Why do you say that the laid-off dot commers are Randites? Certainly many techies are Libertarians and many dot commers are techies, but that doesn't seem a strong enough linkage. Sacrosante aside, since he turns out to be non-representative, there seems to be a misperception about the dot-com boom/bust. The dot-coms were not generally full of techies. The most extravagantly doomed dot-coms gorged themselves on marketing and sales folk. They were not likely to be Randites - more likely fuzzy liberals in typical San Francisco style. Not that I'm claiming a belief in Ayn Rand saved anyone from the axe - on the contrary, when a company sank everyone was laid off. If I really thought any talented techie was homeless out of sheer pride and refusal to accept aid, I'd have great respect for his choice and confidence in his future success. However, it all sounds very much like an urban legend.
It's a nice idea, but I suspect it wouldn't work. It's like free coooperative DNS - too tempting for abusers. I suspect that if Freedom takes off, Abuse will be their biggest cost center. The service is a natural haven for crackers and spammers. Hobbyists might enjoy setting up the FreeFreedom servers, but I doubt they'd enjoy chasing down and disconnecting abusers. Besides, in a free anonymous system there is really no such thing as disconnection. In fact, netblocks housing such servers would very likely end up on the RBL, never to be removed (until our ISP's TOS us).
They were in a strong position to become the linux vendor to many Unix shops and fill the role that HP or Sun would fill. The magic, had they pulled it off, would have consisted of welding the disparate commodified parts of a PC into one shining 'workstation'. Unfortunately they did not do this. I had a VA workstation at one company. I remember when it first arrived - I turned it on and started X. Someting was wrong. The resolution was (I think) 640x480. This was on a machine with a Matrox card. I've never seen a Sun in low-res mode. I don't even know if they're capable - that's the whole point. From the time you shove the install media into a Sun to the time you throw it in the dumpster due to obsolescence, you only see it in the correct video mode. I have no idea what VA was thinking. If I were a naive purchaser, I would have left the box as it was and been somewhat unhappy with the display. Plenty of Windows users run 800x600 after all. A coworker had a VA box with a CD-RW (factory, of course). It never worked very well. After numerous phone calls, VA said there might be a kernel bug causing the problems, and put the fixed kernel on their servers for download. Which was good. The bad part is that I think they shipped this system not only without testing it, but without ever testing one like it (thus the late discovery of the bug). VA had all the chances to offer a Macintosh level of integrating and testing. They controlled the OS and hardware. I don't think they delivered. I should also mention the multi-week lead times for standard workstations. All in all, a bad vendor and the only reason they were used is their support of popular web sites. Which was not very persuasive with management.
Unpleasant? Of course. Like it or not, our modern era is largely rooted in WWII. The Nazis define a certain pole of human behavior - in many contexts they're invoked as the extreme of authoritarianism. Which is not quite the same thing as comparing someone's trivial actions to the holocaust morally. Godwin's law may have been a good idea at one time (although I doubt it - 'the act of censorship is always worse than that which is censored.') but it kills many interesting discussions now. Hitler really introduced 'total war' and the blitzkreig. The parallels to Microsoft are astonishing, but I'm not sure I could do the subject justice. Especially because of the inequality of their crimes - economic predation versus mass murder. At any rate, I'm afraid you can't change the fact that 'grammer nazi' is a natural and evocative phrase. Slang tends to touch on the most powerful things - God, sex and death. The crimes of the Nazis have made their name powerful in metaphor.
Personally, I'll be quite happy when SourceForge shuts down. The worthwhile projects will be hosted elsewhere, and the marginal ones might be abandoned. I admit that SourceForge provides some infrastructure which most projects wouldn't bother creating otherwise. But here's why I hate SourceForge:
It's full of unfinished (unusable) and utterly vaporous projects. It really pisses me off when I'm searching for a solution to a problem to waste minutes on a SourceForge site, sometimes even downloading a tarball and trying to compile, merely to discover I've wasted my time on someone's empty fantasies. You can assert that as the consumer of software I have no right to expect, etc., but I think it's an act of inconsiderate pollution to spread useless software.
Before SourceForge, project web pages were both more usable and more informative of the character and status of the project. More usable because less cluttered - Open Source project web pages are generally some of the best on the web for clarity. More informative, because a half-baked toy project will get one hastily-written page, if that, while a mature, heavily used project will have many pages and current release news.
I shudder to think what Microsoft's PR folks could make of SourceForge. If they want to convince an executive that Open Source is ridiculous, they just have to let him browse this site and take in all the 'discussion forums' with two posts, all the ambitious projects with no release info. It's an ongoing embarassment.
Anyhow, for those who love the enhanced functionality of SourceForge, the software is free and can be implemented elsewhere. Hopefully in a post-SourceForge world it will implemented only when a project has outgrown a simple static site. We're in no danger of reverting to UNC FTP archives. (Not that there's much wrong with them.) Pretty much every ISP offers free static hosting to its customers. Then there's angelfire/geocities for the cash and bandwidth impaired.
Everytime you speed, you run the risk of killing someone.
I remember people like you from the 70's. "Drive 55 - it's not just a good idea, it's the law!" Driving 75 was so horribly dangerous that it was nearly a crime against humanity. Now that the 55 mph speed limit is gone, I don't seen anyone saying "Drive 55 - it's not the law, but it's a good idea." In other words, their concern for safety turned out to be merely a slave's concern with obeying his master's orders. You claim that 'speeding' is unsafe. If a man drove 70 mph every day down the same highway from 1970 to today, it would have been 'unsafe' in the 70's and 'safe' today because we've raised the speed limit.
Most traffic signals use induction loops to detect cars. Essentially an AC signal is sent through a wire loop in the pavement. Putting a big hunk of metal on it changes the impedance. Nothing to do with weight.
That may be your common sense interpretation, but it's not a universal rule. As it happens, you're right about California law. But the reason the law exists is that employers were previously claiming (and compelling employees to grant) ownership of inventions made at home without the employer's resources. Now California employers make employees sign agreements designed to bypass that law by putting the burden of proof on the employee to show that absolutely no company resources were used.
That was a pretty cool response. I was halfway in agreement with the control freaks here - I can certainly understand the fear of having trojaned boxes behind the firewall. But Universities continue to show that openness is possible. My question in these situations is always, "Why do all the machines have to be at the same trust level?" Or to put it differently, maybe it's time to rely more on host-based security and less on firewalls. Given a big enough site, there must always be hostiles behind the firewall. So why not put the desktops on their own network behind a different firewall from servers? Let them infect each other. Of course, even if you completely distrust the desktop machines (best way IMO) it would still be upsetting to have SS7 on them capturing every password. Maybe NSA's trusted linux will solve this stuff.
How dare you pit your puny facts against the mighty river of ignorance that is slashdot? Might as well give this one up, I think. The slashmasses are firmly convinced that M$ is somehow based on BSD code. Whoever invented this meme is a genius because it is unkillable.
I know two techies who are socially inept. Both became unemployed in the crash. One of them is still unemployed. He has gotten two jobs, and in both cases left within a month due to personality conflicts. He is used to being treated as an elite coder code and doesn't like sitting in meetings with idiots. He is also quite frightening to non-techies - streaks of the BOFH. The other guy has a job largely because I hooked him up. I'm not sure how long he'll keep it, although he's doing a very good job. The thing is, he doesn't know how to schmooze his manager at all. I'm not talking about utter corporate ass-kissing - I'm talking about the ability to give your manager more than a blank stare or a dissertation on the latest Linux kernel. Anyhow, it's true that the end of the bubble spells disaster for some talented but unsocial techies.
...they were sueing the kernel hackers because they believed that since the people working on the 386BSD had seen the Bell Labs UNIX source, they were "tainted" with IP and could not legally create a clone.
I think USL, the AT&T spinoff that owned Unix, had more solid grounds than that - BSD contained AT&T code.
A lot of developers misunderstood the trial...
I don't think they misunderstood it. I think they realized that the future of the BSD's was in serious jeopardy. We're very lucky that BSD survived the suit.
It was only by a fluke, AT&T sueing, that Linux ever survived at all.
I have a different interpretation. The lawsuit wasn't a fluke - it was a direct consequence of BSD Unix being built on someone else's proprietary intellectual property and competing with the originator. This is why the GPL is important. Software must be unencumbered by intellectual property claims if it is to have long-term credibility. We need the foresight and caution to avoid licensing tr aps that invite us to contribute our time and energy to the sole benefit of some IP holder.
AT&T lost...
USL didn't exactly lose - they reached a settlement which compelled UCB to strip out the remaining AT&T code.
Anyway, why not upgrade to Win2K? I've upgraded from Mandrake 8.0 to Win2K - now I can run the software I need to *gasp* do my job.
I guess your job centers around the use of 'desktop software' like Excel, Word, etc. Mine centers around programming, with dashes of system administration and Oracle stuff. Moving to W2k would be a downgrade for me. So please realize that for lots of Linux users a Unix environment is essential. If Intel-based Unix didn't exist, we 'd have to use Suns or HP's. "Upgrading" a pickup truck to a sedan only makes sense if you weren't using it as a truck to start with.
You need to use:...1. Asynchronous, Event based socket handling.... Polling is lousy no matter what way you do it. You'll lose most of your performance spent going round a small loop.
Please type 'man poll'. You'll find that poll(2) is asynchronous and event based. Nothing to do with cycling in a tight loop. Which doesn't detract much from your point that the benchmarkers showed no signs of understanding or adapting to the Windows OS.
This is a ploy to strengthen Microsoft's hold on the average users. However, we should still recognize that a content creator has no right to dictate how a user will use his bytes. If you create a web page and put it on the internet, I can download it and
View it in Netscape
View it in Lynx
Derive some data and stuff it in a database (what search engines do)
Run it through a filter that removes words I find offensive
XML is not an alternative to Perl or Java. Those are programming languages. XML is a markup meta-language - a set of very simple ground rules for defining markup languages. It is already very useful. I'm writing an app that receives messages from a custom Windows app. Although the Windows programmers and I hardly share anything in common (they don't know what fork means, for example) we were able to agree on an XML message format with no difficulty. And in case you're wondering, none of us really understand DTD's or the finer points of XML. If XML did not exist I'd probably be asking for messages formatted like RFC822 headers (Key: Value) and we'd run into endless problems with newlines, CRLF etc. For decades programmers have been making ad-hoc markup languages and writing cheesy parsers that work 98% of the time. XML, which has exactly five reserved entities, lets us save a lot of energy and use proven standardized parsers. There is very little to know about XML and it's nowhere near as complex as a programming language. If you've made a web page, you've written something close to well-formed XML. The only difference being that in XML every element must be matched by a closing element or contain a trailing slash. So <P> would become <P/>
Most consumer-grade crypto is pretty trivial compared to these problems
If you're talking about PGP, as you seem to be, you're wrong. A machine capable of brute-forcing a good 128-bit block cipher in reasonable time does not exist. Have a look at http://axion.physics.ubc.ca/pgp-attack.html.
Mac OS X and Windows 2000 are the sumo wrestlers, the Lennox Lewis and Mike Tyson of the operating system world.
Oh, really? Where does that leave OS/390, AIX, HP/UX and Solaris?
live window dragging (where the contents of a window move with it instead of rerendering later)
How impressive. Next thing you know, PCs and Macs will have multiple desktops.
Maybe some Unix geeks and Mac fans need more, but OS X isn't doing a great job with the geek stuff either. You said it yourself: It runs Linux and Unix software "with just a recompile." If it needs to be recompi led, it's not really compatible, right?
Source distribution is the best recipe for widespread compatibility. And I'm sure that someone will package MacOS X binaries of popular free software. Claiming that OS X is less than Unix compatible because software needs a recompile is silly. Unix runs on lots of platforms. Software always has to be compiled for the target platform, whether it's done by the user or someone upstream. A difference between the PC and Unix cultures is that we don't regard compilation as a big deal.
Looks like you missed the part where flikx is facing expulsion and criminal charges. I you had read the entire article, you would know that Flikx does not need "the support of those sponsoring him" - he is quite capable of hosting the site without any help from the University. Someone is facing imprisonment for airing opinions that offended the authorities, and your contribution to the ensuing discussion is a resounding and irrelevant defence of the authorities' property rights. Bravo.
Why don't you put two 12 volt fans in series across 12 volts? Then they'll each get 6 volts, which is close to your initial design. If one of them shorts out, the other will get 12 volts - no harm done.
Of course the tester wasn't trying to isolate CPU - he wanted to test overall hardware performance. And I wonder if there's a fifth variable - how well the compiler was compiled for the host platform. Given a platform X, and a compiler xcc which really optimizes for X, if you compile gcc with xcc it will perform better on the test compilation than if you compile it with gcc. (I'm assuming the test compilation is always with gcc).
That was interesting, but I'm afraid you permitted two variables in your test. If understand correctly, you compiled for Intel on Intel, and for SPARC on SPARC. The benchmarks under discussion were all compilations for MIPS in order to eliminate the effect of some target architectures being more intensive to compile for. Given that your SPARC compilation was slower than your Intel compilation, there are at least three possible explanations:
SPARC architecture is a harder target to compile for, regardless of the compilation platform.
GCC is less efficient on SPARC
The Sun/Solaris machine is less efficient than the Intel/Linux machine.
Use of a constant target architecture would have eliminated #1.
I'll respond to you as one of the many posting similar sentiments. Don't take it personally. You are utterly wrong. The LUG had permission from the event organizers to hand out their stuff in front of the building, as LUG members have made clear in other posts. They were not giving out swag on Microsoft's rented spot, contrary to your assertion. The event organizers permitted several exhibitors to hand out stuff in front of the building, despite the fact (pointed out in other posts) that exhibitors were contractually barred from doing this. They asked (not compelled) the LUG to move simply to avoid unpleasantness from Microsoft, who chose to be unpleasant. My real complaint with you, Jason, is not that you were ignorant of the facts, since the posts to which I refer may have been added after yours (and the many like yours.) My real complaint is the knee-jerk invocation of property rights, the automatic assumption that in any conflict between 'rebels' and 'authority' property rights must be on the side of 'authority'. I see this faulty assumption all the time on slashdot. Yes, the Newsforge article was woefully deficient in facts, but from the skeletal facts it did provide there is no reason to think that Microsoft had any superior legal claim. This was simply a case of a large customer (Microsoft, who rented a ballroom) demanding that a vendor (the event organizer) favor them over a smaller customer (the LUG, who rented a booth). It happens all the time in business. What makes it interesting here is the fear Microsoft displayed.
You are incorrect in asserting that increasingly powerful machines eliminate the need for network transparency. Some of the reasons I like X's network transparency:
There are still many applications available for commercial Unix and not for Linux. X lets me ssh into a Sun and run an app on my Linux desktop.
X lets me access the GUI apps on my workstation from anywhere in the company. If I'm visiting a colleague in a different building, I can easily show him the new program I just installed.
X enables xkibitz, an awesome program that lets multiple users all over the world share an xterm.
Oracle has a GUI installer. If not for X, I would have had to jump through flaming hoops to get permission to enter the data center, round up a keyboard, monitor and mouse, and sit on the floor in a freezing cold room every time I install Oracle on a server. Yes those physical discomforts could be remedied, and the reason they're not is that Unix servers rarely require physical TLC. Anyhow, X lets me do the install from the comfort of my cube.
I've left out the whole "thin client" aspect, because it has been overhyped and may not always make sense. However I think that a really logical IT infrastructure would put an xterminal rather than a PC on the secretary's desk. Every PC is an ongoing cost center, and an invitation to store company documents on un-backed-up disk.
I think that ipf is much more intuitive than ipchains/iptables. It would be very sad if OpenBSD switched to Netfilter, regardless of the license issues. I think it would almost be easier to implement free ipf from scratch. In a way, the hard part was coming up with the conceptual framework. Once you understand ipf syntax, writing the firewall should be less than astronomically difficult.
Yes, the Windows crowd are like children or savages, easily dazzled by shiny objects (shrink-wrapped software) and perpetually dependent on parents/witch doctors (boxware vendors/MS) for the talismans to ward off a scary, complex world. But there are drawbacks to being an adult - we are burdened with the knowledge that the world is held together with duct tape and dried dung. For us there are no happy surprises in shiny packages, only the unerring certainty that software sucks and will continue to suck. With regard to shells, I agree - it's utterly amazing that Microsoft hasn't managed to "innovate" a decent shell yet. But I have used Bash on Windows, and it wasn't pretty. Windows is sluggish in many ways that come annoyingly to the fore in shell interaction.
Slashdot Mirror
Why do you say that the laid-off dot commers are Randites? Certainly many techies are Libertarians and many dot commers are techies, but that doesn't seem a strong enough linkage.
Sacrosante aside, since he turns out to be non-representative, there seems to be a misperception about the dot-com boom/bust. The dot-coms were not generally full of techies. The most extravagantly doomed dot-coms gorged themselves on marketing and sales folk. They were not likely to be Randites - more likely fuzzy liberals in typical San Francisco style.
Not that I'm claiming a belief in Ayn Rand saved anyone from the axe - on the contrary, when a company sank everyone was laid off. If I really thought any talented techie was homeless out of sheer pride and refusal to accept aid, I'd have great respect for his choice and confidence in his future success. However, it all sounds very much like an urban legend.
It's a nice idea, but I suspect it wouldn't work. It's like free coooperative DNS - too tempting for abusers. I suspect that if Freedom takes off, Abuse will be their biggest cost center. The service is a natural haven for crackers and spammers. Hobbyists might enjoy setting up the FreeFreedom servers, but I doubt they'd enjoy chasing down and disconnecting abusers. Besides, in a free anonymous system there is really no such thing as disconnection.
In fact, netblocks housing such servers would very likely end up on the RBL, never to be removed (until our ISP's TOS us).
They were in a strong position to become the linux vendor to many Unix shops and fill the role that HP or Sun would fill. The magic, had they pulled it off, would have consisted of welding the disparate commodified parts of a PC into one shining 'workstation'. Unfortunately they did not do this.
I had a VA workstation at one company. I remember when it first arrived - I turned it on and started X. Someting was wrong. The resolution was (I think) 640x480. This was on a machine with a Matrox card. I've never seen a Sun in low-res mode. I don't even know if they're capable - that's the whole point. From the time you shove the install media into a Sun to the time you throw it in the dumpster due to obsolescence, you only see it in the correct video mode. I have no idea what VA was thinking. If I were a naive purchaser, I would have left the box as it was and been somewhat unhappy with the display. Plenty of Windows users run 800x600 after all. A coworker had a VA box with a CD-RW (factory, of course). It never worked very well. After numerous phone calls, VA said there might be a kernel bug causing the problems, and put the fixed kernel on their servers for download. Which was good. The bad part is that I think they shipped this system not only without testing it, but without ever testing one like it (thus the late discovery of the bug).
VA had all the chances to offer a Macintosh level of integrating and testing. They controlled the OS and hardware. I don't think they delivered.
I should also mention the multi-week lead times for standard workstations. All in all, a bad vendor and the only reason they were used is their support of popular web sites. Which was not very persuasive with management.
Unpleasant? Of course. Like it or not, our modern era is largely rooted in WWII. The Nazis define a certain pole of human behavior - in many contexts they're invoked as the extreme of authoritarianism. Which is not quite the same thing as comparing someone's trivial actions to the holocaust morally.
Godwin's law may have been a good idea at one time (although I doubt it - 'the act of censorship is always worse than that which is censored.') but it kills many interesting discussions now. Hitler really introduced 'total war' and the blitzkreig. The parallels to Microsoft are astonishing, but I'm not sure I could do the subject justice. Especially because of the inequality of their crimes - economic predation versus mass murder.
At any rate, I'm afraid you can't change the fact that 'grammer nazi' is a natural and evocative phrase. Slang tends to touch on the most powerful things - God, sex and death. The crimes of the Nazis have made their name powerful in metaphor.
- It's full of unfinished (unusable) and utterly vaporous projects. It really pisses me off when I'm searching for a solution to a problem to waste minutes on a SourceForge site, sometimes even downloading a tarball and trying to compile, merely to discover I've wasted my time on someone's empty fantasies. You can assert that as the consumer of software I have no right to expect, etc., but I think it's an act of inconsiderate pollution to spread useless software.
- Before SourceForge, project web pages were both more usable and more informative of the character and status of the project. More usable because less cluttered - Open Source project web pages are generally some of the best on the web for clarity. More informative, because a half-baked toy project will get one hastily-written page, if that, while a mature, heavily used project will have many pages and current release news.
- I shudder to think what Microsoft's PR folks could make of SourceForge. If they want to convince an executive that Open Source is ridiculous, they just have to let him browse this site and take in all the 'discussion forums' with two posts, all the ambitious projects with no release info. It's an ongoing embarassment.
Anyhow, for those who love the enhanced functionality of SourceForge, the software is free and can be implemented elsewhere. Hopefully in a post-SourceForge world it will implemented only when a project has outgrown a simple static site.We're in no danger of reverting to UNC FTP archives. (Not that there's much wrong with them.) Pretty much every ISP offers free static hosting to its customers. Then there's angelfire/geocities for the cash and bandwidth impaired.
Now that the 55 mph speed limit is gone, I don't seen anyone saying "Drive 55 - it's not the law, but it's a good idea." In other words, their concern for safety turned out to be merely a slave's concern with obeying his master's orders.
You claim that 'speeding' is unsafe. If a man drove 70 mph every day down the same highway from 1970 to today, it would have been 'unsafe' in the 70's and 'safe' today because we've raised the speed limit.
Most traffic signals use induction loops to detect cars. Essentially an AC signal is sent through a wire loop in the pavement. Putting a big hunk of metal on it changes the impedance. Nothing to do with weight.
That may be your common sense interpretation, but it's not a universal rule. As it happens, you're right about California law. But the reason the law exists is that employers were previously claiming (and compelling employees to grant) ownership of inventions made at home without the employer's resources. Now California employers make employees sign agreements designed to bypass that law by putting the burden of proof on the employee to show that absolutely no company resources were used.
That was a pretty cool response. I was halfway in agreement with the control freaks here - I can certainly understand the fear of having trojaned boxes behind the firewall. But Universities continue to show that openness is possible.
My question in these situations is always, "Why do all the machines have to be at the same trust level?" Or to put it differently, maybe it's time to rely more on host-based security and less on firewalls. Given a big enough site, there must always be hostiles behind the firewall. So why not put the desktops on their own network behind a different firewall from servers? Let them infect each other. Of course, even if you completely distrust the desktop machines (best way IMO) it would still be upsetting to have SS7 on them capturing every password.
Maybe NSA's trusted linux will solve this stuff.
How dare you pit your puny facts against the mighty river of ignorance that is slashdot? Might as well give this one up, I think. The slashmasses are firmly convinced that M$ is somehow based on BSD code. Whoever invented this meme is a genius because it is unkillable.
I know two techies who are socially inept. Both became unemployed in the crash. One of them is still unemployed. He has gotten two jobs, and in both cases left within a month due to personality conflicts. He is used to being treated as an elite coder code and doesn't like sitting in meetings with idiots. He is also quite frightening to non-techies - streaks of the BOFH. The other guy has a job largely because I hooked him up. I'm not sure how long he'll keep it, although he's doing a very good job. The thing is, he doesn't know how to schmooze his manager at all. I'm not talking about utter corporate ass-kissing - I'm talking about the ability to give your manager more than a blank stare or a dissertation on the latest Linux kernel.
Anyhow, it's true that the end of the bubble spells disaster for some talented but unsocial techies.
- View it in Netscape
- View it in Lynx
- Derive some data and stuff it in a database (what search engines do)
- Run it through a filter that removes words I find offensive
- Have it read to me by a text-to-speech program
- Hyperlink the name Bill Gates like this.
Since I believe this in general, I can't make an exception when Microsoft provides the tool.XML is not an alternative to Perl or Java. Those are programming languages. XML is a markup meta-language - a set of very simple ground rules for defining markup languages. It is already very useful. I'm writing an app that receives messages from a custom Windows app. Although the Windows programmers and I hardly share anything in common (they don't know what fork means, for example) we were able to agree on an XML message format with no difficulty.
And in case you're wondering, none of us really understand DTD's or the finer points of XML. If XML did not exist I'd probably be asking for messages formatted like RFC822 headers (Key: Value) and we'd run into endless problems with newlines, CRLF etc.
For decades programmers have been making ad-hoc markup languages and writing cheesy parsers that work 98% of the time. XML, which has exactly five reserved entities, lets us save a lot of energy and use proven standardized parsers.
There is very little to know about XML and it's nowhere near as complex as a programming language. If you've made a web page, you've written something close to well-formed XML. The only difference being that in XML every element must be matched by a closing element or contain a trailing slash. So <P> would become <P/>
Looks like you missed the part where flikx is facing expulsion and criminal charges. I you had read the entire article, you would know that Flikx does not need "the support of those sponsoring him" - he is quite capable of hosting the site without any help from the University.
Someone is facing imprisonment for airing opinions that offended the authorities, and your contribution to the ensuing discussion is a resounding and irrelevant defence of the authorities' property rights. Bravo.
Why don't you put two 12 volt fans in series across 12 volts? Then they'll each get 6 volts, which is close to your initial design. If one of them shorts out, the other will get 12 volts - no harm done.
Of course the tester wasn't trying to isolate CPU - he wanted to test overall hardware performance. And I wonder if there's a fifth variable - how well the compiler was compiled for the host platform. Given a platform X, and a compiler xcc which really optimizes for X, if you compile gcc with xcc it will perform better on the test compilation than if you compile it with gcc. (I'm assuming the test compilation is always with gcc).
Given that your SPARC compilation was slower than your Intel compilation, there are at least three possible explanations:
- SPARC architecture is a harder target to compile for, regardless of the compilation platform.
- GCC is less efficient on SPARC
- The Sun/Solaris machine is less efficient than the Intel/Linux machine.
Use of a constant target architecture would have eliminated #1.I'll respond to you as one of the many posting similar sentiments. Don't take it personally.
You are utterly wrong. The LUG had permission from the event organizers to hand out their stuff in front of the building, as LUG members have made clear in other posts. They were not giving out swag on Microsoft's rented spot, contrary to your assertion. The event organizers permitted several exhibitors to hand out stuff in front of the building, despite the fact (pointed out in other posts) that exhibitors were contractually barred from doing this. They asked (not compelled) the LUG to move simply to avoid unpleasantness from Microsoft, who chose to be unpleasant.
My real complaint with you, Jason, is not that you were ignorant of the facts, since the posts to which I refer may have been added after yours (and the many like yours.) My real complaint is the knee-jerk invocation of property rights, the automatic assumption that in any conflict between 'rebels' and 'authority' property rights must be on the side of 'authority'.
I see this faulty assumption all the time on slashdot. Yes, the Newsforge article was woefully deficient in facts, but from the skeletal facts it did provide there is no reason to think that Microsoft had any superior legal claim.
This was simply a case of a large customer (Microsoft, who rented a ballroom) demanding that a vendor (the event organizer) favor them over a smaller customer (the LUG, who rented a booth). It happens all the time in business. What makes it interesting here is the fear Microsoft displayed.
I've left out the whole "thin client" aspect, because it has been overhyped and may not always make sense. However I think that a really logical IT infrastructure would put an xterminal rather than a PC on the secretary's desk. Every PC is an ongoing cost center, and an invitation to store company documents on un-backed-up disk.
I think that ipf is much more intuitive than ipchains/iptables. It would be very sad if OpenBSD switched to Netfilter, regardless of the license issues.
I think it would almost be easier to implement free ipf from scratch. In a way, the hard part was coming up with the conceptual framework. Once you understand ipf syntax, writing the firewall should be less than astronomically difficult.
Yes, the Windows crowd are like children or savages, easily dazzled by shiny objects (shrink-wrapped software) and perpetually dependent on parents/witch doctors (boxware vendors/MS) for the talismans to ward off a scary, complex world.
But there are drawbacks to being an adult - we are burdened with the knowledge that the world is held together with duct tape and dried dung. For us there are no happy surprises in shiny packages, only the unerring certainty that software sucks and will continue to suck.
With regard to shells, I agree - it's utterly amazing that Microsoft hasn't managed to "innovate" a decent shell yet. But I have used Bash on Windows, and it wasn't pretty. Windows is sluggish in many ways that come annoyingly to the fore in shell interaction.