Slashdot Mirror

I'm a PostgreSQL contributor. Oracle can't buy my copyright. There are dozens of other code contributors just like me in that regard, working for many companies. It was possible to buy MySQL because most of the code was developed by MySQL Ab, and copyright assignments required for contributions to be merged. See Some thoughts on Copyright Assignment for more details. That's not the case for PostgreSQL.

Furthermore, the PostgreSQL community has already been through the process of having a single corporation "buy" many of the top contributors, when a company named Great Bridge hired many of them. The disruption to the PostgreSQL community of Great Bridge failing was such that even starting in that direction is actively rejected now. So even if a company did start gobbling up developers one at a time, they would face increasing resistance at obtaining the remaining ones.

The usual remedy is for the offending party to correct the fault, which in this case would appear to involve (re)attaching the correct copyleft license to some files distributed to developers and a hefty "donation" to the FSF.

FTFY.

the most money we ever ask for is reimbursement of our cost in doing that enforcement effort

Hmm, who to believe, you or Bradley Kuhn?

AFAIK, the GPL has never lost in court. Every case I'm aware of has settled (because the violator realized they could not win); oh here's one that went to court: http://www.jbb.de/judgment_dc_frankfurt_gpl.pdf

http://www.open-mag.com/features/Vol_66/GNU/GNU.htm
"We handled approximately 50 violations last year. We expect to handle about 5-10% more this year. With staffing levels, this is basically what we can handle. There are many more out there that we could pursue. ... I would estimate that 90% or more of violations are simply confusions that can be cleared up by friendly negotiation and explanation. We start every case assuming that it is simply a confusion to be cleared up.

Of the other 10%, almost all are disregard, which requires careful diplomacy to move violators to a point where they take their obligations seriously. So far, we've been able to do it. A very small number of violations are actual willful, concealed infringements. These tend to be the "big cases" that take a long time to resolve.

Because we've been careful not to publicly admonish GPL violators, many people don't realize how often we have enforced the GPL successfully. ...

We have the right to sue for copyright infringement if we need to. We rarely need to threaten a lawsuit, and we've never had to file one. Most companies realize that what we ask for is not onerous and is easily done. Most companies that find copyright infringement sue for huge sums of money; the most money we ever ask for is reimbursement of our cost in doing that enforcement effort."

http://gpl-violations.org/news/20041004-majorupdate.html
"The netfilter/iptables project did not announce every individual case, but has so far settled in more than 10 cases out of court. Among the vendors are major companies such as Siemens, Fujitsu-Siemens, Asus and Belkin."

You buy Microsoft software, and you buy it once and for all, the cost that we tell you is the total cost for ownership

You don't buy the software, you license it until the next version of Windows comes out and your software becomes incompatible with that, and your computer has virtually no resale value as the software has to be totally wiped else you risk a visit from the BSA. And according to Gartner the TCO for a company to support Windows was $9,784 per anum per computer (1997). You think it's come down in the mean while :)

Some solutions, not tested and not reviewed, but simply googled. I have used other solutions for clients in the past. They have worked well on an IBM systems that was networked. The data wound up on a postgresql database that was then linked through a client application and a web server for different kinds of reporting and control.

These are just a few of the listings off of Google on a search:

• One solution reviewed
• More solutions ...
• More solutions from tech republic...
• An article on Infoworld

Remember though that POS is one of the most important aspects of any retail or store front type business. So, be wary of what you choose. Choosing a POS system is more about the relationship you are building with your system provider and maintainers than the actual system itself. Some systems require less of a relationship, but I have never run across a situation where that relationship was not present and integral to the long term use of the system. That relationship could be with the OSS developers and users or a company. I wish you luck on your choice. And, please post back to here as to what you chose, why and then your experiences. I prefer the OSS way in most cases as I can make modifications where I (my client) need them. But some applications I farm out to a company as the requirements are outside of my experience.

InnerWeb

Wow! I knew linux made a good server, didn't know it was that efficient. Thanks for pointing that out. And I'm only 30, happily living with my wife and drive a Holden Rodeo.

udev is part of the Linux kernel project, while HAL and D-BUS are not.

So, why doesn't Linux have a HAL? I can tell you the answer in one word - Tradition. The Linux kernel emanates from kernel.org, which essentially produces a white box OS, supporting x86/IA-32 compatible CPUs. With that Wintel architecture, things like code compatibility, BIOS, and chipsets come together to form what I call the PC/AT "virtual machine." Linux, like Windows, leverages basic knowledge about this platform, so that booting and hardware initialization are taken care of, leaving a kernel to worry about the more interesting things. As one hacker says, "on x86, it just works!"
http://www.open-mag.com/features/10_02feats/HAL/HA L.htm

If you think that's bad try open magazine I signed up to get their paper copy years ago and they helpfully signed me up for their email list as well. The unsubscribe link only pretends to work and I got so desperate I set my mails server to bounce anything coming from them. A year later I removed the filter and two weeks later I got my monthly email from them.

Just didn't expect that from someone claiming to support Open Source.

http://www.open-mag.com/features/Vol_101/MySQL/MyS QL.htm

"And they probably use MSVC. Everyone uses MSVC. They'd be on crack if they didn't."
Why? The Intel compiler seems to in general generate better code then MSVC and seems to do much better at vectorization than MSVC or GCC. The Intel compiler plugs right into Visual Studio so I would say that not using MSVC can be a good choice. I would have to say that to not consider the Intel compiler as a option for game development under windows would require someone to be ignorent or an employee of Microsoft.
Here are some older benchmarks http://www.open-mag.com/features/Vol_27/IntelC6/in telc6.htm
I am surprised how many people just stick with MSVC when they have a better performing option. For some it maybe the added cost. I mean if you already have paid for Visual Studio why pay for the Intel Compiler. However for a game developer I would think it would be worth the effort. Or it could be that the Intel Compiler does not generate good code for the AMD cpu.

Unfortunately, this will probably mean OpenLDAP will fade into insignificance, but I may be wrong !

Why? KDE vs gnome? freebsd vs linux? It's not going to dissapear...

Same happened to QT vs Harmony if anyone remembers (Harmony died when QT went for GPL if I remember)...

The GPL is a COPYRIGHT license. In the USA it gets its power from 17 USC 106 (Copyright law) and since it makes no mention of patents or trademark, it has no power over those (in granting or restricting rights).

In at least one case, it was explicitly determined that the GPL license is not a trademark license.

Indeed. And don't forget:

5. The single file mentioned in 1. can't grow above 16 Gb with the standard edition of Microsoft Exchange. (See also 2.)

Of course, I haven't used OpenExchange yet so I can't really comment on that. I tried hunting round for a few reviews, but didn't come up with much. Here they are anyway:

eweek
pcmag
OpenMag

There is a mention of a "downside" being "lack of a spellchecker in the web client", but of course modern browsers like Konqueror have this built-in anyway.

There is also mention of the web client not being as "feature rich" or "refined" as Microsoft Exchange's, but without any actual qualifications.

Basically, from the reviews everything seems great apart from the backup aspect. This is only a downside because their is apparently little guidance given.

Perhaps you should check out this article (Journaled filesystems on Xeon) from Open, an e-zine covering open source and Linux. It takes a more scientific approach to benchmarking filesystems.

In a recent interview, Jody said a W32 port was the priority. I think that could actually start pushing it over the top and make some real headway, I can see why it would be a priority.

Take a look at this article and this one. They cover perfromance of Intel's C compiler and gcc.

Take a look at this article and this one. They cover perfromance of Intel's C compiler and gcc.

Their case is a simple contract dispute between themselves and IBM.

Yes. SCO is contending IBM breached their contract relating to who owns the code and who is able to release it to others. SCO contends that SCO owns the code and IBM was wrong to release it.

No, their case is NOT based on having the right to license the code.

Yes, it is. SCO is maintaining they are the owners of the code based on the contract, so they have the rights to license it (along with all other Unix code, evidently) and IBM can not give it away to the Open Source people.

The counter-suits, on the other hand, ARE based upon the fact that SCO does NOT control UNIX

Again not correct. The counter suits don't address SCOs licensing rights TO UNIX, they ARE addressing their rights to re-license code that has been already released under the GPL - as you stated.

But the suit isn't about that. It's a contract dispute with IBM over terminating Project Monterey. (my emphasis added to your posting)

Yes it is a contract dispute, and it may have been initiated in a fit of pique of the termination of Project Monterey, but it is disputing IBMs right to release code that SCO claims rights ownership over.

As for the GPL's validity in court, the courts have decided in its' favor.

Thanks for the link, but here they specifically state that that case was about 1) trademark infringment and 2) GPL issues. "The judge ruled that NuSphere can't market products under the MySQL trademark. Saris declined to get into the complexities of the GPL."

Hardly "decided in its' favor", at least in my opinion.

MySQL AB even wrote a thank you letter to the FSF.

Remember that the GPL has already been tested in court, and won?

• CLUSTERING IN TUNE WITH APACHE AND MYSQL (Free registration might be required. Also see Emic Application Cluster for MySQL
• InnoDB Hot Backup (with point in time backup)

MySQL/InnoDB-4.0.1 and Oracle 9i win the database server benchmark of PC Magazine and eWEEK. February 27, 2002 - In the benchmark eWEEK measured the performance of an e-commerce application on leading commercial databases IBM DB2, Oracle, MS SQL Server, Sybase ASE, and MySQL/InnoDB. The application server in the test was BEA WebLogic. The operating system was Windows 2000 Advanced Server running on a 4-way Hewlett-Packard Xeon server with 2 GB RAM and 24 Ultra3 SCSI hard drives.

eWEEK writes: "Of the five databases we tested, only Oracle9i and MySQL were able to run our Nile application as originally written for 8 hours without problems."

The whole story. The throughput chart.

To our knowledge, this is the first time a computer publication has published database benchmark results tested on the same hardware since PC Magazine did so in October 1993.

Two other MySQL products I found interesting (neither of which is open source at this time):

• CLUSTERING IN TUNE WITH APACHE AND MYSQL (Free registration might be required. Also see Emic Application Cluster for MySQL)
• InnoDB Hot Backup (with point in time backup)

MySQL/InnoDB-4.0.1 and Oracle 9i win the database server benchmark of PC Magazine and eWEEK. February 27, 2002 - In the benchmark eWEEK measured the performance of an e-commerce application on leading commercial databases IBM DB2, Oracle, MS SQL Server, Sybase ASE, and MySQL/InnoDB. The application server in the test was BEA WebLogic. The operating system was Windows 2000 Advanced Server running on a 4-way Hewlett-Packard Xeon server with 2 GB RAM and 24 Ultra3 SCSI hard drives.

eWEEK writes: "Of the five databases we tested, only Oracle9i and MySQL were able to run our Nile application as originally written for 8 hours without problems."

The whole story. The throughput chart.

No, the results are NOT valid, since the back-end code generators on the two compilers are different (GCC/x86, or "i386" as they call it, and whatever they call the PowerPC backend).

It's widely known that Intel C/C++ compiler and even Visual C++ produce far superior code on x86 than GCC - see, for example, this page.

It appears from my googling this space is dominated by crappy software that requires the use of IE for the application and only run on Windows. This and the Intuit offerings are prime examples of places where windows is being used where it had better not be used, and poorly written software is requiring data which should be well-secured to be placed in an environment where security is eliminated. Indeed it might bear looking into from the various developers here. If someone can save these guys money and give them a proven solution, they will be rather rich and will have done a good thing.

I know about Gnucash, I support its development, and hope it eclipses all Intuit has to offer. However I see the parent is right w/r/t Insurance Agency software, and I would suspect travel agency software is still as scary and abysmal as it was years ago when I looked into it.

It may not be Mac OS X or Darwin-specific, but it seems that Intel is working with Wasabi Systems/NetBSD to work on getting iSCSI to run on NetBSD. Some of this may work up to Darwin, then into Mac OS X... but it will take a while, mostly when Apple is focusing on Fibre Channel with their XServe RAID units.

Whenever you hear "TCO study", think "FUD". I'm in IT in a large corp and regularly have to dissect some inane TCO "study" which supposedly proves that $VENDOR_A's product has a much lower TCO than $VENDOR_B's.

I have never, ever, seen a TCO "study" that is worth the paper it's printed on, or that even remotely applies to the particular circumstances in our company. Every single one was, to paraphrase from this interesting article, every bit as accurate as Enron's books.

Basically, TCO studies are marketing intruments: You pay a hefty sum to a "neutral" consulting firm and, in return, get a paper that says your product has better TCO than the competition's.

The problem is that the management layer with its skewed perception of reality is only too willing to grasp for those numbers and quote them as if they were hard facts. Once, when a PHB wanted to use the Gartner NT TCO figures in a presentation, I kindly reminded him that these figures were based on a company which had nothing whatsoever to do with ours, that they left out important but hard-to-assess TCO-relevant factors, and that they came from Gartner Group, who once predicted that OS/2 would rule the world. I got my point across.

For a 25 foot fall with (nearly) no drag, the drive will get up to a speed of 40.0 ft/sec (27.3 MPH). If the drive stops over a 1/8" distance, with -uniform deceleration- (this is pretty generous for a fall onto concrete), this equates to 1600 G's. Halve the distance, and quadruple the force. Decelerate it in a non-uniform fashion (as it realistically would) and you'll get even more spectacular results.

See this review of a hitachi drive. Note that they say a drive designed for a non-operating shock of 800G's can take a fall of -one foot- onto concrete. I destroyed a maxtor by dropping it 3 feet onto carpet in a past life, and I'd suspect it was rated for a non-operating shock of at least 50G's.

I'd love to see you try it with your drive with your valuable data sometime though.

The infamous Marx Brothers -vs- Warner Brothers case comes to mind.

So in other words, they created a new executable format (instead of Windows emulating Linux or visa versa, an much more useful goal), which will work on Linux/x86 if and when Linus ever adds support.

Yes. And Linux didn't support ELF until Linus added support. But being that ECMA335 is public, we won't have to reverse-engineer support, we'll simply have to read the standard.

Actually we won't. Miguel will do it for us with Mono. And you're right about Windows/Linux emulating Linux/Windows - because Wine and VMWare are *so* fast.

It's been doable in GNU C or GNU Ada for a long time now. It just needs a recompile. And there's all of what, 5, x86 OS's out there. Compile for x86 Linux and you will be able to run on most x86 BSD's anyway, so you only really need to compile it twice.

I guess you missed that article where Intel and MS VS.NET beat the crap out of gcc.

I am not 100% sure of this, but endianness is the only portability issue I've seen with the executables...and given that it's just a simply matter of inserting a bunch of bswap instructions in the assembly stage of compilation, it most likely won't be an issue by the time .NET and the ECMA standards are complete, so it really isn't an issue.

Frankly, I hope not to be running x86 in a few years. Either Itanium, or Sledgehammer or maybe some nice PowerPC hardware. Nothing you've suggested is in the least portable to non-x86 hardware, unless you're willing to emulate the x86, which puts us back to where we started with bytecode.

You can buy an Itanium now...it will run 64 bit XP or Linux...oh wait, is that Linux ia64 port done yet???

Not necessarily. Recall from this slashdot story about this article the intel compiler also showed similar results over GCC when targetting the Athlon.

GCC's mission statement is not the running time of executable code, we've recently been having a thread about it on the plan9 mailing list (or comp.os.plan9). (although ours started as a flame from Thomas Bushnell that plan9's 8c was nothing more than a "cute toy" - 8c is more concerned with compilation speed than execution time where it beats GCC hands down, if you want raw execution speed look elsewhere).

It could well be that Intel's compiler will show similar performance gains over GCC on the Hammer.

I wonder if every problem will start to look like a nail when the hammer claws it's way out of the AMD tool box.

There are more important things than Windows vs. Linux.

It's not relevant to clustering, do but recall the story in Open Magazine (also discused on /.) a couple of weeks ago reporting that on a single machine, numerically intense code ran about 10% faster on Windows XP than on Linux (compiled with the Intel compiler in both cases).

This speedup is probably irrelevant next to questions of efficient distribution of processing power, but is worth knowing about.

More significant is the finding in the same story that numerically intense code compiled with Intel's compiler ran 47% faster than when compiled with gcc, so you should be looking at getting your hands on some funding to buy the Intel compiler rather than worrying about MS and Windows.

Better compiler optimizations could be worth the price of many nodes in your cluster!