Notice the dates on these references, the secret APIs have been in NT all along.
MSFT hasn't hesitated to use the secret APIs either. From the July 10 InternetWeek: Microsoft has historically achieved market dominance by controlling APIs and forcing competitors to write software to Microsoft's APIs, then changing the APIs. "Instead of satisfying their own customers' demand, competitors are busy catching up with Microsoft," said IDC analyst Dan Kusnetzky.
From the October 8, 1998 NY Times: And Microsoft, the people added, did what it has always denied it does -- used access to its technology as a powerful lever in business negotiations, by offering Netscape preferential access to the Windows "application program interfaces," or A.P.I.'s, the links that enable other companies' programs to run smoothly on the Windows operating system. By turning down the deal, Netscape, they say, would not have that preferred access to Microsoft technology -- a threat that Microsoft fiercely denies making.
Think about it - can you, using only Win32, write all of the stuff that MSFT provides with NT/W2k? No. Clearly, MSFT keeps APIs to themselves. MSFT wants to allow itself the latitude to write faster, more functional programs than the ordinary developers can write. MSFT has proven time and time again that it will use secret APIs to its own advantage, or to the advantage of selected partners (Executive Software, for example). This practice is certainly bad for the consumer. Secret APIs raise the cost of entry into the NT system software market, which will keep out competitors, raise prices, and reduce choice.
What's the minimum software that you'd have to replace in order to de-Microsoft a typical PC?
... a huge selling point. People can write to other API's later as time goes on, like Apple moving to the Cocoa API by just putting it out there, and developers can adopt it as they please. In the meantime, people still have a lot of Win32 apps that they want to use on their PC's, and Carbon apps on their Macs. If somebody wants to replace Windows with Linux in great numbers then getting the Win32 API going might be just as important to their vision of Linux as having the Carbon API and the Classic environment is on Mac OS X. Computers exist to run the apps that are on them RIGHT NOW. To be able to switch to Linux and still use all those old Netscape and SmartSuite and DOS game discs you have from 1997 might be cool for some people. Also, the Enhanced CD's in people's music collections have Mac versions and PC versions. The PC versions are Windows versions. If Linux were to replace Windows tomorrow, then adding support for those CD's would be a priority. Whatever replaces Windows (if anything) would probably have to make that level of commitment to the PC platform's non-Unix past, so why not just enable the stuff to be run natively?
I don't think the average frustrated Windows user wants to stop running IE and MS Office. You don't have to replace those. Users probably like most of their other applications, too. What most people want to stop is that the whole shooting match is dependent on Microsoft Windows. It gets better, or it gets worse, according to Microsoft. Any security in Windows is always open to Microsoft's applications. Features come and go and are reliable or unreliable based mainly on who is currently having their air supply cut off, etc. It's a drag.
One of the nice things about switching from using Windows full-time to using a Mac full-time is that suddenly Microsoft is just another software company. Replacing the core of Windows with something non-Microsoft could give Windows users that feeling that they can really hold Microsoft accountable for their dirty tricks and just switch off the Microsoft software if they want to, without losing their computer's functionality and capabilities. At least you could close IE when you wanted to, and run an Adobe or other app on a non-Microsoft foundation, perhaps with better performance as well.
Imagine a Windows software installer that you run, and after a reboot, the computer functions in basically the same manner, except that it can function with all of the Microsoft apps shut off. They are running a non-MS kernel and certain systems that are acting as a traffic cop or conscience for Windows. You can run IE or any MS app, but they are in their own processes and effectively being contained from any scripting mischief or whatever.
Wouldn't this be a good way for the Linux kernel to get onto more machines full-time? If it could replace the NT kernel and the user could still keep their apps at least
These guys at Trumpet don't seem to look at it as OS and applications and who owns what. If you think of the user's PC, the Trumpet core OS product was replaced by yet another Microsoft core OS product. Now Trumpet is working on new core OS products that replace some of Microsoft's at the user's discretion. Even if they only replace the functionality of Windows 95, then that gives them an instant application base that is very large. Also, they would give Windows developers a reason to stick to core Win32 stuff and make good cross-platform code in order to run in both places, even just for the principle. If you are making subnotebooks or something, why make one with Pocket Windows when you could make a Microsoft-free Windows95-compatible Pentium-class subnotebook that runs real Windows and DOS apps (not all of them, but LOTS) with excellent performance? There are interesting things going on when you can get the Win32 application base or a subset of it without having to build your business on Microsoft.
Well there is long-lasting wine project that is capable of running windows apps. It has been developed for many years, but we're still far from running "most" windows apps on linux.
Here is a list from their "competitors" page:
These are the dead or very slow projects. The statistics can be wrong, but usually this is not the case.
There are also "alive" projects. One is twin, but it's slow. Another is odin which was former OS/2-WIN code. But I do not know much about that one.
'nuff said.