Slashdot Mirror
Brew your own SPARC: SPARC IP Core SCSLed
Tekmage writes "Sun has just announced the release of it's SPARC IP core under their Community Source License. " The dialogue over whether or not the SCSL is a good license continues, but it's better nothing, IMHO. Interesting move on their part, especially given IBM's recent moves with the PowerPC designs.
Finally a company, Sun, is doing exactly what hackers have been demanding all these years: letting them have a look at interesting technology, to learn from and satisfy their curiosity.
This isn't about open source evangelism and your imagined right to free computer hardware. It's about giving students and other interested people a tool to learn from, in a way that doesn't hurt Sun's business.
Nobody loses anything from this. People interested in microprocessor design gain. Why are you complaining?
If you expect Sun to license people to compete against it with it's own technology, you're living in a dream world. It's not going to hurt the GPL to have more information available in the world, and with your hypothetical binary choice between GPL and total secrecy, most companies would choose total secrecy. Be glad some have chosen to imagine a third alternative that *is* much better than nothing.
Even long before this, I never liked Sun Microsystems very much. Aside from obvious exceptions such as Perl, I never cared much for programming languages that are constantly getting updated. That's what is supposed to happen to applications, not programming languages (of course, Perl is a C application.. ha!), for the most part. Sure, when C++ first emerged it wasn't finalized. But then again, you aren't going to see C++ 1.0, 1.1, 1.2, 2.0, and 2.1 within the next year or so. It has a standard. Java just keeps growing.. and growing.. And why is it growing still? Because when it was first released it wasn't worth a whole lot. It was not a finely developed language. The most obvious point being its extremely poor performance issues (what, write once, -walk- anywhere?).
That was strike one. Strike two is probably even worse. A commercial programming language trying to define itself as a universal standard? Conceptually, Java really rocked.. But the finer details of reality stopped it almost dead cold (all hype aside). However, even if it had lived up to expectations, who would really want to have to pay licensing fees to Sun for making commercial applications with Java? I'd rather just stick to C/C++ and Perl and not worry so much about licenses.. If the only license I ever have to agree to is the GPL I'll die a happy man.
And then there's the most annoying thing about Sun, the very thing they are known for: Java. I could have just left them alone, never said a bad word about them.. just ignored them and went on my merry way.. but they have to prepend the word Java to, well, everything! Am I the only person who gets sick of JavaWhatever products being a quarter a dozen? This company is known for its crazed marketing and over-capitalization on key products more than for anything all that useful to the world at large, unless you actually like their horrible licensing agreements.
~ Kish
Someone should write up a License Generator webpage.
Have it pick an acronym at random using a database of buzzwords ("open", "community", "free", "public", etc) and have sliders for the level of openness you want, of protection for your source, your patents, or whatever else it is that people write up their own licenses for...
Intel also has the virtue of not claiming its software is Open Source when it isn't. I don't mind pure commercial software, the world has a place for it as long as the commercial vendors don't try to restrict the development of free software. I do mind when people pose their product as Open Source when it's clearly not Open Source at all.
Thanks
Bruce
Bruce Perens.
The Community source is not "better [than] nothing." Given that it's not completely open:
1) If people adopt and develop under SCSL, Sun has no incentive to open the license further.
2) If people don't adopt SCSL, Sun is likely to drop further free/open source involvement.
Either way, SCSL is a bad thing.
Whilst more openness from previously closed companies can never be a particularly Bad Thing, I question Sun's motives - how many people are going to have the time, ability and resources to contribute to this?
Is Sun hoping to benefit more from the traditional advantages of Open Source (strong peer review, new viewpoints providing new enhancements) ? Or is this move intended to attract more press to Sun's recent announcements, and form more of a statistic for future Sun comments on their commitment to Open Source?
PenguiNet: the (shareware) Windows SSH client
Bruce
Bruce Perens.
The point is that if we want to influence Sun, we can do it by not supporting them as enthusiasticly as we might otherwise. Voting with our feet, in other words.
Thanks
Bruce
Bruce Perens.
HP asked for real Open Source
Sun stopped short of Open Source
HP decided that stopping short of Open Source didn't give them a reason to cooperate with Sun.
Can't you read it that way?
Thanks
Bruce
Bruce Perens.
And Linux is just a pawn in the game.
Bruce Perens.
Up to that point, if you built a computer you had to write your own custom OS - in assembler. That's not a garage operation. But now along came UNIX:
You were supposed to have a source license. But the source code circulated freely (due to a bunch that had been handed to universities) and Mama Bell didn't bother with you until you were ready to sell it, and then didn't hold any grudges for your "illegal" use of the source when it came time to price your license. The kernel was tiny and easy to port - and mostly in a compiler language yet! So a whole new model of OS construction became prevalant:
Suppose you've got your new box (or a design for it and a prototype coming together), with it's new processor and new peripherals.
First: If it's a new CPU, add a code generator for it to the Portable C Compiler (PCC). Compile once to get a cross-compiler (to use on your development platform), then use that to compile again to get a native compiler (to run on your target, once it's up).
Second: Port (and configure) the kernel. You probably have to modify the memory management code, the task-switcher, and the raw disk and console driver. Write drivers for any new-fangled peripherals (though you can probably modify them from stock stuff, too) - but that can wait 'til you're running native.
Third: Port the ROM bootloader. (It uses the drivers you already ported, above.) Burn it into a PROM. Plug it into the lab box.
Fourth: Compile all the utilities with your cross-compiler and build an initial root disk - using your current UNIX platform to write it.
Fifth: Plug the disk into your new box. Boot up. You're live. Debug and expand on your shiny new lab box.
Sixth: Call up AT&T and negotiate price of a license to distribute this puppy, after you got it to work so there's no longer any risk.
Seventh: Show the vulture capitalists your working shiny new box. Get your working capital with a high valuation on your company (so you still own most of it).
Eighth: Build it and ship it.
This you can do in a garage shop (or as a grad school project). And it happened just as a couple decent microprocessors hit the market, too. So there was a decade or more where dozens of UNIX boxes, on diverse platforms, sprang up like weeds.
Looks to me like Sun wants to use the same model with the SPARC CPU core, to penetrate the ASIC market (which MIPS and ARM currently dominate). They're making the SPARC processor core free to the shoestring fabless-chip-house startups, during that difficult design period when they're still hanging by their shoestring.
First bag is free, dude!
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Yes, it probably did, but then it was almost certainly using either a 60 or 70MHz chip (they did a 110MHz version, and later a 170MHz version based on the TurboSparc, but they weren't veyr common). From personal experience, a Sparc will feel about that same as an Intel with 1.5 to 2 times the clock speed. This is probably mostly due to the on-chip cache (even the 60MHz MicroSparc II had 1MB of onboard cache).
"The invisible and the non-existent look very much alike." -- Delos B. McKown
The thing is that Sun, like any corporation of that size, is not a single entity. The hardware side of Sun, after an initial shaky start, have been quite helpful to the Linux community, even going so far as lending developers high-end hardware and providing technical info so that Linux could be made to work well on it. See starfire.c, somewhere under arch/sparc64 (sorry, I dont' have a kernel tree to hand to check the exact location), to see the code for supporting Sun's Ultra Enterprise 10000 (high-end server, up to 64 CPUs -- we've got 4 of them here :-)
"The invisible and the non-existent look very much alike." -- Delos B. McKown
--Jamin Philip Gray
jamin@DoLinux.org
Celebrate the finer things in life
Reminds me of the ALSA project slamming Creative for not open-sourcing its drivers....should we really be critical of the companies who won't open source their code when they are building drivers, or should we concern ourselves with those producers who won't even acknowledge Linux at all?
Funny and I thought Perl == Paid employment recently located
Of course, now that I think about it, the subject of this post makes me wonder what kind of half-assed programmers these "weird licensing people" really are. However, back to the point..
One of the few licenses most of us really trust is the GNU General Public License. Especially in light of things like the Netscape Public License. Most people use the term "open source software" in the way Richard Stallman talks about "free software", but they're not exactly the same thing. Just because it's "open source" doesn't mean the software is free (liberated, whatever.. I'm talking about freedom here, ok?), as licenses such as the NPL makes quite clear.
These companies aren't very likely to gain the trust of the free/open source community if they continue to develop spin-offs of the GPL that often end up trying to screw over contributors.
~ Kish
They made the announcement back in March. I remember talking to one of my professors about the research & academic opportunities it presented. /03/02/1133216.shtml
http://slashdot.org/articles/99
Christopher A. Bohn
cb
Oooh! What does this button do!?
Sun's license may be better for hardware than software, since hardware doesn't "want to be free" :).
The MicroSPARCII is not a particularly exciting processor. It's the chip that was used in the SS5, and it ran up to 170MHz in the last SS5 offered. It's not designed for SMP (Turbo- and HyperSPARCs did that in SS10 and SS20). It's a 32bit chip, a bit faster at floating point than an equivalent Pentium. (it's also the chip in my SPARCBook 3 :).
:) But it does on the ARM, Motorola and MIPS chips too.
So don't expect cheap high powered crazy Suns floating around soon. Sun wants people using MSII's where they're now using R4xxx's and ARMcores and m68k's in PDAs.
An advantage is of course Linux already runs fine on it
Microsoft estimates that 65% of NT crashes are due to third-party drivers. Why is this significant? In Linux, there are VERY few non-GPL'd drivers, but if the trend increases, then there will be a LOT of binary-only drivers for Linux making it almost as unstable as NT.
No driver at all may very well be better than one that is not open source, as it prevents people from developing their own GPL'd drivers, which will work more correctly and be more stable in the long run.
Unfortunately, I think it's one doomed to failure.
There are two main reasons why Open-Source is practical in software. One, software is available in infinite supply; I could make as many copies of a given program as I wanted and always have one more, and furthermore I can do it at zero cost. Two, software is pretty easy to modify; all one needs is a compiler (which GNU and the EGCS team have given us already) and source code (which is readily available). Because of these two facts, anyone with a computer can get into the business quickly and cheaply.
Hardware doesn't have these two attributes. First, it's not available in infinite supply; if I have one chip and give it away I have no chips. I can, of course, make more if I have the right machine. The problem is, the cost of said machine (several million dollars, last I checked, and I don't seem to be able to find any of them on EBay for less) keeps pretty much every man, woman, and child on the face of the planet from getting one. Even if I have the machine, I still have to buy the materials, which gets expensive if I want to make many chips.
Second, hardware is very difficult to modify on the level of the individual chips. Many people on Slashdot probably built their computers from preexisting parts. Some probably have managed to build one from preexisting chips and constructing even the boards themselves. But I'd love to see someone here running Linux on a chip that he or she made as a Computer Engineering project in college (granted, such courses do tend to include constructing a simple microprocessor as a final project, but now try making a whole computer out of it). Besides which, EPROM's and EEPROM's notwithstanding, one cannot modify a chip which has already been made; you must literally throw it out and start again if you want to change the chip.
These two major factors are going to keep the idea of an Open-Source processor from being truly feasible. It's not that no one will work on it (a fallacy often used as FUD against Open-Source software). It's that almost no one is able to work on it (certainly not enough to derive much of an advantage), and those that are typically already work for a chip company which is going to take a very dim view of an employee who's helping out someone else's chips, so in the interest of job security they aren't going to work on it either. Opening the specs is still a Good Thing from a trustworthiness standpoint, but I seriously hope Sun isn't hoping to get the next generation of Sparcs this way.
'Build your own SPARC' in an absolutely ANCIENT issue of Byte I have lying around somewhere. I seem to recall it also had the first ever local bus PC, a discussion about OOPS vs DDE, a bit on Unix fragmentation and a sort of SGML tutorial. I'll have to dig it out and read it again.
-Yarn - Rio Karma: Excellent
The SCSL requires that you get a license from Sun in order to do a commercial redistribution of the product in question. For software, this is kind of okay because software can be redistributed for free pretty easily. It costs just about nothing to put up an FTP server somewhere.
On the other hand, to redistribute hardware you have to setup an assembly line and distribution chain at considerable expense. I don't see how this can possibly work under the SCSL. As soon as you want to recover the cost of manufacturing your modified hardware, WHAM! you have to pay sun a (hefty) license fee. Think of this as a "open source" license which is open until you compile the program, because that's exactly what it is.
It remains to be seen whether even GENUINELY open sourced hardware will work -- I can't see any way that this will work.
Amphigory
-- Slashdot sucks.
I obviously shoulda checked my references rather than do this off the top of my head....
:)
... that would be interesting. I wouldn't mind an Ultra based subnotebook :)
I'm only saying: I don't see what advantage microII has in embedded applications -- I don't see a particular incentive for embedded chip designers to pick it up.
Does anyone know what its power needs etc. are? Have there been advanced, low-power versions of this chip produced? The microII 85MHz in my SB3 is pretty power hungry
Now an embedded Ultra
At least someone in the industry catches that.
I found out that Sun did say the SCSL was an Open Source license at the StarOffice press conference. Of course, that's a bald-faced lie. A reporter who was there, and seems to be a responsible person, asserted that fact to me in private mail. I'd like to know if they did the same thing at this most recent press conference.
I wonder if they're just trying to buy the idea of Open Source by releasing so much almost-Open-Source software that they confuse people into believing that what Sun does is really Open Source?
I personally am not going to have anything more to do with Sun and its products while they insist on foisting the SCSL on the world. I'd suggest that people who maintain GPL-ed SPARC port of Linux and the SPARC port of GCC consider if they are really helping the cause of free software.
Thanks
Bruce Perens
Bruce Perens.