[alexander]

Sun released the source for Solaris 10 OS and have finally shifted to proprietary off their backs putting more pressure on microsoft. They have released 1600 pattents to the open-source community, finally realising that it is the way to go!

Put one up for Sun! Actually they are the first major Unix supplier to do so, now we wait for sisco and netgear to do the same and microsoft will start loosing their cool...
anyways here's the link
http://enterprise-linux-it.newsfacto...rc#story-start

[Stargazer]

Good news! I'm not into Solaris, but I understand it's an important system. I especially like how they released 1600 patents. I suppose that's the only way to use patents in the future if software patents becomes a reality in the EU, that is to patent everything you can and then release it. That way no one else can control certain techniques in order to get royalty.

[alexander]

if microsoft follows, Linux would rule the world!

[Buffy]

Yoo hoo! Linux != Solaris...

Solaris is based on the original UNIX code (by now a wild mix of the System V and BSD branches). Many rags have commented on the fact that the greedheads at SCO will be filing lawsuits any day now...

The more open source the better though...

Cheers,
Buffy

[alexander]

Yoo hoo! Linux != Solaris...

Really? I must have forgoten...

SunOS was once based on Berkley Systems Unix code, but not for that long, Solaris was much modified, most Unix parts were rewritten speciffically for their system, Sun is smarter than using the original code, they have lawyers who know copyright, as well as patent laws, and i dont think they hold license to use Unix, Solaris 10 is such a huge mix of code from different Unis systems. The only true Unix out there is BSD, because BSD is Unix, well technically there's no Original Unix left, for a good reason, 1960 technology is a bit old...
What i meant was the Linux ways. I guess I could have clarified that line by saying GNU will rule the world, but they pretty much mean the same in a sense...
here's why i said what i said:
1) windows releases their code
2) crackers prove that windows code is the most insecure, by releasing a supervirus to destroy the system, and hackers dont want anything to do with windows
3) people doubt their safety with windows OS and realize that they have to switch and as open-source is the way to go, what better system to switch to as the one that defined open-source to start with, and guess what, it is not Solaris or BSD or OS/2 or Minix....

[Buffy]

Originally Posted by alexander

SunOS was once based on Berkley Systems Unix code, but not for that long, Solaris was much modified, most Unix parts were rewritten speciffically for their system, Sun is smarter than using the original code, they have lawyers who know copyright, as well as patent laws, and i dont think they hold license to use Unix, Solaris 10 is such a huge mix of code from different Unis systems. The only true Unix out there is BSD, because BSD is Unix, well technically there's no Original Unix left, for a good reason, 1960 technology is a bit old...

Ya missed a step. Orig Sun OS was a BSD* port (cuz Bill Joy wrote so much of it at Berkeley, whadda ya expect him to do when he goes to start Sun with Scott?). Unfortunately, AT&T got most everyone except for DEC (Ultrix was also BSD) to go with the System V code base, so to be compatible, after much grumbling by the true believers, Solaris was built and was a System V port. Since then, lots of the old SunOS code got put back in where it made sense to, along with a LOT more code developed in house. You'd be amazed how much old code is still around in the code base, as the "if it ain't broke don't fix it" rule is golden. There's no question that SCO has a plausible lawsuit. Most are doubting that every line of code was rewritten to the point that you could not argue that the rewrite was superficial and retained the "spirit" of the orginal code (the rights are based not only on copyright but on licensing restrictions that are more vague). SCO did buy the System V license rights...

Linus was brilliant to say, "I've never seen the code, I'm gonna write it from scratch", which is almost EXACTLY what Compaq did when they built the first IBM PC clone. WRT Linux, the only hope SCO has is to say that some of the contributors had access to Unix source and "willfully and without regard to SCO's license rights" contributed the System V-derivative to the Linux code base. Of course since Linus has never seen that code, even if he had inspected it, he wouldn't know that it was derivative, thus giving SCO an opening on legal (but pretty darn likely baseless) grounds. Most people who know don't believe that any infringing code amounts to enough that it couldn't be rewritten easily to be non-infringing. Solaris sounds like a much more problematic situation tho....

Old code all rewritten? Don Knuth, Kernigan & Plaugher, Niklaus Wirth, and maybe Chris Date are ALL you need to read to understand good programming (and okay, okay, I'll throw in Bjarne Stroustrup, but so much C++ design was influenced by Smalltalk from 10 years earlier). Mebe yur not old enuf to remember Y2K

Originally Posted by alexander

What i meant was the Linux ways.... <snip />

Won't argue about any of the rest of what you wrote!

Cheers,
Buffy

* BSD = Berkeley Standard Distribution: branched off of the AT&T System III Unix code base and heavily redone as a research project at UC Berkeley mostly coded by Bill Joy, Eric Allman, Michael Toy, and a zillion other grad students who were there from about 1975 to 1984. BSD/OS was the commercialized version that used to be from BSDI, which got bought up by Windriver Systems. Got rewritten linux-like into NetBSD, OpenBSD and FreeBSD among others. "Berkeley Systems" were the great folks that brought you the classic screen saver "After Dark" but didn't have anything to do with Unix....

[alexander]

Well, to start, there really arent that many ways you can write the kernel, and many pieces of Linuses code were similar to that of Thompson and Gates.

If you really want to go deep into Unix, than you also missed a step also:
MIT, AT&T and GE were working on an OS for one of their mainframes, I think it was called multis or multics or something like that, i beleive with the main goal to create a secure operating system, but they had a problem with performance so everybody split (but there were releases).
Thompson wasnt one of the greater programmers, but he wrote a game after that (split) that ran really slow on the mainframe and decided to rewrite it in assembly which gave him enough experience and confidence to hack up the multi(s, cs or whatever) to be able to multitask and have a command line, something rare at the time i should probably add. Its only a bit after that that it actually became known as unix. but it only ran on one type of a machine (as it was written in assembly), so in mid 70s unix was rewritten in C to add the ability to easily modify code to be ran on other hardware systems.

I agree that old code is not rewritten, there were real programmers back then, that knew how to propperly program, although i havent read the books you proposed, I've read quite a lot about good programming, both style and habbits, as well as algorithm design and other things. Old code is much better than most of new code that is produced. The old ways to code are mostly gone, sadly, i mean look at Windows, I bet programmers at UCB whould have had good times going through all that code, lots of jokes and laughter would have been heard anyhow...

Mebe yur not old enuf to remember Y2K

yup i'm 3...

"The year 2000 problem (also known as the Y2K problem and the millennium bug) was a flaw in computer program design that caused some date-related processing to operate incorrectly for dates and times on and after January 1, 2000. It turned into a major fear that critical industries (electricity, financial, etc.) and government functions would stop working at 12:00 AM, January 1, 2000, and at other critical dates which were billed as "event horizons." This fear was fueled by huge amounts of press coverage and speculation, as well as copious official corporate and government reports. All over the world companies and organisations checked and upgraded their computer systems. The preparation for Y2K had a significant effect on the computer industry." -Wiki-

[Buffy]

Originally Posted by alexander

Well, to start, there really arent that many ways you can write the kernel, and many pieces of Linuses code were similar to that of Thompson and Gates.

Very good point! [Clarification: Bill didn't write the kernel! NT was designed by David Cutler who was the architect of VMS...I'm sure even Bill himself wouldn't take credit for it.]

Originally Posted by alexander

If you really want to go deep into Unix, than you also missed a step also: MIT, AT&T and GE were working on an OS for one of their mainframes, I think it was called multis or multics or something like that, i beleive with the main goal to create a secure operating system, but they had a problem with performance so everybody split (but there were releases).....

Yah, "MULTICS." Unix was a play on it which couldn't be used by Bell Labs cuz it was owned by the folks who paid Ken to write it, but the main point of both was a shared resource, multi-user OS (pretty novel in 1969!)...

Originally Posted by alexander

Thompson wasnt one of the greater programmers....

Ken would be modest enough to agree. One of the reasons his next OS was "Plan 9" named after the Ed Wood film...

Originally Posted by alexander

I agree that old code is not rewritten, there were real programmers back then, that knew how to propperly program...

Ha! We used to have contests to see who could write the most obscure single-line C programs...but I say too much...

Originally Posted by alexander

although i havent read the books you proposed, I've read quite a lot about good programming, both style and habbits, as well as algorithm design and other things.

Ya really should. Honest. Its good stuff! Donno what I'd do without my copy of Knuth! Too many people reinvent algorithms that they could pick right off the shelf....

Originally Posted by alexander

Old code is much better than most of new code that is produced. The old ways to code are mostly gone, sadly, i mean look at Windows, I bet programmers at UCB whould have had good times going through all that code, lots of jokes and laughter would have been heard anyhow...

We have! Its a hoot! BUT I will tell you that I worked for a company that was spun out of a CS project at Berkeley and the first thing we did was to REWRITE the ENTIRE thing. Grad students being paid almost nothing actually don't write the best code all the time!

Originally Posted by alexander

yup i'm 3... Y2k

I did have a client (and there were lots of reports of others) who had no source and only the machine code for a 30 year old application in 1999... Not pretty!

Warning: don't take anything I say too seriously....

Cheers!
Buffy

[alexander]

Bill didn't write the kernel

ok, ok it was Paterson who wrote MS-DOS...

We have! Its a hoot! BUT I will tell you that I worked for a company that was spun out of a CS project at Berkeley and the first thing we did was to REWRITE the ENTIRE thing. Grad students being paid almost nothing actually don't write the best code all the time!

i would still think that the ratio is waaaay off, the industry standard for commercial software right now is about 20 - 30 lines of buggfull code per 1000 lines of code, i think that back then the ratio must have been a bit smaller than that... BTW 2.6 kernel contained 985 bugs in its 5.7 million lines of code when it was released, and that was a while ago, WinXP i estimate to contain minimum of 800,000 (based on 20/1000 lines on the scale of 40 million lines that meke up XP)

Ha! We used to have contests to see who could write the most obscure single-line C programs

and now that tradition had transformed itself into who can write the most obfuscated Perl code, or best game in 20 lines on DarkBasic, or most obfuscated C code and such...

Ya really should. Honest. Its good stuff! Donno what I'd do without my copy of Knuth! Too many people reinvent algorithms that they could pick right off the shelf....

I think reinventing the wheel should be against the law! Although new solutions are sometimes great, most of the time, the greatest way to solve an algorithm has already been created by somebody else. I would like to have time to read, but i dont have it, got my programming projects, programming, networking, program, Linux and everything in between manuals, college work, college projects that will benefit the school in future and stuff. (for example my newest one is how to have linux on all the machines under our departments supervision, but still dont deny students the ability to use microsoft... the solution is vmware, we will have linux be the host OS on all the machines in a labs, have vmware with images of XP for all the classes in that particular room with only the software that the class needs and no privilages to install or remove anything, and then write a short script for xorg to start an appropriate virtual machine after user chooses to start their class from the menu... should be fun, but there are things that need testing and stuff)

Grad students being paid almost nothing actually don't write the best code all the time!

who would unless they want to? if their motivation is money, then they wont, if their motivation is the excitement of coding and the gain in experience (their betterenmentishness (dont even ask me what that is supposed to be) of the code (ooh, i think i thought of a definition, how does "an act of bing betterer(and that is betterer not just better), sort of, like" sound?)), then, and only then they would... (you know we whould have a stickey here with all the new and weird words and their definitions that we come up with when we talk or type something)

sources:
http://www.wired.com/news/linux/0,14...w=wn_tophead_1

[Buffy]

Originally Posted by alexander

i would still think that the ratio is waaaay off, the industry standard for commercial software right now is about 20 - 30 lines of buggfull code per 1000 lines of code

Be careful of these statistics: I know programmers who fill their code with mostly irrelevant comments just to beat the "industry standard."

Originally Posted by alexander

I would like to have time to read, but i dont have it, got my programming projects, programming, networking, program, Linux and everything in between manuals, college work, college projects that will benefit the school in future and stuff.

You're excused....

Originally Posted by alexander

if their motivation is money, then they wont, if their motivation is the excitement of coding and the gain in experience..., then, and only then they would...

Increasingly here in sillycon valley, a strong motivation is proving to your boss that its worth paying you 5 times more not to send your job to India, China or Russia....Its an incredible incentive for writing good code!

Originally Posted by alexander

their betterenmentishness (dont even ask me what that is supposed to be) of the code

I secondontilarily this motion....